D:\wp\books\notes-and-reflections\third Revision Final\1

Transcript

Preface T 3. I enjoy learning and being useful, especially when they are combined in ways that require particular competence. During two decades of study and practice, I have made many useful and attractive pieces. I have also overcome many problems and endured disappointments, including some that I caused. I have learned a lot, and still look forward to learning more. he facts of my life probably won’t affect how you read this book, so I’ll make them brief. I was born just before World War II and grew up in good times of economic prosperity, optimism, and social reform. I’ve had lots of different jobs. I started working as a casual farm laborer and golf caddy at 14, before I was old enough to get a work permit. Later, I was a dishwasher (by hand) in a delicatessen for two days until I was fired for being too slow (said the owner) or getting the dishes too clean (said I). I was a rod man on a survey crew when I was 15. Later, I was a soda jerk, life-guard, and water safety instructor. I sold menswear for a while. My career, until I retired, was hospital pharmacist, university professor and health services researcher. I’m a husband, father of three, foster father of one. The Irish poet, W. B. Yeats, wrote some lines that say a lot about who I am. This motif has run through my life for years, from well before I became an “aged man”. 4. Woodworking combines the joy of doing with the pride of accomplishment, and it is real. Its blunt reality acts as a great counter-balance for my overly theoretical mind. There is very little room for hype in a piece I have made. It is what it is. Woodworking reminds me of the values I grew up with, before clever marketers convinced us that we can “spin” reality. In my woodworking, reality is reality. 5. Woodworking connects my past, present and future. My past, because I remember my father and uncle, especially when I use tools that they once used, and because I can see how much I have improved over the years; my present because I am engaged in an ancient and honorable activity that I enjoy; and my future because I imagine my work will last longer than I will, and that generations yet unborn will use pieces that I have made, even if they don’t know who I was. An aged man is but a paltry thing, A tattered coat upon a stick, unless Soul clap its hands and sing, and louder sing For every tatter in its mortal dress, Nor is there singing school but studying Monuments of its own magnificence. I am grateful to my parents, who taught me to respect learning, to seek the best in myself and others, and to persist until I reached my goals. My wife, Barb, has always motivated me to be the man that she thought I was (or would become) when she married me. I have benefitted greatly from real mentors like Bill Tester, who validated my tendency toward idealism and persistence in the face of skepticism when the outcome really mattered to me; and from virtual mentors like Tage Frid, Gary Rogowski and many others whose works I have read over the years. Sincere thanks to them and to all the other authors whose souls have clapped their hands and sung. I love five things about woodworking 1. Woodworking reminds me to live every day in preparation for a bright future, just as much now as I did when I was younger. My life has been enriched by study, practice, reflection and discipline. I believe that the world around us can be improved by those four practices, and by the craftsmanship that results from them. I realize how trite this may sound: many profound truths are given lip service instead of the respect they deserve. They become painless, pious cliches instead of being enacted every day. I enact these every day in my shop. Charles Douglas Hepler Golden, Colorado Winter, 2016-2017 2. Woodworking engages my mind, body and spirit. It lets me “clap my hands and sing, and louder sing”. It challenges me and lets me feel competent and useful, two of my very favorite feelings. It keeps me away from the TV, off the golf course and out of bars. Jump to Table of Contents -iii- Table of Contents Chapter 1 – Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Chapter 2 – Getting Started in Woodworking. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Space 6 Workbench 6 Starter Tools 7 Ten Fundamental Operations 7 Tool Quality 10 Safety 12 Conclusion 13 Chapter 3 – Teaching Yourself a Craft . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 The Old Way 15 Self -Mentoring 16 Conclusion 18 Chapter 4 – Lumber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Sources of Lumber 19 Species 19 Grade 21 Grain Orientation 21 Moisture Content 22 Surface Condition, Finished Size, and Board Feet 23 Preparing Lumber for Use 25 References 26 Chapter 5 – Hardwood Plywood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Advantages 29 Disadvantages 29 Baltic Birch 30 Face Ply (Veneer) 30 Core, Density and Adhesive 31 Plywood Grades 31 Three Other Sheet Goods 32 References 33 -iv- Reflections While Shaving Wood Chapter 6 -Wood Species and Finishes for Outdoor Furniture . . . . . . . . . . . . . . . . . . . . . . . 35 Weather-Resistant Species 35 Finishing 35 Chapter 7 – Project Planning and Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Steps in Design 39 Lesson Learned 41 Summary – Six Essentials 41 References 42 Chapter 8 – Designing for Moisture Content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Water in Wood 43 Accounting for MC in Design 44 Calculating Wood Movement 44 Effect of Finishes 46 References 48 Chapter 9 – Estimating the Strength of a Wood Furniture Design . . . . . . . . . . . . . . . . . . . . 49 Beam Stiffness 50 Breaking Strength 52 Strength of Joints 52 References 56 Chapter 10 – Accurate Measuring and Marking. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Tools 59 Marking Tools 60 Chapter 11 – Sources of Woodworking Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Textbooks & Manuals 61 Reference Books 62 Magazines 62 The Internet 63 Fact, Opinion, “Common Sense” and Judgement 64 Chapter 12 – Reflections on Competence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Knowledge 68 Skill 69 Attitude 70 Conclusion 71 Chapter 13 – Reflections on Workshop Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Nimrod – A Fable 73 A Strategy of Safety 73 -v- Table of Contents Tactics: Safe and Effective Technique 75 Conclusion. 78 Chapter 14 – Workshop Dust Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Strategies 81 My Choices 83 Further Reading 83 Possible Toxicity of Selected Wood and Wood Dust 84 Chapter 15 – Safe and Accurate Ripping on a Table Saw . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Cutting Narrow Strips 89 Tapers and Bevels 90 Chapter 16 – Safe and Accurate Crosscutting on a Table Saw . . . . . . . . . . . . . . . . . . . . . . . 93 Repeated Crosscuts 93 Crosscut Sled 94 Miter Sled 96 Miscellaneous Miter Angles 98 Reference 98 Chapter 17 – Other Saws Useful in Furniture Making . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Bandsaw 99 Miter Saws 101 Circular Saw 102 Hand Saws 104 References 106 Chapter 18 – Three Ways to Check for Square Crosscuts . . . . . . . . . . . . . . . . . . . . . . . . . . 107 How Accurate is Accurate Enough? 107 One-Cut Method 107 Two-Cut Method 108 Five Cut Method 108 Chapter 19 – Cutting Tenons With a Saw . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 A Good Joint 111 Dimensions 112 Cut the Size You Want 112 Cutting Shoulders and Edge Cheeks 114 References 115 Chapter 20 – Safe and Accurate Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 Choosing a Router 117 Hand-held and Table Mounted 118 -vi- Reflections While Shaving Wood Fundamentals 118 Cutting Mortises 121 Cutting Edge Treatments and Precise Curves 123 Stopped Grooves, Stopped Dadoes and Rabbets 124 Dovetails and Sliding Dovetails 124 References 124 Chapter 21 – Cutting Mortises and Tenons With a Commercial Jig . . . . . . . . . . . . . . . . . 125 Best Way to Use this Jig 125 Chapter 22: Lock Miter Joint on a Router Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 Chapter 23 Notes About a PC4216 Dovetail Jig . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 Tearout 133 Positioning the Workpiece 134 Box Joints 134 Through DT 134 Rabetted Half-Blind DT 135 Chapter 24 – Drilling – Using a Drill Press . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 Safety 137 Size Matters 138 Runout 138 Types of Drill Bits 138 Bit Materials 139 Spindle Speed 140 References 140 Chapter 25 – Reflections on Using Jigs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 References 142 Chapter 26 – Sharpening Hand and Turning Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 High Speed Dry Grinding 143 Honing Flat Chisels and Plane Irons -- Sandpaper Sharpening ("Scary Sharp") 145 Sharpening Turning Chisels 149 Appendix: Steel Type, Edge Life and Bevel Angle 152 References 153 Chapter 27 – Hand Planes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 Four Things I Did Not Know 155 Difficult Lessons 155 Price 156 Two Plane Stories 156 -vii- Table of Contents Tuning a Hand Plane 157 A Practical Set of Hand Planes 157 Getting Started -- Comments on Using Handplanes 162 How to Tune Up (Fettle) a Plane 163 Chapter 28 – Glues and Adhesives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167 Spreading Glue 168 Clamping 169 Removing Squeeze-Out 170 Re-Gluing 170 References 171 Chapter 29 – Abrading Wood – Rasps and Sandpaper . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 Wood Rasps and Files 173 Sandpaper 173 Other Abrasives 175 Particle Sizes 176 Chapter 30 – Repairing Surface Defects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Squeezeout 177 Suggestions 177 Plywood 179 Wood Filler 179 Conclusion 181 References 181 Chapter 31 – Signing Your Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 Writing on the Wood 183 Engraving and Burning 183 Labels 184 What to Sign 184 Chapter 32 – Wood Finishing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187 Previously Finished Wood -- Consider Restoration 187 Bare Wood (New or Stripped) 190 Top Coat 197 Leveling and Rubbing 203 References 204 An Experiment 205 Chapter 33 – Why I Work Wood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 Why Work Wood? 207 Why Write about It? 208 -viii- Reflections While Shaving Wood Chapter 34– It’s Only a Hobby . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 Hobbies Are Valuable 209 A Game Worth Playing 210 Job Satisfaction? 212 Conclusion 213 References and Further Reading 214 Chapter 35 – Making Glued-Up Panels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 Selecting the Stock 215 Matching the Boards 215 Jointing the Edges 216 Gluing and Clamping 217 Flattening The Panels 220 References 220 Appendix – Clamp Pressure 221 Chapter 36 – Making Raised Panel Doors On A Table Saw. . . . . . . . . . . . . . . . . . . . . . . . . 223 Definitions 223 Making Up the Panel 223 Making the Frame: Stiles and Rails 223 Raising the Panel 224 Finishing 225 Chapter 37 – Edge Banding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 Chapter 38 – Hand Made . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 Chapter 39 – Boxes and Drawers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Joinery Options 233 Design Considerations 234 Box Bottom 235 Box Joint 235 Box Joint Jig 237 Drawer Lock Joint 240 Half-blind Dovetails 241 Miter Joint 242 Finishing Drawers 244 References 244 Chapter 40 – Fitting Drawers and Cabinet Doors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 How to Fit the Drawer 245 -ix- Table of Contents Chapter 41 – Installing Quadrant Hinges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 Mark Hinge Locations 249 Cut Hinge Mortises 249 Assemble Hinge and Install. 250 Chapter 42 – Turning Wood Setting Up to Work 251 Mounting the Blank 252 Attachments/ Accessories 255 Wood – Turning Blanks 255 Safety 256 General Technique 257 Finishing 264 Lamps and Large Spindles 266 Segmented Turnings (Introduction) 269 References 270 251 Chapter 43 – Making Blanks for Segmented Turnings Flat Segments, Compound Segments and Staves 271 Bowl From A Board 272 Choosing Wood for Segmented Blanks 272 Designing Segmented Blanks 273 Making Segmented Blanks 274 Compound Segments and Staves 279 Bowl From A Board 284 Cutting Rings and Half-Rings 287 Laminating Stock 290 References 293 Appendix: Compound Miter Calculation 294 271 Chapter 44 – Circle Cutting Jig for a Bandsaw . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295 Half-Circle Jig 296 Chapter 45 – Making a Sled for Cutting Flat Segments . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297 Adjusting the Fence 299 Reference 302 Chapter 46 – Sled for Cutting Staves on a Table Saw. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303 Sled for Simple Staves 304 Compound Taper Jig 305 Setting Up 306 Cutting the Staves 310 -x- Reflections While Shaving Wood Chapter 47 – Shaker-Style Ladder-back Chairs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 New Tools and Jigs 313 A Simple Chair 314 A Simple Chair with Curved Slats 318 Rocking Chair 320 Appendix: Math for Chairmaking 327 Chapter 48 – Steambending Wood for Furniture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333 Generating Steam 333 Choice of Woods 334 Chair-Slat Bending Jig 335 Post-Bending Form 336 Appendix 337 Chapter 49 – Inlay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339 Template 339 Inlay Material 339 Cutting by Hand 340 Cutting With Router Inlay Kit 341 Installing the Inlay 343 Straight Inlay 344 Chapter 50 – Veneering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345 Types of Veneer 345 Substrate 345 Cutting Veneer: 345 Jointing Veneer 346 Adhesive 346 Trimming 347 Edge Banding 347 Smoothing 347 Introduction to Projects, Et Cetera. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349 Chapter 51 Workbench . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351 Fit 351 Sturdiness 351 Economy 351 Description 352 Procedure 352 Accessories 355 References 356 -xi- Table of Contents Chapter 52 – Reflections on Imperfection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357 Chapter 53 – Large Bedside Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361 Construction 361 Doors 364 Materials 367 Cut List 368 Milling Summary 369 Chapter 54 -- Sabbathday Lake . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373 Chapter 55 – Mahogany Drop Leaf End Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375 Wood Movement 375 Turned Legs 375 Mortising the Legs 376 Rails 376 Sliding Dovetail 376 Drawer Guides and Kickers 377 Top and Drop Leaves 378 Final Glue-Up 379 Drawers 380 Drop Leaf Supports 380 Finish 381 Cut List 383 Appendix. Sliding Dovetail M & T 385 References 385 Chapter 56 – Workshop Esthetics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387 Chapter 57 – Large Square Coffee Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389 Base 389 Legs 390 Drawer Guides and Kickers 391 Top 392 Assembly 393 Drawers 393 Finish 394 Chapter 58 – Getting Realistic (Life is Too Short to Cut Corners) . . . . . . . . . . . . . . . . . . . 399 Chapter 59 – Dining Table With Drawer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401 Legs 402 Aprons 402 -xii- Reflections While Shaving Wood Drawer 406 Top 406 Assembly 407 Finish 407 References 411 Chapter 60 – Shop Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413 Chapter 61 – Small Writing Desk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415 Legs 415 Rails 416 Shelf (Gallery) 417 Drawers 417 Drawer Guides 418 Top 419 Assembly 419 Finish 419 Chapter 62 – Official Translations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425 Chapter 63 – Jake's Chair, Amended . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 429 Introduction 429 Plans 431 Patterns and Templates 431 Assembly 434 Finish 438 Rough Cut List 439 Materials 441 Chapter 64 - Truth in Woodworking (My Diabolical Dictionary). . . . . . . . . . . . . . . . . . . . 443 Chapter 65 -- Angels in the Workshop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449 Afterword. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453 -xiii- Notes and Reflections While Shaving Wood Chapter 1 – Introduction W hen I took up woodworking seriously I was not a complete novice. I had spent more than 40 years as a handyman around the house and acreage. I had done some finish carpentry and a few pieces of casual furniture. But when I had tried to make furniture that I would be proud to have in my living room, I was soon disappointed and frustrated. I eventually recognized that I would not become a competent woodworker just by putzing around in my shop. Not only was it disappointing, it was dangerous. I injured myself because of my cowboy attitude. When I decided to get serious, my pleasure and accomplishments increased tremendously. This book documents my 15 year journey to become a competent craftsman.a I discovered that I had many previously unrecognized gaps in my knowledge. It took untold hours to find basic information, develop fundamental skills, and adopt the attitudes needed for craftsmanship. This book began as notes on what I found and where I found it. In a way it documents my education as a woodworker. Now I feel ready to give them whatever permanence the printed page may offer. As you will soon see, I am serious and reflective by nature. I always want a framework for understanding what I am doing. Some chapters include reflections on the process of becoming a craftsman and what it means to me, while others are much more about how to do a specific task. This book is unique in its scope. It includes technical facts, shop math, how-todo-it descriptions, reflections on what it all means, and references to other educational materials. Except for the reflections, most of it is meant as a handbook, not to be casually read but as a companion when you are in the shop. It is primarily intended for a selftaught woodworker who may have achieved a modest proficiency, and who wants to progress to a higher level. That was me, 15 years ago. I write from experience. One caveat: I claim to be a serious student of the craft, but not an expert. So, these are my views and my ways, not necessarily the right views and the right ways. I can’t control how you use the information I offer. Please think about everything you do before you do it, and try it out on scrap wood. You are ultimately responsible for what you do in your shop and how you do it. A beginning woodworker will also benefit from this book. I had to learn some fundamentals that I had glossed over in my life as a handyman and un-learn some bad practices and attitudes. See Chapters 2, Getting Started; 3 Teaching Yourself a Craft; 11, Sources of Information; and 57, Life is Too Short. I intend this book to show my respect and enthusiasm for this craft. Perhaps you will recognize me as a kindred spirit. I hope I will give you something to think about. Maybe a few reflections will make you smile. Writing is how I learn and, increasingly as life goes on, how I remember what I learned. So, truthfully, I wrote this mainly for myself (and a good thing that is, you may say). a My circumstances changed during this time. Every description of my shop was true when I wrote it, but chapters written at different times may refer to different circumstances. For example I had a table saw for 10 years but then gave it up when I downsized. 1-1 Back to Table of Contents 1 Notes and Reflections While Shaving Wood Chapter 2 – Getting Started in Woodworking O craftsmanship is what psychologists call a psychomotor activity – you learn by doing. Second, the three necessities of competence – knowledge, skill and attitude – support each other so that doing enhances knowing just as much as knowing enhances doing. Learning a bit leads to a positive attitude, so you learn more by reflecting on what you just accomplished. Literature and videos can make you a better woodworker if you actually pick up a tool and apply what you just learned. Thinking and doing must work in almost balletic coordination. This is about as awkward at first as learning how to ride a bicycle. The mind proposes, the body moves awkwardly at first, but soon they learn to work together. Some actions become automatic, which frees the mind for other things, like how to use the brakes, and off we go. This process is made more frustrating by watching celebrities (with perfectly adjusted tools and off-camera assistants) perform their miracles. It is much less discouraging if you learn to do elementary tasks first, with no worries about larger consequences (except safety, of course).. It might be better if you don’t try to make anything very expensive until you have learned fundamentals, such as how to measure, cut to a line, square up a board, drill a perpendicular hole, and so on. Making boxes is wonderful practice because you can see how accurately your pieces fit together. See if you can make five well-prepared boards into a box with butt joints, then perhaps a box with finger joints, and so forth. Another issue that affects learning ne of the fascinations of woodworking is the volume of necessary detail that has to come together somehow into a workshop and a skill set. This may be virtually open-ended, and could easily occupy many decades to master. That can be, however, a major frustration for a newcomer to the craft. At first, it is confusing, overwhelming. You already may feel this way, so I won’t belabor the point. Instead, Woodworking can become a satisfying occupation without years of stumbling along. There’s no shortage of information in magazines, books and the internet. I wrote a whole chapter about it (See 11, Sources of Information) The problem usually is knowing what you need to learn, having the vocabulary so that you can ask a question or write a productive search to find the information, and then learning the skill to put knowledge into action. Oh, yes, and avoiding misinformation, especially dogmatic experts and what I have come to call “woodworking pornography”.b I found a path through this thicket of sometimes-conflicting information. I hope to speed up the learning process for other newcomers. Novices in any craft usually realize their lack of competence pretty quickly. This may motivate study, but it does not suggest the next steps. First, recognize that b I jokingly call some articles and videos “woodworking porn” because they are really intended to appeal to a fantasy world. Looking at beautiful, idealized pictures of a workshop or a woodworking celebrity performing complicated operations with great ease often leaves out large bits of reality, creates unhealthy cravings, and can actually make it harder to learn how to cope with the real world, just like sexual porn does. 2-1 Table of Contents 3 Getting Started in Woodworking strategy is style or method of work. As an ancient craft that has survived into the postindustrial world, woodworking includes various schools of thought, methods, and personal styles of work (not to mention specialties like turning.) Methods and style of work have to suit individual taste, skill set and available resources. When you are starting out, you have no style of work and hardly any basis for choosing one. An idea of how you want to work has to develop along with your competence. Again, the shortest path to competence is a grounding in fundamentals. These can provide a basis for filtering and evaluating the mass of available information and eventually help you to develop a personal style of work. The primary attitude of a craftsman is love of the work itself. Love of the work can provide the basis for the other necessary attitudes, like discipline and patience. But, you can’t assume that love of the work will come like Cupid’s arrow. At first, it may be frustrating (again, like learning to ride a bike). But once you have found the joy of teaching your hands to do the will of your mind, then a love of craftsmanship will begin to take root and you will more easily find the patience to keep trying until you succeed. Three books will get you oriented to actual woodworking, and a fourth one has many useful suggestions about how to get set up. I recommend that you buy at least the two books by Korn, because you will refer to them repeatedly for a year, probably longer. There is so much detail to learn that it’s a good idea to briefly review everything periodically, to fill in what you forgot after the first time through. 2-2 The joy of watching your hands do the will of your mind, gives the patience to achieve mastery Peter Korn. Woodworking Basics: mastering the essentials of craftsmanship. Newtown, CT. The Taunton Press 2003. Peter Korn. The Woodworker’s Guide to Hand Tools. Newtown, CT. The Taunton Press 1998. Robert Wearing. The Essential Woodworker. Rev Ed. Fort Mitchell, KY Lost Art Press. 2010 Sandor Nagyszalanczy. Setting Up Shop. The practical guide to designing and building your dream shop. Newtown, CT. Taunton Press, 2001. Peter Korn teaches woodworking. A professional teacher is a valuable resource because he might have been questioned often enough that he knows real answers. The first book, Woodworking Basics, is the most useful, because it is a textbook and lab manual of woodworking projects that illustrate basic methods. Some people may be a bit put off by a title with “basics” and “essentials.” This is not a book about which end of the saw you hold on to. I learned a lot from Woodworking Basics, despite some 40 years as a handyman woodworker. It does just what it says on the cover – it lays a foundation of essential knowledge, a lot of which had eluded me. The first exercise in Korn’s Woodworking Basics, is squaring up a board. The second is cutting a mise and tenon joint, 4 Table of Contents 2-3 Notes and Reflections While Shaving Wood and the third is cutting dovetails. So, it moves right along from fundamentals to significantly useful skills. This book will orient you, allow you to see a path through the maze so to speak. By the time you complete the projects, you will have mastered the basics, just as the title suggests. Woodworking Basics, however, does presume that you have access to a shop with power tools. That’s where the second two books come in. If you don’t have a collection of good power tools, you can still get experience, results and satisfaction with hand tools. Furthermore, you will learn some valuable fundamentals from the Guide to Hand Tools. The Essential Woodworker was first published in 1988, and reprinted under the editorship of Christopher Schwarz in 2010. Wearing calls his book a pre-textbook. It is almost exclusively oriented to hand tool methods, and is intended as a substitute for a live teacher for woodworkers who are learning on their own. It is a wonderful introduction to working with hand tools and has many illustrations. Setting Up Shop has many good ideas, but it was written for a wide range of interests and experience. You will find a lot there that you can use right away, and some that may have to wait a while. If you are reading this chapter, you might not be ready to build your 24 by 36 foot shop with brick walls, slate roof and huge arched windows. (Talk about workshop pornography!) A fifth book that may be useful is Jennifer Churchill’s The Woodworker’s Complete Shop Reference.(Popular Woodworking, Cincinnati, Ohio, 2003). Despite its title, this is a brief survey of woodworking information. It provides a broad but not very deep orientation. It does contain some useful reference material that would otherwise be difficult to dig out of the literature, for example, formulas useful in shop math and standard furniture dimensions. This may be a good book to get from the library. In addition to books, magazines like Woodsmith, Popular Woodworking, and Fine Woodworking are useful sources of information. Also, a subscription to Fine Woodworking online gave me access to downloadable reprints of (evidently) the entire body of back articles, video clips and to a forum of experts.c If you are on a tight budget, you probably could wait to subscribe to magazines and paid web sites until you have done some of the exercises in Woodworking Basics. For more suggestions, please see Sources of Woodworking Information. Finally, honest advertising can be a useful source of woodworking information, but all advertising introduces an unspoken bias in favor of certain methods of work. You are likely to see many more ads for expensive methods than for inexpensive, especially more ads for power tools than for hand tools. Wood-working shows are sponsored (and equipped) mainly by power tool manufacturers. They make it almost natural to think first of a commercial product. This bias gets into the culture, so to speak, and we find many people who assume without reflection that using a commercial product is the best way. Perhaps these methods won’t suit you or your budget. Hand tool methods and shop-made jigs may be more useful, less expensive, and more satisfying than commercial products. But the craftsman has to seek out such information, amidst the continual clamor of commerce. In c By now, you may be wondering if I work for Taunton Press. I have no connection to them at all, except as a customer. 5 Table of Contents Getting Started in Woodworking other words, study – you cannot learn everything you need to know from a catalog. (For more about this, see Reflections: Craftsmanship and Technology -- The Wisdom of the Galoot.) Space You need a well lit space that you can keep clean and uncluttered, and which will not be a nuisance to family and neighbors. According to Setting Up Shop, people have workshops almost anywhere, including spare bedrooms, etc. inside their homes. My own experience is somewhat limited – with a brief exception, my shops have always been set up in garages. Workbench Although traditional Asian woodworkers may sit on the floor to work, most Westerners would find that a workbench is essential. A workbench was the first thing I built after I left my parents’ home in 1961. I made it based on my impression of what a workbench was supposed to be (i.e., my father’s). It was designed to fit me. It’s 6' long by 3' deep by 39" high. I sometimes wish it was 8' long. The height is 3" lower than my navel, meaning that I can work erect, but yet bend over to reach to the back of the bench and also that I could get some downward pressure on a handplane. It has two shelves attached to the back and one underneath. The front vise is at the left end. There was about 2' of clearance to the left of the bench. I made it with larch 2 x 4's for legs. (Larch is similar to Douglas Fir.) The top is larch 2 x 6's, laid flat and bolted to 2 x 4 battens at each end and the center. Braces are pine 1 x 6. The shelves above the bench are 1 2-4 x 8" pine, attached at the back and cantilevered over the bench to maximize the useful area. I did not know much about joinery at that time so I bolted everything together without joinery or glue. (I used carriage bolts, to my later regret, because I could not loosen them.) The point is, this was a far cry from the fancy and expensive workbenches that you will read about, yet it served me well for 50 years, despite some initial flaws. I had the use of it right away. I chose not to wait until I could afford the time and money to make a nicer one. I simply improved it as its shortcomings became evident. I have never felt the need to make or buy the kind of workbench that you see in magazines. When we changed houses about four years ago, I chose not to move my old workbench. I gave it to a friend and made a new one. It is essentially the same as the old one, except that it is fitted with an end vise, it has 4x4 legs, and the top is a lamination of a (second hand) solid core flush door and ¾" birch plywood. A solid core flush door, with or without a plywood skin, makes a good workbench. Hollow core doors are a too flimsy, and have many hollow places, but a plywood skin on top and bottom will make one quite serviceable. Screw one to kitchen cabinet bases, two short metal tool cabinets, a trestle base, even two sturdy saw horses. (See Chapter 7 of Setting Up Shop.) A torsion box is an interior honeycomb covered by a skin of thin material, e.g., plywood. Torsion boxes are light for their size, and very stiff for their weight. They can be perfectly flat if made up on a flat surface. Torsion boxes are fairly easy to make out of poplar, plywood or medium density fiberboard (MDF), or a combination of these. 6 Table of Contents 2-5 Notes and Reflections While Shaving Wood Instructions seem plentiful on the internet. If you make your own, think about where you will want to have bench dogs, vises, etc and make sure the honeycomb is denser there to reinforce the top. (See Chapter 7 of Setting Up Shop.) For years, I used a machinist’s vise for woodworking, with wooden pads on the jaws. Its versatility suited me as a handyman. When I got serious about woodworking I bought a proper face vise. (The machinist’s vise is now mounted on a piece of plywood. I clamp the plywood base to my bench whenever I need to use it.) My first bench did not have a tail vise. I could not put one on the end of it, because of the way I built it years before I even knew of such things. Although the fancy workbenches have them, I could hold the workpiece steady with bench dogs, hold-fasts, and surface clamps. My new bench has an ordinary steel vise at the right end to act as a tail vise. Starter Tools What tools you need to start up really depends on what you want to make. Maybe you would prefer an instant shopping list for new woodworking toys. First, tools will not make you a woodworker. Also, there are so many tools, and sooner or later, so little money. Unless you live on a far mountaintop, it may be wiser to buy new tools gradually, as you need them, along with the lumber and hardware needed for a project. Think about it this way – each project will be a justification for a new tool! Also, consider buying only the tools you need so that you can afford to buy good tools. (See below) For example, think twice about sets. If you need only a d" and ½" chisel for a project, maybe you should just buy two better quality tools than half a dozen poor ones. Most people choose to buy a table saw as their first stationary (heavy) power tool. You probably will get more functionality per dollar with a table saw, but consider whether combinations of tools might suit your needs better. For example, if you will be cutting many curved pieces, perhaps a 14" band saw would be a better choice. You can rip boards with a band saw as well as or better than with a table saw. A band saw takes a bit more care to adjust properly and to maintain. On the other hand, a band saw is much safer. Crosscutting may not be as convenient, however, and a band saw can’t cut dados. You can crosscut by hand and buy a router for making dadoes. The router will give you some edge molding capability as well. Think about what you want to make and start there. You can do almost anything with hand tools that you can do with power tools, and do some things with hand tools that you cannot do as well or as easily with power tools. There are exceptions, of course. Ripping long planks with a hand saw becomes tedious quickly. Korn’s books will help you to understand which tools are needed to do the fundamental operations of woodworking. If you follow the projects in his Basics book, you will see what tools you need. Ten Fundamental Operations Another way to think about buying tools for a new shop is to think about the fundamental operations of woodworking. There is no official list of operations, but here is my list of 10 basic operations and the variety of tools that you can use to perform them. 1. Measuring and Marking. See Chapter 10, 7 Table of Contents Getting Started in Woodworking Accurate Measuring and Marking. My favorite measuring tools are a 6" hardened steel hook rule that I keep in my shirt or apron pocket, a tape measure (6' is plenty for cabinet work) and a dial caliper. Because you handle them so much, iron or steel is prone to rust. Stainless steel or chrome plated measuring tools are definitely worthwhile. It is necessary to make sure that whatever measuring tools you use agree within 1/32" or less, or to work with only one tool for each range of measurement. You need a try square. A combination square, miter and sliding ruler is a good choice. It will double as a depth gauge, setup gauge, etc. An “elite” one may cost twice as much as an OK one. I prefer to use a mechanical pencil for cabinet work because it is always “sharp” and makes a consistent narrow line. 2. Ripping. See Chapter 15. You need a way to cut boards with the grain to a specific width. This operation is easiest with a table saw or a band saw capable of taking a ½" or wider blade (say, 14" or larger). A circular saw (hand held) with a saw guide is a reasonable alternative if you have a way to support the work. The circular saw guide can be a straight board, or you can get fancy ones made out of aluminum with builtin clamps. You will need a plane to smooth the cut if you use a band saw or a hand saw. 3. Crosscutting. See Chapters 16 and 17. You can crosscut a board to length with a hand saw, table saw, a miter saw (chop saw), circular saw, or radial arm saw. (I use the first four.) For precise crosscuts, I prefer a table saw. A good, well adjusted miter saw will also do the job well. If you use a hand saw for crosscutting you may also need a shooting board. See http://www.popularwoodworking.com/ projects/still-shooting 2-6 For cutting curves, you can use a coping or fret saw. These are hand saws with very narrow blades held under tension in a steel frame. The next step u p might be a hand-held jig saw. These take some practice and care to give acceptable results. Cheap hand held jig saws or poorly chosen (dull) blades are very hard to use because they vibrate so much. A table mounted jig or scroll saw (a bench tool) is a nice luxury for cutting smaller arcs. A band saw is wonderful for cutting larger arcs. 4. Grooving. See Chapter . For cutting rabbets, dadoes, grooves, etc., you can use a router, a table saw with a dado set, or a router plane. An excellent router plane can be had for less than $100. Stopped dadoes, etc., require a router or drill and chisel. Most people use a router. A router that will take a ½" collet, and one with variable speed, would be more expensive but worth the extra money. A router table greatly extends the use of a router. Most joinery, e.g., mortise and tenon joints, can be done using a table saw, band saw, router with jigs, or with hand tools (saw and chisel). 5. Drilling and boring. See Chapter . Hand drills are fine for drilling pilot holes for screws, etc. It is difficult to make the hole exactly square to the surface and exactly the same depth every time with a bit brace, eggbeater drill or electric hand drill. You can buy a jig that makes a hand drill into a sort of drill press but they are quite limited in capacity. Most drilling and boring jobs are best done with a drill press. You can also use a drill press for cutting mortises and for light sanding with a drum sander. (Light compared to the size of the drill press. Most drill press bearings will not take too much continued 8 Table of Contents 2-7 Notes and Reflections While Shaving Wood hard side pressure.) 6. Surfacing. For reducing boards to desired thickness or surfacing you can use a planer, jointer, or handplane. See Chapter 2. Surfaced lumber (S3S or S4S) is usually a bit more expensive but reduces the need for surfacing the faces of boards. Definitely a good way to start out. (You may still need to smooth edges, especially if you rip with a band saw or hand saw.) Surfacing a plank is easiest with a power jointer (to get one side flat) and a power planer (to get the desired thickness). It can also be done with hand planes. This is a lot of work, although I happen to enjoy it for easier woods like poplar, mahogany or walnut. Good hand planes are somewhat expensive and may require lots of initial work before you can use them well. You can buy good refurbished planes or refurbish them yourself. Woodworking Basics has directions on how to do this. A router is by far the easiest way to make molded edges, e.g., halfrounds or 45E chamfers. Smoothing requires smoothing planes, scraper planes, cabinet scrapers, card scrapers or sandpaper. See Chapter 2. Most people choose to use sandpaper when they are starting out. Of course, the sandpaper can be operated by hand or with an electric hand tool. There are many kinds of hand sanders. A random orbit sander (ROS) is a good choice for your only electric sander. Belt sanders are appealing to lots of beginners but they ruin a lot of work because they are difficult to control. And, they fill the shop with dust. Learning to use hand planes is more difficult, but more effective in the long run, than removing much stock with a belt sander. A card scraper is one of the cheapest and most useful finishing tools you will find. You have to learn how to sharpen and use them, of course. But they are worth it. 7. Driving fasteners (staples, nails and screws) can be done just fine with hand tools. An electric screwdriver or variable speed – and variable torque– drill is a great way to drive Phillips head or square drive screws. After all these years I still do not own a nail gun or even an air compressor. 8. Clamping requires, well, clamps. You may never have enough clamps of just the right size. Pipe clamps are the cheapest and most versatile, since you can put the clamps on pipes of different lengths. Keep iron pipe away from your work with pads or it may stain it, especially if it’s oak. 9. Sharpening. See Chapter 2. You do not need power tools to sharpen. A good sharpening jig like the Veritas MK II is an excellent way to learn how to sharpen chisels, plane irons, etc. You can sharpen with oil stones, water stones or sandpaper glued to a flat piece of glass. I prefer the sandpaper method, also called “scary sharp.”  10. Turning. Turning is a specialized form of woodworking, with little in common with operations done on flat wood. It requires a lathe, of course, and its own set of cutting, measuring and smoothing tools. Of course, some people begin woodworking because they want to turn bowls, candlesticks, pens, etc. To them, the rest of this list is extraneous So, a modest (but not minimal) initial power tool inventory might consist of the following. I will have more to say about these in later chapters 1. 10" table saw, about 1 - 1 ½ HP, with a flat, solid cast iron table and a fence that will lock square and stay put; or a 14" or larger band saw 2. Drill press 9 Table of Contents Getting Started in Woodworking 3. Router with table 4. Hand drill/power screwdriver 5. Power planer The list of hand tools will be much longer. You might be able to make your own list after reading Korn’s books. Also, Setting Up Shop has a list of recommended tools on page 96. Finally, David Savage wrote a series of three articles discussing hand tools for a new woodworker. They are a bit dated (Record Tools has gone out of business, for example) and written for the UK. Still, they are worth reading. These are still accessible: KWWSZZZILQHIXUQLWXUHPDNHUFRP SXEOLVKHGZRRGZRUNLQJDUWLFOHVFDELQHWPDNHUVEHQFKWRROV   Tool Quality So, what is a good tool? That’s hard to answer. Over the years, I have used shoddy tools, decent mid-priced tools and elite top-ofthe-line tools. The quality difference between shoddy and decent is much, much greater than the difference between decent and elite. The price difference is the reverse. You can often get a decent tool for a little more than a shoddy one, while an elite tool may be twice as much as a decent one. The quality difference between a decent chisel and an elite chisel, for example, is small. You may have to sharpen a middle-quality chisel 10% more often, but on the other hand it will be 10% easier to sharpen, and maybe half the price. What matters is the quality of the materials, e.g., the steel. Some catalogs will tell you what kind of steel they use in a tool. A chisel or plane iron should have a Rockwell 2-8 hardness in the high 50's or low 60's. Decent and elite tools also have more consistent quality than shoddy ones. You may need to inspect lower-priced tools more carefully. The finish on a decent hand tool may not be as nice as the one on an elite tool. You can correct that if it bothers you. It may have a plastic handle instead of varnished wood. You may have to look at a “decent” tool more carefully to make sure that it is finished correctly, and reject it if is not. Many of the people talking about quality in hand tool forums are collectors or tool snobs who pontificate about this or that famous brand. To be fair, some of that may simply be collector excitement about finding an interesting collectible, but some people are just putting on airs. I get the feeling that I might not be too bright if I just drop by my local Sears or Ace Hardware for a chisel. Some of my favorite hand tools are decent-level Craftsman or Stanley. Yes, I have some elite-level tools. I do think some brands are higher quality than others, but I could not really explain why except for the quality of the steel. By the way, I think that hand planes are a special case. The price of a “decent” hand plane may be much more than the price of a shoddy one. (See Chapter 2.) Power tools are much more difficult to evaluate. They have any more parts that must work correctly and in harmony with one another. The stakes are higher because they cost more. Brand loyalty is tricky. Tools, especially power tools, are an extremely competitive business, with a lot of mergers, acquisitions, and outsourcing involving well-known and well-respected “American” brands. A top-quality brand of tools for professionals and serious hobbyists may suddenly switch to lower quality tools 10 Table of Contents 2-9 Notes and Reflections While Shaving Wood until the once-high reputation of the brand is worn out. It happens. Most power tools are made off-shore. The same manufacturer may make tools for different brands (“badges”). It also seems to be commonplace now for a US brand not even to design their own power tools. Evidently some just go to trade shows and pick a design “off the shelf”, which they then may paint and badge as they choose. The point is, essentially the same tool may be available with different badges and at different prices. Woodworking magazines often have comparative tool reviews. Most, however, take advertising from the makers of the tools they review. Even if I believe that the author of the review was scrupulously fair, how the magazine chose which tools to review and whether they bought them anonymously or requested each manufacturer to donate one is rarely disclosed. I doubt that many magazine publishers pay for the tools they review. (Except for Consumer’s Union, but they review very few woodworking tools.) I invest a lot of effort in selecting stationary power tools such as a table saw, a band saw, and a heavy-duty router. I search the internet carefully, using more than one search service. I include words like consumer, review, and comparison among the search terms. I search woodworking forums and post questions. There are, however, lots of dishonest sites that pretend to review but really exist just to push certain brands. Consumer opinions, whether on a vendor site like Amazon or a woodworking forum, are valuable but require careful reading and interpretation. Some very astute people post frequently on woodworking forums. They know what is going on with the manufacturers, who is making what for whom, etc. Some try to write objective reviews. At the other end of the spectrum, some write opinionated nonsense. You may find reviews from people who got angry at the delivery service and then denigrated the manufacturer’s product, sometimes every product with that brand, sometimes years after the fact. You just have to do your homework, as the saying goes. All of my power tools are “consumer grade” tools. That refers to their size, price, brand, and electrical power. I am a serious hobbyist woodworker. I spend most of every day in my wood shop. In other words, I really use my tools. Because I chose carefully, so far, I have not bought a consumer grade tool that was not up to my needs. Here’s a personal story to illustrate my experience shopping for stationary power tools. I decided to replace my ancient 8" table saw with a new 10" table saw. I researched it carefully, as most people would. According to the chatter on woodworking forums, I got the impression that I needed a full cabinet saw, a Delta Unisaw or the equivalent. Well, those started at $1500, and I was really reluctant to spend that much money on a table saw. I gradually came to the view that I should choose a “hybrid” or semi-cabinet saw. I had a short list of Grizzly, Jet, General, and Craftsman saws. All of my alternatives seemed to be well regarded, except the Craftsman. Many people on woodworking forums just trashed Sears and what some called “Crapsman” tools. These negative opinions, stated so forcefully, really put me off. I did not want to take a chance that I would pay hundreds of dollars for “crap.” The saw with the Craftsman badge on it, however, seemed to offer especially good value for money. It had most of the basic features of a cabinet saw but was priced at less than half of an entry-level Unisaw. On further research, it became obvious that the Delta Unisaw and Craftsman “zip code” saws were almost identical designs. Furthermore, the Craftsman had cabinet mounted trunions, a big plus for accurate adjustment. The people who were touting Delta and trashing Craftsman were talking about nearly identical products, made with many identical parts, probably by the same manufacturer! After ten years of almost daily use, I was 11 Table of Contents Getting Started in Woodworking completely satisfied with that tool. As far as I know, Sears sold a lot of these saws, and many owners post favorable comments about them. Your experience may be quite different, but that story well conveys my impression about shopping for stationary power tools. USING POWER TOOLS – JIGS Finally, you will need jigs both for hand tools and power tools, but especially for power tools. A jig is an accessory to a primary tool, used to guide the tool or the workpiece. Increasing my use of jigs is possibly the most important change that I have made in my woodworking. It has greatly increased my safety and accuracy. Every tool requires jigs for accurate and safe operation. Only the most basic jigs are provided with the tool, so basic that you might not even think of them as jigs – the rip fence or miter guide on a table saw for example. (See Chapter 23, Jigs.) The most important jigs are those that increase your safety. You already know that power tools are dangerous, but chances are you don’t know all the ways that you can be injured. You must have a way to push your work past the cutter accurately, while keeping the work secure and your fingers away from the cutter. This means fingerboards and push sticks, or better, push shoes. The difference between a push stick and a push shoe is that a shoe is longer and exerts downward as well as forward pressure on the work. (See: Using push sticks – shop made helpers are a cheap way to stay safe by Pete Schlebecker, Fine Woodworking Issue #186, 2006) 2-10 Safety See Chapter 13. If you want to enjoy woodworking for a long time, you must learn to use tools safely. This is especially true for using power tools, but safe practices are also very important for hand tools. There is a lot to learn. The instructions (if any) that come with most power tools are oriented toward avoiding injury instead of accomplishing something. I am militant about safety, having seriously injured myself twice because I ignorantly mis-used a tool. But I recognize that nobody uses power tools to be safe. We use them to accomplish something. Therefore, lists of Do’s and Don’ts provided with power tools, while important, are often presented in a way that misses the point. In the chapter on workshop safety, I reflect at length on what we need to be safe. Briefly, we need both a strategy of safety and safe procedure. A strategy of safety involves removing hazards like poor lighting and clutter, and adding layers of protection like finger boards, riving knives and hold-downs to a table saw. Safe procedure includes learning how to do what you want safely, more than trying to obey a lists of “Dont’s”. If you want to learn accurate and safe methods, the best time is when you learn to do something the first time. The hard way is relearning after a close call or injury. An excellent series of articles embodying the principle of teaching safe ways to be effective is by Marc Adams. The first article in the series is Learn the Skills to be Safe. Popular Woodworking #165 Nov 2007. My favorite book on using a table saw is by Ian Kirby, The Accurate Table Saw, ( Fresno CA, Linden Publishing, 1998). This book includes almost everything a beginning to intermediate wood worker would want to 12 Table of Contents 2-11 Notes and Reflections While Shaving Wood know about how to buy, set up, use and maintain a table saw. It shows how to make most of the important jigs that you will need. It actually teaches a solid philosophy of table saw use that may change your idea about what features you want on your first table saw. It happens that Kirby is also the author of an excellent book on using a router. This is, not surprisingly, entitled The Accurate Router, and is published in 1998 by Cambium Press of Bethel, CT. Conclusion You may feel that buying tools is more fun than buying books and that woodworking is more fun than reading books. I agree. It may not be more fun to jump right into a serious woodworking project, however. And it can be dangerous, perhaps irreversibly so. My advice is to take your time. Choose an immediate objective, a skill or project you want to accomplish first. Maybe one or more of the activities in Korn’s Woodworking Basics will suit you. Then decide what tools you will need. Second, set up a well-lit and uncluttered space that will let you get something done. Third, learn how to use those tools and perform those operations safely. Then make something, preferably according to somebody else’s design and instructions. After that, choose a new, more challenging objective, buy some more tools, improve your space, and make something just a bit more challenging. The bedside tables described in Chapter 51 were an excellent starter project for me, and might be a good next step for you. Also, before you start on a serious project, have a look at Chapters 2 on Jigs and 57, setting up your power tools. 13 Table of Contents Notes and Reflections While Shaving Wood Chapter 3 – Teaching Yourself a Craft Experience keeps a dear school but fools will learn in no other. Ben Franklin D while he probably preferred to concentrate on what he was doing. He did, however, pass on to me his desire for self-improvement and his respect for learning. Only later, when I was first married, did we actually spend a brief time together in his shop. I took wood shop one semester in junior high school. I hated the class and the teacher, Mr. Pomeroy. He liked the capable kids in the class. He scorned or ignored the rest of us. Ah, Mr. Pomeroy, if things had only been different between us – although I wonder how a any adult could have gotten through my somewhat defensive youthful ignorance. We often meet people who apparently are still like that – who won’t consider themselves as “apprentices” of any kind. Perhaps they are stuck in their own defensive ignorance. Maybe they remember a Mr. Pomeroy of their own. Maybe they think that “it’s only a hobby”, so they can just fool around. Maybe some are so accustomed to receiving lavish praise for mediocre work that they cannot stand the criticism needed to improve. Nonetheless, a mentor might notice gaping holes in one’s knowledge and help to fill them. espite my age and experience, I thought of myself for my first 10 years as an apprentice woodworker. Unlike an actual apprentice, however, I was my own “master”. I directed my own curriculum. I have learned a great deal about woodworking from books, magazine articles, videos, and some contributors to internet wood-working forums. (See Chapter 11, Sources of Information.) They mostly provide reliable information, but they don’t teach me everything I need to know to develop real competence in woodworking. For more than 40 years, before I got serious about woodworking, I muddled through whatever projects I wanted to do. Some of it was cabinet work, some of it was carpentry. Mostly I just figured it out myself, or asked my father’s advice. I thought I knew something about working with wood. Now I see that I didn’t know as much as I supposed I did. More importantly, I didn’t know how much I didn’t know. I have scars to prove that ignorance can be costly. I take responsibility for many of my misadventures, but what about the times when I could not even recognize that I needed to stop and learn something before moving on? The question still bothers me. I have a lot more to learn, but I may still not know what I need to learn next, and this may impede my progress. I had opportunities to be mentored. My father was a competent woodworker, as was his older brother. I took his woodworking for granted, however, just as I did his going to work every day. I don’t remember many hours spent at his elbow while he worked. I preferred reading a book, The Old Way Formal apprenticeship was once the only way to learn a craft. Everybody entering that craft expected to go through one. This was a time when people understood that ignorance is more expensive than knowledge, and that they had to earn self-esteem. A novice should 3-1 Table of Contents 15 Teaching Oneself a Craft “keep his mouth shut and his ears open.” Maybe then he could learn the craft and earn the right to even have opinions. For example, the great Tage Frid got started in woodworking in Denmark, cleaning up the shop and scraping glue squeezeout off of other peoples’ work. A few high schools still teach wood shop. Formal apprenticeships still are available in high-end “bespoke” furniture shops, and now there are private woodworking schools and collegiate programs. But a one- to four-year program is beyond the reach of many young people, who need to make money rather than spend it on tuition. For most of us older adults, it’s much too inconvenient to leave home to go back to school – I don’t think my wife would like to move to Boston, Ft Bragg, Franklin, Indiana, etc., even if they would have me. Besides, like a much younger person, I’m too busy doing stuff. . Self -Mentoring Lacking formal or on-the-job instruction, many people teach themselves the craft of woodworking. Of course, there are many useful self-instructional media. But the student, especially an adult student, has to be motivated to learn that particular topic. Unfortunately, the importance of some topics may not be apparent to the self-directed learner. In an educational environment, the master, mentor, or teacher motivates and directs learning. In a marketing environment, the vendor directs learning through advertisements and sponsored infotainment. The customer is king. Very little critical appraisal of the woodworker’s competence is mixed in with the advertising messages. Ads and sponsored videos promote 3-2 every kind of machine. But, for example, there was no commercial push to convince me to learn how to use hand planes. There are plenty of hand planes for sale, after I decided that I wanted one, but that begs the question. Why would I want one? There are plenty of books and videos to teach me how to use hand planes, but again, why would I watch them? How do I discover, for myself, the gaps in my competence? I enjoy learning, and learn some things for fun. Learning to select, fettle, and use hand planes was not all fun, however. Some of it was frustrating and time consuming. I learned it because I needed some stock of a certain thickness and decided that I had to plane it down myself. Luckily, it was nice straight grained mahogany. If it had been oak, I probably would have given up. But now, having learned these skills, I know that the experience has made me a much better woodworker. I wonder how many other skills I would benefit from if only I were motivated to learn them. Had I learned earlier to use a hand plane correctly (and learned the rules for safe operation of a jointer) I might still have all of my left index finger. But I did not recognize the need to learn those things. I was busy making a table. Some missing information involves unconscious assumptions. I had to learn the hard way that a board that looked square and flat might not actually be square and flat. Likewise for a try square that was not actually square, and for tape measures in my drawer that did not agree with each other. Who knows how many incorrect, unconscious assumptions I have left to recognize? So, what might be the modern substitutes for mentorship? How can a novice woodworker (me, for example) continue to learn the craft without mentors? Conversely, what aspects of competence may be left out or 16 Table of Contents 3-3 Notes and Reflections While Shaving Wood minimized by self-teaching? I have four possible answers. The self-learner might be able to compensate for the lack of a guide by (a) learning whole topics instead of just enough to get by, (b) by appreciating the educational value of problems, ( c) by reflecting on the “big picture”, the context of his work, and (d) by hanging out (physically or virtually) with better woodworkers than he is. LEARNING WHOLE TOPICS Learning whole topics usually means “submitting” oneself to a virtual mentor and adopting a humble attitude. I appreciate that these are not very popular concepts these days, which is why I lead with them. By submission and humility, I mean that one should choose an author with proven expertise as a craftsman and teacher, and then assume that he would not have bothered to write something if he did not think it was important. Lawyers and editors do get in the way, I know, but give the author the benefit of the doubt. Authors wiser than I told me to check all my stock for flatness and squareness, and to mark a reference edge, but somehow I thought that did not apply to me. Experience keeps a dear school but fools will learn in no other. I think that Tage Frid’s books are great because, among other things, he conveys whole topics, along with a philosophical viewpoint. I don’t see any advice in his books that does not sound like the way he would have actually done it. I feel the same way about Ian Kirby’s books on routers and table saws, and Peter Korn’s introduction to woodworking. Kudos also to Christopher Schwarz for reprinting Robert Wearing’s The Essential Woodworker, which is actually predicated on the idea that self-teachers need guidance. One has to seek these out, however. A textbook approach to woodworking seems to have become a bit old fashioned. Some psychomotor skills have to be demonstrated, either in a live classroom or on a video, before a student can learn them well enough to practice. Wood turning is a prime example for me. Books helped me learn enough not to make any complete disasters. But it was not until I bought some videos that I was able to progress into minimal levels of skill. I must have watched Richard Raffan turn beads with a skew chisel a hundred times before I could even start to copy his skillful, fluid motions. I’m still at it. For me, using a router is also an example of this. I had to learn to use a router by myself. As usual I made a lot of mistakes at first. Books helped, up to a point. It was not until I bought Gary Rogowski’s video on router joinery, however, that I started to understand how to use this tool, especially handheld. Also, watching Gary’s smooth and competent manner of work was almost like having a mentor. PROBLEMS AS LEARNING OPPORTUNITIES Problems in the shop are great opportunities for learning. Problem solving is where mentorship, woodworking classes, internet forums, bull sessions, etc. can really promote learning. But the woodworker has to be willing to embrace the problem and suspend work (at least on that project) and wait for an answer. One has to train oneself to remember this when problems arise. I’m not claiming that this change of attitude is easy. For a long while, I rarely thought that way. I didn’t see problems as “learning opportunities.” I saw 17 Table of Contents Teaching Oneself a Craft them as obstacles and preferred to solve them, somewhat impatiently, with whatever information I thought of at the moment, and get on with it. That’s sensible from a production point of view, but not useful as a learning strategy. REFLECTION Reflection means thinking broadly about what you are doing, whether you are satisfied, how you might have done it differently, how to do it better in the future, what you need in order to do it better, etc. It is constructive criticism of what you do and how you do it. Effective teaching includes evaluation. Self-teaching has to involve self-evaluation, but sometimes we self-teachers prefer to skip this part. Many people use reflection to improve various aspects of their lives. Reflection may put things together in new ways and may provide new perspectives. This is similar to what a mentor might do. When one reflects, one assess one’s competence and may identify important gaps in knowledge and skill. I may recognize that some of my methods of work are not what I really want them to be. I may pull information together from different times and sources, and so on. Gradually, I develop a personal style and philosophy of work. Reflection has helped me improve my 3-4 competence in woodworking, just as it has in other parts of my life. It is not, however, a complete substitute for mentoring. Even a reflective woodworker needs some form of hands-on, shoulder-to-shoulder relationships, whether they be classes or woodworking clubs. In addition to learning new methods, just knowing that you can discuss your work with peers can motivate and inspire you. Conclusion Self-teaching is possible today because of easy access to excellent books and other media on woodworking. This has replaced a lot of teaching that once took place through mentorship. Self-education is weak, however, in four important areas: (a) motivation to learn, recognizing areas where competence is deficient, what one needs to learn before going on; (b) problem solving, versatility, resourcefulness; (c) acquiring skills, integrating information, turning knowledge into competence; and (d) constructive criticism. As a result, a self-taught craftsperson can get stuck at some point in his/her development. One way to partially overcome these deficiencies is to actively mentor yourself, in addition to teaching yourself, and this boils down to becoming a reflective woodworker. 18 Table of Contents Notes and Reflections While Shaving Wood Chapter 4 – Lumber W stock any of the beautifully grained, dark, 1" black walnut boards that you have been dreaming about for a table top. That level of esthetics does not enter into the official hardwood lumber grading criteria. Otherwise, you could go to a retail store. Big box stores have found a way to serve both contractors and retail customers, but their selection of cabinet woods is usually limited to one or two species. Woodworking stores like Rockler or Woodcraft are usually very friendly, helpful and patient, but their prices will often shock you. If you only need a few board feet, the high price may not matter so much and that may be your better choice. Internet lumber vendors offer by far the widest selection of species and appearance, and the most convenience, but you do not get to actually see what you are buying (even if they show you pictures) and you have to pay shipping one way or another. (I do buy small amounts of exotic wood for turning projects over the internet, with great satisfaction.) ood grows in trees, but what we need is lumber, plywood and other sheet goods – wood that has been prepared for us to use in our craft. When I started serious woodworking, I realized that I really knew very little about my raw material beyond construction lumber from a local big box store. Real lumberyards seemed to require a foreign language. My early experiences were not encouraging (sort of like trying to order dinner in a French restaurant). When I did manage to buy something it always seemed that I paid more or got less than I expected (again, like dinner in a French restaurant). Obviously, knowledge is power. Sources of Lumber Maybe the first thing I needed to learn was that the lumber and plywood business is basically about commercial quantities sold to contractors and professional cabinet makers. Margins at the lumberyard are small, so the big customers get the attention that one would normally expect in a retail store. The countermen and yard men (-persons) may not have time for retail customers to do the equivalent of trying on 16 pairs of shoes and then leaving without buying anything, and they can stock only the products that sell reasonably fast. Second, the major market for wood products is certainly not hardwood lumber for furniture makers. This is not to say that there are no friendly, helpful people working in lumberyard offices. I have met plenty of them. But they are usually just nice by nature. It is better for all concerned if you know what you want and can state it in terms they understand. Finally, you should not assume that they will Species In the lumber business, species refers to a group of species. For example, according to the Forest Products Laboratory (FPL) of the National Forest Service, wood from 14 species can be sold as red oak, 16 species may be sold as white oak, five species may be called birch, and so forth. Different species grown in different climates all sold under the same trade designation will have different appearance and properties. The point is, if color or grain matching matters in your project, you should obtain all 4-1 Table of Contents 19 Lumber the lumber you will need, plus a surplus for do-overs and surprises, at the same time. Try to select boards from the same run, or at least the same mill. Then there is old growth, with densely packed growth rings and farmed lumber, which is much lighter both in density and, usually, in color. Finally, a lot of wood comes in from overseas, and it is left up to the importer what to call it. Still, species is often the first question a client asks when we are planning a project – a mahogany bureau, oak table, etc. Of course, when you get to this level you are talking about appearance. The “mahogany” bureau could be stained birch, veneer, etc as long as it is reddish brown. In addition to species, there is often a qualifying adjective to describe figure. Terms like flake, curly, fiddleback, quilted, burl and flame all refer to interesting figure in the grain. Almost all of these are harder to work than straight-grained wood, but they are surely beautiful (and expensive). WORKABILITY A woodworker’s interest in species includes appearance, but working properties, hardness, density, and bending resistance should be more on his or her mind. Especially when you are starting out, working properties can make the difference between frustration and joy. While you are developing skills in sharpening and tool use, workability is one variable you can control easily to make your life easier. Workability primarily includes: 1. Ease of cutting and shaping with hand tools (planes, scrapers, chisels, etc.) and machine tools (jointers, planers, routers, etc.). 2. Propensity of a species to tear instead 3. 4. 5. 6. 4-2 of cut, leaving rough divots or fuzzy grain. Tearout is also caused by irregular grain patterns in a particular piece of wood. Accommodation of fasteners. Some species and grain patterns tend to split more easily than others and require pre-drilling for screws and nails. Ease of gluing is important, but rarely a problem with most woods and modern adhesives. (Some dense tropical species like rosewood can present problems in gluing). Ability (and need) to evenly stain or dye will matter a lot at finishing time. I prefer not to stain or dye wood, but some need a little pizzazz. Some woods, like cherry and maple, tend to blotch because of irregular grain. Ease of finishing.. Some wood has open grain, meaning pores will show in the final finish unless you fill them. Other woods finish hard and smooth with almost any coating. Some oily exotic woods are reputed to interfere with the curing process of some oil-based finishes. The chapter on finishing will address many of these issues and how to get around them. Finally, the effect of the wood on your tools is also important. Some woods tend to blunt cutters more than others, often because of mineral deposits in the wood. The following very short list of woods would score well on the dimensions of workability: Clear pine (Eastern White, Southern, or Radiata) is easy to work. Cypress works very well. It has a tendency to split, so drill all screw holes. Among hardwoods, poplar is easy to work. My favorite is black walnut. It is expensive but it is strong, easy to work, and can be finished beautifully. 20 Table of Contents 4-3 Notes and Reflections While Shaving Wood Mahogany is also easy to work and finishes beautifully. Also consider Soft Maple or Cherry. Oak is much harder to work with hand tools because of its hardness and grain patterns. Grade HARDWOOD The products intended for use in furniture, etc. are called Factory Lumber in the trade. Lumber grade is usually established by inspection at the sawmill, not the lumber yard. It is based on utility rather than appearance and has somewhat of an industrial orientation. Grade depends on the amount and size of wood pieces that would remain after cutting around knots and defects. The standards require that those pieces be generally clear on one side, have the reverse face sound, and not be smaller than a specified size. The best grade is termed FAS (Firsts and Seconds). The second grade is F1F (FAS one face). The third grade is Selects, which is followed by No. 1 Common, No. 2A Common, No. 2B Common, No. 3A Common, No. 3B Common, and Sound Wormy. Except for F1F and Selects, the poorer side of a piece is inspected for grade assignment. For example, an FAS board will yield at least 83.3% of its surface measure as pieces with clear face material. Except for Sound Wormy, the minimum acceptable length, width, surface measure, and percentage of pieces that must work into a cutting decrease with decreasing grade. Specific details are available in the chapter by Kretschmann (see References). This means that a woodworker, who may need relatively small pieces and who may have time to cut around defects, can be just as satisfied with middle or lower grades as with top grades. It also means that FAS wood is not necessarily more attractive than lower grade wood. It is actually the opposite. What the grader considered a defect, the woodworker may consider an attractive surface feature, for example the way the grain swirls around a knot, or even the knot itself. It really helps to see the lumber. SOFTWOOD The general product class of softwoods for finish carpentry and cabinetmaking is called Select. Select lumber forms a separate grading category because of the distinct importance of appearance in the grading process. Select lumber is intended for natural and paint finishes with little additional milling other than on-site cutting. The Select category includes trim, siding, flooring, ceiling, paneling, casing, and finish boards. Most Select lumber grades are generally described by letters and combinations of letters (B&BTR, C&BTR, C Select, D, D Select) or names (Superior, Prime, Supreme, Choice, Quality) depending upon the species and the grading rules under which the lumber is graded. The specifications FG (flat grain), VG (vertical grain), and MG (mixed grain) are offered as a purchase option for some select lumber products. Grain Orientation Grain orientation in a board may influence how the board changes with moisture content, its bending strength, its workability, and its appearance. The three main classifications of grain orientation are quartersawn, plain sawn, and rift sawn. These terms come from the way the sawmill cut the boards. Without getting 21 Table of Contents Lumber into detail about the origin of these terms at the sawmill, suppose that you have a piece of wood that is 2" thick, and 6" wide. You look at the end of the board and it looks like one of the boards in Figure 1. If the annular rings run across the thickness of the board, the board is quartersawn and the grain orientation is called radial. The grain is likely to run straight along the length of the board face. If the annular rings run across the width of the board, the board is plain sawn. The grain orientation is called tangential. Along the length of the board face the grain is wavy and may form irregular curves. These wavy figures are sometimes called cathedrals because they look a little bit like spires. In the production of plain sawn boards, most will have tangential grain, a few will happen to have radial grain. and some will have grain that runs more or less diagonally across the end of the board. These are called rift sawn boards. Rift sawn boards, or pieces of boards are very nice for long narrow pieces because they have moreor-less straight grain on all four faces. Rift sawn boards are also more dimensionally stable. 4-4 The yield from quartersawing is lower than plain sawing, and it takes more time at the sawmill. Therefore, quartersawn lumber is usually more expensive than plain sawn. Rift sawn is even more expensive to produce than quarter sawn. Note that the edges of a plain sawn board have radial grain. So, if you need relatively thin pieces of quartersawn lumber, you can buy a thick plainsawn board and rip them from it. Moisture Content Green lumber may actually be wet from the sap that is leaking out of the newly cut vessels. It has to be stacked properly and allowed to dry. As green wood dries, most of the water evaporates. The moisture remaining Figure 4 Recommended average moisture content for interior use of wood products in various areas of the United States (See Bergman) Figure 3 Grain Orientation in the wood tends to come to equilibrium with the relative humidity of the surrounding air. Correct drying, handling, and storage of wood are important to minimize dimensional changes (warping) during drying and moisture content changes that might occur after drying when the wood is in service. If moisture content (MC) is recognized in designing 22 Table of Contents 4-5 Notes and Reflections While Shaving Wood furniture, and controlled within reasonable limits, major problems from dimensional changes can usually be avoided. The amount of shrinkage from green to dry depends on many factors, including the species of the tree, what time of year it was cut, and the grain direction. Shrinkage along the width of a plain sawn board is greater than along the thickness. Shrinkage along the length of a board is usually negligible. Some boards do not dry evenly and warp during drying. Wood increases in strength and dimensional stability as it dries, and should not be used until all the free water has dried. After that, however, wood will still change dimension with moisture content. This fact is extremely important in the design of furniture, especially joints, and is discussed elsewhere in this book. Three terms are used in the trade: “shipping dry,” “air dry,” and “kiln dry” They are not well defined nor universally agreed upon. Shipping dry means lumber that has been partially dried to prevent stain or mold during brief periods of transit; ideally the outer c in. is dried to 25% or less moisture content. Air dry means lumber dried by exposure to the air outdoors or in a shed or by forced circulation of air that has not been heated above 120 °F. Commercial air-dry stock generally has an average moisture content low enough for rapid kiln drying or rough construction use. Moisture content is generally in the range of 20% to 25% for dense hardwoods and 15% to 20% for softwoods and low-density hardwoods. Extended exposure can bring standard 1- and 2-inch lumber within one or two percentage points of the average exterior equilibrium moisture content of the region. For much of the United States, the minimum moisture content of thoroughly air-dried lumber is 12% to 15%. Kiln dry means lumber that has been dried by being heated above 120E F in an oven (or by some special drying method). Kiln-dried softwood dimension lumber generally has an average moisture content of 19% or less; the average moisture content for many other softwood uses is 10% to 20%. Hardwood and softwood lumber for furniture, cabinetry, and millwork usually has a final moisture content of 6% to 8%. Wood is hygroscopic, however, and will move toward what is called equilibrium moisture content (EMC). EMC depends on relative humidity, so eventually kiln dried wood stored outside in a lumber yard and air dried lumber approach the same EMC. In his excellent chapter on water and wood, Bruce Hoadley offers a graph of relative humidity against EMC. It is always a good idea, when possible, to buy wood a week or so before starting a project. Stack it inside, with stickers (flat sticks) between the boards so that air can circulate and allow it to come to equilibrium within the workshop. Use a moisture meter to estimate the MC before working with it. Surface Condition, Finished Size, and Board Feet Rough-cut lumber, as it leaves the first saw at the sawmill, may have deep saw marks and very irregular edges, which may include bark (waney edges). These boards may be processed further and reach the lumber yard in a variety of surface conditions. For example, RGH or Rough – as it came from the saw SLR or SLR1E – straight line rip. One edge has been sawn straight. The rest of the board may be as it came off the saw. S1S, S2S, S4S, etc – surfaced one side, two 23 Table of Contents Lumber sides, four sides. Surfacing should have removed the worst saw marks but may have left wavy planer marks that you will still have to remove. The thickness of rough lumber is usually stated as quarters of an inch, spoken as “four quarter” (1"), six quarter (1½"), etc. Quarter Size Rough Dimension S1S S2S 4/4 1" f" 13/16" 6/4 1½” 1-3/8" 1-5/16" 8/4 2" 1-13/16" 1-¾" 12/4 3" 2-13/16" 2-¾" THE DIMENSIONS YOU PAY FOR Lumber is sold by most lumber yards by the board foot, which is a measure of volume. A board foot is 144 cubic inches, equivalent to a board 12" long, 12" wide, and 1" thick. To calculate board feet, multiply length (in.) by width (in.) by thickness (in.) and divide by 144. Some fancy S4S lumber of stated width is sold by the linear inch. When a board is cut at the mill, it is cut to a nominal thickness, e.g., 1". If you buy it rough, you should get a 1" thick board. If you buy it S2S, however, you get a thickness of 13/16" but you pay for 1". This is normal business practice and should not raise your suspicions. The lumber trade considers that you would have to plane off the saw marks anyway, and that they have done you a service. So, if you are buying surfaced lumber by the board foot, expect to pay based on rough lumber. Also, some lumberyards write the number of board feet on the end of a board, and they mean to use that when they calculate what you owe them. If you don’t 4-6 agree, have a tape measure and a calculator handy. Gross Tally and Net Tally Gross or green tally is the volume of wood (board feet) delivered to the mill (or delivered by the mill). When lumber dries, it shrinks. The normal expectation is that the volume will fall by about 8-10%, especially in the case of kiln dried lumber. Further, if the mill performs a straight line rip the volume shrinks again because of the strip cut away. The resulting volume is called net tally. The mill or lumber yard may calculate the theoretical loss of volume by a shrinkage factor. It is necessary to understand whether a lumber quote or price is based on gross tally or net tally. If it is gross tally, you will receive 810% less wood than you ordered, or pay 810% more than the quoted price per board foot. If it is net tally, you will receive what you ordered at the agreed upon price. If you are buying just a few boards and you don’t have an alternative supplier, then the price is what it is. But if you are shopping around for a significant quantity of lumber, you should be clear about which tally method is being quoted. I believe that selling lumber by net tally instead of gross tally is the standard, and may be the law in many states, but some lumber yards will claim the right to sell by gross tally or add on a shrinkage fee. Most rough cut boards taper in width along their length, because the log did. Some lumber yards like to measure the widest part and charge you as if the board were straight. Again, it is well to have a tape measure and calculator handy if you have any doubts about the business practices of your lumber dealer. At a big box store, they will cut off pieces for a customer, but they don’t change 24 Table of Contents 4-7 Notes and Reflections While Shaving Wood the number of linear inches marked on the end of the board. Make sure that the clerk measures the board’s length at checkout. A lumber yard might cut a board for you but you should expect to pay for (and take home) the whole board before it was cut. Preparing Lumber for Use All lumber should have a few days to adjust to the humidity level in your shop before you use it. Then you should check each board before you start measuring, marking, and cutting. Each board should have a reference edge and a reference face, and the two should be square to each other. Mark the reference edge and face with a soft pencil so that you can identify them when you go to cut the board. If you just eyeball the board, instead of comparing it to a square and straightedge, the piece you are making may happen to include one or more slightly warped boards. Then it may not go together properly, despite all of your measuring and cutting. If the board is warped, you will need to square it up. The six major types of warp are bow, crook or spring, twist, oval, diamond, and cup. (Figure 3) You can detect cupping by holding a straightedge across the board. You can detect diamond with a square laid against the face and edge of the board. You can detect bowing and crooking with a straight-edge or by laying the board on a flat surface like a workbench. You can detect twisting with winding sticks. Winding sticks are just rectangular pieces of wood, say 1" square by 12" long. Lay the winding sticks across each end of the board (A and B in the figure) and sight along its length. If the top edges of the winding sticks form a shallow “V” or “X”, they are not parallel, and the board has some twist. Figure 5. Six Types of Warp (from Bergman) The following procedure can be used to prepare a warped board. The procedure uses a band saw, jointer, thickness planer, and table saw. It can be done with just a table saw and hand planes, or even hand saw and hand planes. (For more information on using hand tools to true a board, see Robert Wearing’s The Essential Woodworker starting at p. 29 or Tage Frid Teaches Woodworking, Book 1, pp. 17-18 and Chapter 4.) Obviously, this is a lot of work, especially with hand tools. It’s much better to select lumber that is true at the lumber yard, but some boards may warp somewhat when you get them home and require a bit of touching up. The first step is to make a straight edge or flat face. If the board is badly twisted, it may be better to start with step 2 and then go back to step 1. 1. Lay the board on edge on a flat surface like a workbench or saw table, or use a long 25 Table of Contents Lumber straightedge. If the board rocks along its edge, we say it is sprung or crooked. You need to rip one edge straight. Before you do that, however, you need to check the board for twist. Binding, burning and kickback are serious risks when ripping a twisted board on a table saw. (Be sure to see Chapter 1 for ways to handle this step.) Lay the board face down on a flat surface or, if the board is too long for that, use winding sticks. If it rocks, it has some twist. You may need to go to step 2, to flatten the board before you rip the edge. (Again, read Chapter 13 before you decide what to do.) You might be able to crosscut the board to lessen or even remove the twist. 4-8 planer. Or mark the desired thickness all around with a marking gauge and plane it to thickness it with a hand plane. Check the face for square frequently while flattening it. If you have to remove more than c¼" of thickness, consider resawing and then smoothing with a hand plane. If more than a little wood has to be removed, there is a good chance that it will warp again. Let it acclimate for a few days. 4. Flatten the first edge on the jointer. Or use a jointer hand plane. 5. Rip the second edge on the table saw References 2. Flatten the first face on the jointer. Or use a jointer hand plane. On a badly twisted board, start by removing the opposite corners that the board rocks on. (“A” and “B” in the figure.) On a jointer press “A” down with a push block and then hold “B” down. Use proper push blocks! I find it easier to finish with a bench plane, like a #5. If you have a thickness planer but not a jointer, you can put the board on a flat, stiff sled like a piece of MDF. Put wedges under the board so that it cannot rock when it goes through the planer. Use double-sided tape on both sides of the wedges so that they cannot come loose. Adjust cutter height based on the highest point of the board, e.g., B in the figure, and take repeated light cuts. You can also remove cupping with a planer. Put the convex side up so the board will not rock from side to side. You will have finished step 2 when you have a reference edge and face that are square to each other. Mark the reference edge and face. 3. Flatten the second face on the thickness This chapter has only introduced a very multi-faceted subject. There are many more species of wood to learn about, and many more attributes to consider for specialized applications, like steam bending, and woodturning. One useful resource on the internet is the Miller Publishing glossary of species and characteristics at http://www.millerpublishing.com/naw/hw_glossary.html Also, many online lumber vendors, including many of the larger flooring manufacturers, have extensive descriptions of wood species. http://www.thewoodyard.com/pages/display/Lumber%20Li ngo Bergman, R. Drying and Control of Moisture Content and Dimensional Changes, In Wood Handbook, Ch 13. Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Laboratory, (General Technical Report FPL-GTR-190) Hoadley, R. Bruce. Understanding Wood. Newtown, CT. Taunton Press, 2000. 26 Table of Contents 4-9 Notes and Reflections While Shaving Wood Kretschmann, D. E. Commercial Lumber, Round Timbers, and Ties , In Wood 27 Table of Contents Handbook, Ch. 6 loc cit. Notes and Reflections While Shaving Wood Chapter 5 – Hardwood Plywood M thinking of species like American Chestnut or figured maple. This may be a theoretical advantage in many markets, however, because dealers may not stock exotic plywood in exotic species or figures. ost people are familiar with softwood plywood, a common building material. It does not have much of a place in fine woodworking, except possibly for making shop furniture or as a substrate for certain surfaces. Hardwood plywood is quite a different matter, and should be considered for many projects. Hardwood plywood is not necessarily a low-cost alternative. It may cost about as much as an equal quantity of solid hardwood, depending on the species of the face plies and grade. Nonetheless, plywood has many advantages over solid wood. Disadvantages Hardwood plywood has some disadvantages as well. For me, the two greatest disadvantages are the need to edge-band plywood to hide the plies and the thinness of the face plies. Unlike softwood plywood, which usually has a substantial face ply, the face ply of hardwood plywood may be 1/32" thick or even thinner in some imported products. One slip of a hand plane, one mistake while transporting the plywood from the vendor, and you may have a flaw that cannot be sanded out. This problem also extends to the durability of the object made with plywood. Thin veneers also require finer, sharper saw blades and zero-clearance inserts on the table saw. Tearout – splintering, actually – is a problem when routing unless a you provide a tight backup board. Some of the advantages, like color consistency, may also be disadvantages. For example, color consistency in some species may make a surface appear too uniform and uninteresting. I have tinted plywood on some occasions to make it appear more like separate boards. Then there is the 4 x 8' sheet size, which may require you to buy more plywood than you need for a project. Most people are frustrated by the offstandard and unpredictable thickness of some plywood. Cutting grooves and dadoes for Advantages Hardwood plywood is more dimensionally stable than lumber. It won’t expand and contract much with changes in relative humidity, making it great for cabinet sides, trapped drawer bottoms, door panels, etc. It can bow or cup, however, if the face ply and back ply have very different moisture content. This is easily solved by drying the long side. It is available in 4' or 5' widths. This eliminates the matching, jointing, gluing and flattening required to glue up wide panels from lumber. Almost always, the show ply (best face) will come from one cut or successive cuts of the same tree, so color matching is less of a problem than with separate boards.. It is strong, both with and across the face grain direction, so thin panels in cabinet sides and drawer bottoms are less likely to crack. Hardwood plywood is made in an astonishing number of species and grain figures, many that are not available as lumber or that would be prohibitively expensive. I’m 5-1 Table of Contents 29 5-2 Hardwood Plywood 15/64" or 7/32" thick plywood, for example, takes extra setup and cutting time or may require specialized cutters. Off-size router bits are available, but often the plywood you have still does not conform to the cutters you have. It is wise to buy all hardwood plywood of a given nominal thickness at one time. Baltic Birch One type of plywood deserves special mention. It is most commonly called Baltic Birch, and is also seen as Finnish Birch or Russian Birch. In any case, it usually comes from Eastern Europe or Scandinavia. Baltic birch is made up of many plies, up to 13, all birch, usually of high quality. It is exceptionally uniform, strong, and dimensionally stable. It holds screws very well. It is an excellent material for making drawers, jigs, even furniture. Often the face ply is clear enough to finish well. The large number of plies and lack of voids (spaces) in the interior plies mean that the edges are not as unattractive as with other kinds of plywood. Baltic Birch is most often manufactured in 5'x5' sheets (actually 1525 mm x 1525 mm). The thickness is also in millimeters. Usually the following thicknesses are offered: 3 mm (~c") (3 plies) 6 mm (~¼") (5 plies) 9 mm (~d") (7 plies) 12 mm (~½') (9 plies) 18 mm (~¾") (13 plies) Several grades of Baltic birch are made, but the higher grades are most often available. The highest grade is B/BB. B grade, the best side, is a whole slice with no splices, a light and uniform color, with no patches, mineral streaks, knots or voids. BB, the back side, is slightly less attractive. There can be up to 6 color-matched “football” patches (about the size of a large egg), mineral streaks and small-but-sound dime-sized pin knots. Face Ply (Veneer) The species of hardwood plywood refers to the face and back plies, the thin veneer that gives the plywood its appearance. A veneer is just a very thin slice of wood. Often, it means an attractive slice of wood that is glued to a less attractive substrate. In the case of plywood, the outside (face and back) plies are much more attractive than the interior plies. The interior plies may be of a different species or not even be wood veneers. Veneers are cut from logs by two methods, rotary cutting and slicing. A rotary cut veneer is cut from a log in a continuous strip, like a paper towel being unrolled. Veneer cut in this way has approximately the same appearance as a flat sawn board, with wavy, irregular grain. All veneer cut for inner plies and all plies for structural softwood plywood are cut in this manner. Plain sliced veneers are cut much as plain sawn boards are cut, and plain sawn veneer gives the appearance of plain sawn lumber, with inverted “V” or “cathedral” patterns in the grain. As with plain sawing, some slices of plain sliced veneer will look like quarter sliced or rift sliced veneers. These are sometimes inferior, however, because they come from the outside of the log. Veneer can also be quarter sliced or rift sliced, analogous to quarter sawing and rift sawing. This gives a straight grain pattern. The piece of log from which the veneer is sliced is called a flitch. Veneer slices are always kept in the order that they were sliced from the flitch, so that adjacent slices can be matched when the veneers are 30 Table of Contents Notes and Reflections While Shaving Wood 5-3 assembled into plywood. Rotary cut veneer usually covers the entire face in one piece. Rotary-cut veneers are cheaper to make and usually cost less. A surface covered with plain-sliced veneer may resemble a series of glued-up boards. This type of veneer is somewhat more expensive than rotary cut. Slip-matched veneer has consecutive flitches as they come off the log butted up side by side. Book-matched veneer uses consecutive slices from the flitch, but every other one is flipped over to make a mirror image, like the right and left pages of an open book. least 70 pounds. Core, Density and Adhesive ADHESIVE CORE Four types of hardwood plywood are recognized based on the type of adhesive used. Technical Type and Type I is made with waterproof adhesives that can withstand prolonged wetting. Type II is made with moisture-resistant adhesives. Type III is less moisture resistant than type II. The water resistance of the adhesive does not address the resistance of the core, however. A particleboard core will swell when wet, so it would not matter much whether the plies delaminated or not. The core of plywood is what occupies space within the face and back plies. All plywood has a second layer of veneer under both the front and back, with grain direction perpendicular to the surface, called a crossband. It is usually an inferior veneer, at least in appearance, and may not even be the same species as the face. This fact, and its perpendicular grain direction, is why you cannot sand or plane out deep scratches in the face veneer. Within the crossbanding, the core of a plywood sheet may be more veneers, solid lumber, or a composite material like particleboard. DENSITY Hardwood plywood species are divided into three density categories based on oven dry weight and volume at 12% moisture. This may seem obscure, but the density translates into weight, and may tell you something about what is inside the face plies. For example, a 4 x 8 sheet of ¾" oak plywood should weigh at [(3456 cubic inches) x (0.56 specific gravity) x (0.0361 conversion factor) = (69.9 pounds)] You are going to need a helper to unload that on a windy day. Category A, the densest, has a specific gravity greater than 0.56, and includes, such species as beech, birch, hard maple, and oak. Category B has specific gravity from 0.48 0.55. It includes species such as cherry, chestnut, soft maple, lauan, mahogany. Category C has specific gravity less than 0.48. It includes many softwood face plies, including red cedar. Plywood Grades Grade of hardwood plywood is all about appearance. Each face should be specified when choosing hardwood plywood. Plywood grades are (best to worst) A, B, C and D for the face ply. The back grades are (best to worst) 1, 2, 3, and 4. Plywood grades of A1, A2, B1, B2 are a good two sided product and would be used where both sides are going to be seen. The back side (number grade) may have a few more blemishes than the front. The premium 31 Table of Contents 5-4 Hardwood Plywood grade (A) should have matched slices. A3, B3, C3 are good for work that will be seen from only one side. The back side must be clean, although there may be some stain or discoloration on the back. The back does not have to be the same species as the front. Use these grades for drawer bottoms, bedside table sides, etc. A4, B4, C4 are a one-side-good product. The back may have open knots, filled knots, bad veneer joints or seams. This material would work well for drawer bottoms, and such. Official hardwood plywood standards are published by the Hardwood Plywood & Veneer Association (HPVA). They are proprietary and copyrighted. The following is a paraphrase taken from the website of a plywood manufacturer. Face Grade Description AA Architectural grade. The best quality face grade for high-end uses, such as architectural paneling, doors and cabinets, case goods and quality furniture. Generally available only by special order. A Excellent appearance, as for cabinets and furniture. Select grade veneer for quality and color. A select veneer is composed of entirely heartwood or sapwood and is matched for both grain pattern and color. If spliced, leaves must be book-matched for a pleasing effect of color and grain. Minor infrequent burls, pin knots and inconspicuous small patches are allowed. Frequency of defects depends on species. B Natural characteristics and appearance of the species are desirable. B grade is composed entirely of heartwood or sapwood but is matched for color only. Similar to A grade, but allows more numerous and larger burls, pin knots and color streaks. C Allows more numerous and larger burls, pin knots, color variations and rough-cut veneers. C grade is sound and smooth. D Sound grade free of open defects, but allows unlimited and more repaired defects than the C grades. E Sound grade with all repaired defects allowed. SHOP Minor imperfections, a minimum of 85% useable. May be higher grade with a broken corner, a small area of void, or a sanding error. Back Grade Description 1 Allows color variation, no large knots, tight knots cannot exceed 3/8" in diameter. Core laps are not permitted. Worm holes and splits must be filled. 2 Color is not a consideration. Sound knots cannot exceed 3/4" in diameter. Repaired core laps and repaired knots permitted. 3 Knot holes of 1" in diameter are permitted 4 Reject back grade not sound. Allows many open defects Three Other Sheet Goods HARDBOARD, Hardboard is also called Masonite or high-density fiberboard (HDF). It is an engineered wood product similar to particle board and medium-density fiberboard, but is denser and much stronger and harder. The density of hardboard is usually about 50-65 lbs. per cubic foot. Masonite has one rough and one smooth side. It is available in c" and ¼" thicknesses. I use it for making jigs, templates and drawer bottoms in shop furniture. MEDIUM DENSITY FIBERBOARD Medium density fiberboard (MDF) is an engineered wood product formed by breaking down hardwood or softwood into wood fibers, combining it with a resin binder, and forming panels under high temperature and pressure. Typical MDF has a hard, flat, smooth surface that makes it ideal for veneering, as there is no underlying grain to telegraph through the thin veneer as with plywood. MDF is generally denser than plywood. Furniture made with it can seem very heavy. It does not contain knots or rings, making it more uniform than natural woods during 32 Table of Contents Notes and Reflections While Shaving Wood cutting and in service. It is softer than plywood and saws easily, but the sawdust may be dangerous. MDF contains urea-formaldehyde which is a probable carcinogen. Its sawdust may cause allergy, eye and lung irritation. It tends to dull blades more quickly than many woods MDF may be glued, doweled or laminated, but smooth-shank nails do not hold well. It may split when woodscrews are installed without pilot holes, especially into the edge of a board, when it may delaminate. Typical fasteners are T-nuts and pan-head sheet-metal screws. Often, pilot holes have raised dimples around them and should be lightly countersunk to allow two pieces to mate well when screwed together. MDF is often cheaper than plywood, but it is heavy and breaks more easily than plywood. It swells and breaks if it gets wet with water. Painting with latex paints is difficult due to rapid water absorption. Most finishes appear uneven and nail holes tend to pucker somewhat. MDF is used in commercial furniture, from low-end kit furniture to medium high end. I use it only to make jigs when I need a dead flat surface. 5-5 MELAMINE-COATED FIBERBOARD Melamine-coated fiberboard is useful for jigs. Melamine is not adherent to PVA glues, even less so after I wax it with paste wax. I use it to clamp veneer when I am crossbanding. A piece of melamine-coated Masonite is a good runner underneath a shooting board, for the plane to run on, a slippery, flat surface for fine cutting. I think it is too slippery to serve as a general purpose work surface. References ANSI/HPVA HP-1-09 American National Standard for Hardwood and Decorative Plywood. Schramm, A. A Complete Guide to Hardwood Plywood and Face Veneer (Google eBook) Purdue University Press, 2003 Wikipedia monographs on Hardboard and MDF (January, 2014) 33 Table of Contents Notes and Reflections While Shaving Wood Chapter 6 -Wood Species and Finishes for Outdoor Furniture If you will apply oil stain, finish sand with 180 grit, with the grain, by hand. Unfinished redwood, cedar or cypress, (heartwood in particular) left outdoors, will weather to a silver-gray patina. It may also check and raise splinters on the surface and so require periodic smoothing if used for furniture. Wood will last indefinitely if it is kept dry and out of sunlight. Water and strong sunlight cause wood to fade, split, warp and rot. Weather-Resistant Species Redwood, cedar, and cypress are decay resistant softwoods. Old-growth cypress and redwood are significantly more decay resistant than new-growth, and heartwood is more resistant than sapwood. Pressure-treated pine lumber (treated with ACQ or copper azole) is very weather resistant, of course, but is usually much too wet when purchased. It might need to be stacked and dried under cover for months before being used for outdoor furniture or doors. Among hardwoods, teak is at the top in appearance and durability but is very expensive and very hard on tool edges. The high price of teak has made other tropical hardwoods popular, such as, ipé, jarrah and eucalyptus. All of these woods are dense, durable, and stand up very well to weather for decades. White oak (not red oak) is decay resistant. Other domestic hardwoods that stand up well outdoors are black locust, Osage orange, mulberry (all hard to find) and black walnut. American mahogany is used for outdoor furniture, if you can find it and afford it. (African mahogany is only moderately resistant to decay.) (See the FPL Wood Handbook) CHOICE OF FINISH The best way to protect wood exposed to sun and rain is to apply paint or a clear finish that blocks sunlight and moisture and that can itself withstand sunlight, water, and temperature changes. Paint is probably the most practical finish for furniture that will be left out in the sun and rain or a south-facing exterior wooden door. (Think of an old house with wood siding. If it is kept painted, the wood may be preserved for hundreds of years.) If you paint, first prime the wood with an exterior grade wood primer. Zinsser 1-2-3 is excellent, but will not reliably prevent resin bleed-through from softwood knots. Softwood knots must be primed with at least one coat of shellac or shellac based primer or discoloration will bleed through the paint. Although some wood dyes claim that they are light-resistant, I believe that all dyes fade somewhat. I recommend that only pigment stain be used on exterior wood that will be exposed to sunlight for extended periods. As will be clear in a moment, applying real protection to outdoor furniture is a laborious process. One easier option is to apply oil, e.g., linseed oil. Oil finishes, including commercial exterior oil stains (deck and siding stains) do not protect wood very Finishing A final sanding with 120-150 grit should be adequate if you plan to apply varnish or paint. 6-1 Table of Contents 35 6-2 Outdoor Wood long, although they may put a pigment on the surface. You might need to re-apply the oil every year, depending on conditions. Also, oil finishes won’t protect the wood from spills, skin oils, etc. I’d use an exterior oil stain product only if it could be over-coated with clear finish. Opaque oil stains are just weak paint, so why not use paint? The only clear finishes that contain enough UV absorbers to do any good are higher-quality marine spar varnishes, but they are expensive and require many coats. The finish itself can still deteriorate, even as it protects the wood underneath. This means that the finish may need to be renewed periodically, every five years or so at best. Alkyd spar varnishes are preferable to polyurethane spar varnishes because the alkyd resins tend to be less brittle than polys. Possibly, staining with pigmented stain (oil stain) before applying spar varnish would also help to filter out UV radiation. According to tests done by Fine Woodworking magazine, no clear finish prevented discoloration and damage to wood left outdoors for a year, except for seven coats of Epifanes High Gloss Marine Varnish. Next best was four coats of McCloskey Man O War Marine spar varnish. Watco Oil and water based finishes were of little value. Other tests tend to confirm that Epifanes clear varnish and McCloskey Man-o-War Marine Spar Varnish usually perform at or near to the top of one-part spar varnishes and rival two-part epoxies. Both varnishes are long-oil varnish with phenolic-modified alkyd resins. Total solids are 62% for Epifanes and 51% for McCloskeys. These facts matter because (1) Long-oil varnish tends to dry to a more flexible but softer film than short-oil varnish. (2) Phenolic-modified alkyd resins tend to dry to a more flexible film than urethane (polyurethane) resins. Flexibility is important in exterior applications because wood tends to expand and contract more outdoors with temperature and humidity changes. (3) The total solids constitute what is left after the varnish dries. 62% is higher than most other spar varnishes. Some manufacturers are reluctant to divulge their products’ solids content, but it is a major indicator of what you get for your money. I suspect that the resins, UV light blockers and total solids explain the excellent performance of these products. To be fair, Epifanes recommends eight coats, much more than most other spar varnishes. McCloskeys recommends 2-4 coats. A properly done side-by-side product comparison should follow manufacturers' recommendations. So part of Epifanes superior performance must be due simply to how much varnish is laid down between the sun and the wood. It may be that eight coats of other high-quality spar varnishes (e.g., McCloskey's) would perform as well. Eight coats, with 24 hours of drying time between coats, takes some persistence and patience. Also, multiple coats of these products will darken wood somewhat. The label instruction to thin the first coat of Epifanes is one you should follow for any spar varnish. The varnish soaks into the wood, sealing the pores. I go back over the first coat immediately until every part of the surface stays wet. Epifanes is very thick (almost syrupy). It requires good brush technique to apply unthinned. It levels brush marks fairly well but lap marks can be a problem because it is so thick. It's very important to let each coat cure before you apply the next coat. Varnish dries 36 Table of Contents Notes and Reflections While Shaving Wood in two stages. First the solvent evaporates and then the varnish cures (polymerizes). Polymerization can actually take days. Until the varnish has cured, it remains partially soluble in mineral spirits and can be softened by subsequent coats. Furthermore, no time is saved because subsequent coats simply retard the curing of earlier coats. It is important to sand lightly between coats with 180 grit paper. These products sand well. If you do not get fine dust while sanding a coat of varnish, if it tends to roll up into little balls, you should stop and wait another 12 hours. If you apply more varnish before the varnish has cured sufficiently you may get a gooey mess. It’s fine to apply the final coat(s) unthinned, to maximize the build. On the other hand, however, after many coats of varnish, it seems OK that the last coat or two could be thinned to the consistency of wiping varnish. This will be easier to apply smoothly and will tend to self-level better than full-strength varnish. PERSONAL EXPERIENCE My experience with furniture left out in the Florida sun and rain is that 2-3 coats of polyurethane spar varnish will protect the wood for a year or two. Three coats of top quality exterior marine paint did not protect pine furniture 6-3 exposed to an irrigation sprinkler in addition to sun and rain. Water penetrated from the end grain and began to rot the wood after two years. I think it was my choice of wood that was to blame, not the paint. But I should have sealed the end grain. Eight coats of Epifanes protected cypress chairs for 10 years before the finish started to fail and the wood began to darken. The chairs were left outside in Florida yearround, in partial shade. References Tom Begnall, Torture Test for Outdoor Finishes, FWW May/June 2009 Bob Flexner, Protecting Exterior Wood, American Painting Contractor, Douglas Publications, April 1999. Forest Products Laboratory. 1999. Wood handbook—Wood as an engineering material. Gen. Tech. Rep. FPL–GTR–113. Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Laboratory. https://www.fpl.fs.fed.us/products/ publications/specific_pub.php?posting_id= 18102&header_id=pNew Text 37 Table of Contents Notes and Reflections While Shaving Wood 7-1 Chapter 7 – Project Planning and Design Design is the fueling station on the way from dreaming to doing C areful plans make woodworking a pleasure. My first serious project followed plans from a magazine. The plans were accurate, and they allowed me to concentrate on accurate cutting and assembly. That, in turn, contributed to the success of the project and the pleasure that I took from it. I still prefer published plans for anything I want to build, although I often modify and always verify them as I go along. Most of the time, however, I must draw original designs to fit unique circumstances or client wishes. Even with experience, developing an original design can be as challenging as the actual woodworking. Design is a bit easier if you “walk” yourself through a series of steps, and avoid jumping ahead to decisions for which you are not quite ready. By design, I mean detailed planning to reach my objective of attractive, useful and durable furniture or a decorative object. The term has taken on an arty connotation which really applies only to the first step, the concept. Whether a concept is very original or very imitative, it deserves to be designed. Steps in Design The first step should develop a concept. This word has also taken on connotations of high originality, but it simply means an idea formulated as a picture or list of attributes. These include dimensions, especially if a piece has to fit in a particular space; style (a style that you will actually be able to execute well); and functions and features (doors, drawers, shelves, etc). If the piece will have a function, imagine it in use. Of course, this stage should include discussions with the client, another fancy collective word for customer, family member, or friend. Discussions with a client are usually fun for me, but if we can’t get along at this stage, maybe I should not take the job. Although we strive for perfection, the closest we can come to perfection in fine furniture making is when the client thinks our piece is perfect. (See Chapter 51, Imperfection.) A design that is exactly what the client wants will go a long way toward obtaining that verdict. Will it suit the client’s needs? Some furniture, e.g., a side table, has to fit the chair or bed beside it. If it is an inch or two too high, the client may hit his elbow on the table every time he sits beside it. Other furniture, e.g., a dining table or chair, has to fit people. There are standard dimensions for that, but you should fit it to the client. Illustrated Cabinetmaking by Bill Hylton is very helpful at this step – it’s like a catalog of joinery designs, details of construction methods, and recommended dimensions. It helps me to decide many details. Will the items intended to go into the piece actually fit? Will the drawers or doors open in the space available? Will people, especially children, be hurt if they bump into sharp corners? Read books and magazines and browse the web for ideas. You may find a design that suits you or that you can modify. When I have at least a fuzzy idea of what I want, I draw a rough pencil sketch, even 39 Table of Contents 7-2 Planning and Design though I am no draftsman. An overall sketch, with notes, will clarify and then fix the objective in my mind. The second step is to think about joinery and design details. My objective is to make attractive, useful and durable furniture that meets a client’s needs and desires. Of course, I have to be able to build it. Therefore, I accept some conventional constraints. During this second step, you should also begin to think about assembly. How will you put the pieces together? Like the joke about painting yourself into a corner, you can design a piece that cannot be assembled. A bridle joint instead of a mortise and tenon can be crucial in some places, and it is better to recognize this early (before it has glue on it). The third step is to produce working plans and perhaps a rendered three dimensional drawing. Sketchup® is very convenient for that. I can easily produce professional looking drawings with shadows, textures, etc. to show to a client. Drawing and designing go back and forth as I see the results of an idea take shape on paper. Sketchup is not exactly easy to learn, but it is preferable to paper and pencil. An almost full-featured version is available for free. Sketchup is built for speed and (I think) mainly for large-scale architectural work like houses or shopping centers. It has many features, and the Sketchup interface is surprisingly subtle and intelligent. It will try to infer what you want to draw, and show you its inference. When you draw a simple, straight line, for example, the color of the line will show you which axis it is following. That’s a terrific feature, but you have to know what to look for. I now draw all of my woodworking designs in 3D with Sketchup. I calculate most dimensions rather than try to read them from the drawing. I draw a piece of furniture as if I were building it. That is, I draw the individual parts. While I am drawing, I am thinking about how I will cut the parts and assemble the piece. I think out details before I cut my first board. Many details of construction and assembly are difficult to visualize and anticipate. Drawing helps me to do that. Making a detailed drawing is very tedious for me. It’s not the drawing itself that I find tedious, however. Once you learn to use it, Sketchup makes that part fairly easy. The tedious part is working out the details, including dimensions. I endure the tedium because careful design helps me to think through the details of cutout and assembly, to avoid surprises that can really waste time (and money). Some experienced woodworkers say that they don’t make careful plans because the piece never comes out exactly like the plan. They expect that they will have to adapt on the fly, and then the time spent in making plans would have been wasted. I don’t assume that my careful plans will be executed down to the last 1/32". For example, I make and measure the actual opening before I cut out the pieces for the door or drawer. But I design as if the piece will be made exactly as I drew it. Unless the piece is quite simple, the final design step is a cut list: a list of every part with material type, dimensions, and milling instructions. A specific cut list lets me figure out the material requirements for a project and helps me to avoid confusion while I am cutting. Finally, a cut list lets me use a program like CutList to lay out the pieces on the stock, and figure out how much wood I 40 Table of Contents Notes and Reflections While Shaving Wood need to buy.d I do not, however, cut out all of the pieces in advance. Many pieces have to be cut to fit. Lesson Learned When I first drafted this article, the memory was fresh in my mind of a project that I did not design carefully enough. It was moderately complicated, my most challenging until that time. It had many unfamiliar elements, maybe too many. I got distracted by the details of learning Sketchup and some difficult joinery. Also, I was impatient to get started. Instead of thinking through the details, I left many important decisions for construction time. As a result, I made my mistakes in wood rather than on paper. That significantly lengthened the construction time. For example, I did not plan out and draw details of how the edge trim would fit together, how I expected to attach it, and how I was going to assemble the project. Then, having recognized more than the usual number of potential improvements in my original design, I was in the unhappy position of finishing the piece, knowing that I could have designed a better one. But that was just the beginning. Because I spent too little effort on design, and especially on planning the assembly and glue-up I spent way too much effort on construction. Even though I dryassembled it, after the glue was on, I discovered that I could not apply clamp pressure to some joints. As a consequence the glue-up was a nightmare. To say that my failure to plan reduced my enjoyment of the project would put it too mildly. I had a near d Cutlist http://www.delphiforfun.org/Programs/CutList.htm 7-3 meltdown during glue-up and almost scrapped the project, to which I had given hundreds of dollars and weeks of time. I made my mistakes in wood instead of on paper Ben Franklin said something like, “Experience keeps an expensive school, but fools can learn in no other.” Taking the time to design a project carefully will – I hope – prevent me from repeating this experience. Summary – Six Essentials Before I buy the lumber and get to work, I need to have decided six important issues: 1. The final appearance of the finished item(s), overall dimensions, suitability and acceptability to the client. 2. The dimensions and milling of each part, how it will fit the others, whether each joint will look right, whether it will be strong enough, and whether I can make it with my present tools, skill level and available time. 3. What kinds of stock I’ll use: species, hardwood, plywood, etc., and how much of each to buy. 4. Any special requirements for stock, e.g., grain direction or appearance, board widths. 5. Necessary new tools and hardware, e.g., router bits, hinges, etc., especially if they have to be ordered by mail 6. The rough order of assembly, whether to make sub-assemblies first, 41 Table of Contents 7-4 e.g., panels. References 7. How I’ll clamp it, whether I’ll have the necessary clamps, cauls, etc. Hylton, Bill. Rodale’s Illustrated Cabinetmaking. Emmaus, PA, Rodale Press 1998. Editors of Fine Woodworking. Practical Furniture Design: From Drawing Board to Smart Construction. Newtown, CT. Taunton Press, 2009. Chopra, Aidan. Google Sketchup for Dummies. 2007. Hoboken, NJ Wiley Publishing. (I’m not a dummy and I don’t like a marketing strategy that appeals to low selfesteem. But this is a very accessible and useful book. Sketchup itself has excellent online help and tutorials.) 42 Table of Contents Notes and Reflections While Shaving Wood Chapter 8 – Designing for Moisture Content W wet. ood swells and shrinks with relative humidity (RH). A door, drawer, panel, or table top made of solid wood may shrink or expand as RH changes seasonally, especially from summer’s high RH to the desert conditions inside a heated house in wintertime. Careful fitting may be for naught if a drawer sticks, a panel shrinks and cracks, or a table top strains against its attachments. In most cases, design must account for seasonal or environmental changes in wood dimension, usually by allowing the wood to move without damaging the item. As the wood dries beyond the FSP, it loses weight, gains strength, and shrinks along both width and thickness (across the grain), but not along its length. It is important to appreciate that air-dried or kiln-dried lumber still contains moisture. Lumber exchanges water with its surrounding air. It exists in a slow equilibrium with the moisture in the atmosphere. This equilibrium is called the equilibrium moisture content (EMC) of the wood. The MC of kiln dried wood varies greatly, but suppose that wood came out of the drier at 6% moisture. As the wood sits in a lumberyard or in your workshop, the MC may go up or down depending on RH. In northeastern Florida, the MC of kiln dried wood will increase in the air. If the RH is high enough, a plain sawn board may distort (cup), especially if you plane it, because it changes dimension about twice as much in the direction of the annual growth rings (tangentially) than it does across the growth rings (radially). (Compare the cross sections in Figure 1 at the top center and at the left). Also, in high RH the freshly planed surface may be drier than the factory surface. After a while the cup may subside as the planed surface absorbs moisture. In fact, a common cure for a cupped board is to lay it out in the sunlight with the convex side up. The sunlight may dry the longer surface, shortening it. A quartersawn board may shrink in thickness but not distort. Theoretically a rift-sawn board would shrink along a diagonal. This is why quartersawn and rift sawn boards are said to Water in Wood Freshly harvested (green) wood is actually wet with sap. If you hit it with a hammer, you can see the sap squeezed out of it. Initial air or kiln drying removes the water inside the wood cells (called free water) but leaves water bound to the cell walls. This degree of dryness, when the last of the free water has gone, is called the fiber saturation point (FSP). FSP varies by species but is Figure 1. Grain direction of boards, and possible changes in dimension with moisture content typically around 35% by weight. Wood whose moisture content (MC) is above the FSP is still considered green, even if it is no longer 8-1 Table of Contents 43 8-2 Designing for Moisture Content be more dimensionally stable. Accounting for MC in Design So, relative humidity (RH) affects equilibrium moisture content (EMC), which affects wood dimensions. The degree to which RH affects wood dimension depends mainly on the grain direction (radial or tangential) and wood species. The relationship between dimension and RH is nearly linear over a wide range of RH, but the calculations become more approximate at very high RH. Suppose that you are designing a raised panel door like that in Figure 2. (Detail on how to make these is covered in Chapter 36, on raised panel doors.) This door consists of a frame with a ¼" wide groove cut around the inside. Let’s say the panel is white oak. Its thickness tapers from ½" down to ¼" to fit in the grooves. What size should the panel be, and how deep should the grooves be? The distance between the stiles is 8" and the distance between the rails is 13". The first thing to notice is that our door has grain directions perpendicular to each other. True, the joints between the rails and stiles are cross-grain, but the big issue is the panel, which is cross-grain to the rails. The width of the panel itself will change with changes in RH, but the width of the frame will not. Ignoring wood movement, you might think that gluing the panel into the groove all around would really reinforce the door. That’s true. It would make the door much stronger, for a while. This would be a good choice if the panel were plywood. But, since the panel is solid white oak, its dimensions will change with changes in RH. From what we said in the first section, you can predict that the length of the rails and stiles is long grain, and will change negligibly, while the width of the panel will change appreciably. Wood shrinkage could conceivably allow the panel to fall out of the frame, while expansion could warp the door, or even tear it apart. Often the largest piece of solid wood in a table, bureau, etc. is the top. A solid wood top should not be screwed to the base directly. Wood movement across a wide top may be substantial and may, at best, loosen the screws. Likewise, although it may seem logical to reinforce a glued up panel with cross-members glued to the underside or with breadboard ends, these must allow for expansion across the width of the panel. Calculating Wood Movement THE SIMPLEST WAY Choose the maximum change in EMC that your piece may experience. A typical figure would be 6-10 percentage points, corresponding to about a 30 - 50 point change in RH. Look up the coefficient of change for the grain orientation and wood species in Table 1. For plain sawn white oak that’s .0037. Multiply these by the 8" width of the board: 6 x .0037 x8" = 0.18" (about 3/16") 10 x .0037 x 8" = 0.296" (about 5/16") (I rounded up to the nearest 16'th of an inch.) Depending on your assumption about how much change in EMC the panel will experience, it may change width as much as 5/16" with seasonal variation in RH. Now, the panel has to be at least 8" wide or it may fall out of the frame. So, specify it to be 8-5/8" wide, giving a 5/16" edge on both sides. If it shrinks 5/16" it will still be larger than the opening. Adding 5/16" for expansion gives 8e + 5/16" = 9d” of space for the panel. The grooves in the stiles should be 11/16" deep 44 Table of Contents Notes and Reflections While Shaving Wood (half of 9d - 8). The grooves for the rails do not need clearance for wood movement. If you specify 13½" length for the panel, the grooves can be ¼" deep, maybe a bit deeper. You can glue the panel in the frame near the center of the top and bottom, where wood movement will be minimal. GETTING SERIOUS we want. At least, we can take note of the RH when we are making the door. If it is winter, then the door is about as narrow as it will get, and we need to allow mainly for expansion. If it is summer, the door may be as wide as it will get and we need to allow mainly for contraction. We can take this a step further with a moisture meter that we can use to estimate the MC of the wood. If it is high, we can assume that the wood may shrink somewhat, and allow mainly for that. And vise versa. Finally, we can sharpen our estimate of seasonal changes in EMC. All of the following discussion boils down to a simple idea. Figure out what range of EMC applies to your work, use it, and save it for the future. 1. Figure 2. Raised Panel Door 2. In the preceding discussion, we may have made the clearance much larger than necessary. We allowed for both 5/16" expansion and contraction, as if we did not know what the EMC would be when the door was constructed. We can do better than that, if 8-3 Table 2 is based on a table by Bergman in Wood Handbook that lists EMC for each month of the year, based on outdoor RH, for many US cities. It condenses the original detailed information to the annual mean and range of EMC. No listed city has a maximum change in EMC more than 10 percentage points, and most have a change around 2 percentage points. Even though RH is often artificially controlled within limits comfortable for people, the 6 or 10 percentage points assumed in our simplified approach above may be higher than necessary. If you know the range of RH in your locality, you can estimate the range of EMC change. The relationship between RH and EMC is like a lazy S curve – not linear over the possible range of RH. Plots of this relationship are available in Hoadley’s book, and you can read the data there. However, 45 Table of Contents 8-4 Designing for Moisture Content from point (0,0), where both EMC and RH are zero, to about (80, 20) where RH is 80% and EMC is 20%, a straight line is within +/- 2 percentage points of EMC. So we can say, approximately, that when RH is less than about 80-90%, EMC . 0.2 RH. 3. After 90%, each percent increase in RH increases EMC by about one percentage point, up to the saturation MC about 30-35%. The iso-EMC maps in Figures 3 and 4 show EMC for large regions of the country in July and January. These maps originally appeared in U.S.D.A. Forest Service Research Note FPL 0226. Effect of Finishes The finish applied to a wood product, especially to the end grain, will slow changes in MC due to RH. With seasonal fluctuations in RH, the finish can prevent a piece of wood from actually reaching its EMC before the RH changes again, and it certainly prevents surface swelling. Multiple coats of oil based varnish or shellac offer the best protection against water vapor. Shellac offers excellent protection against water vapor but is somewhat permeable to liquid water. Drying oil, and furniture wax offer little protection. It is best to apply the same number of coats of the same finish to both sides of a wide board like a table top. Otherwise, different rates of vapor penetration may cause an imbalance in the MC from one side to the other, which may cause it to warp. In practice, I rarely apply more than three coats of varnish to the underside of a table top. I find that is a sufficient vapor barrier. Since I am rubbing between coats on the top surface, the build of six rubbed coats on the top surface may be about as much vapor barrier as three un-rubbed coats on the bottom surface. Eventually, wood moved to a very different environment will reach its EMC or at least reach a new cycle of highs and lows. For example, a piece of furniture made in Atlanta (mean EMC= 13) and then moved to Las Vegas, NV (mean EMC=6) would eventually shrink and then find a new range of expansion and contraction in Las Vegas. Table 1. Dimensional change coefficients (CR, radial; CT, tangential) for shrinking or expanding within MC limits 6%- 14% (From Bergman) Species CR CT Alder(Red) 0.0015 0.0026 Ash (White) 0.0017 0.0027 Aspen (Quaking) 0.0012 0.0023 Basswood (American) 0.0023 0.0033 Beech (American) 0.0019 0.0043 Birch (Yellow) 0.0026 0.0034 Butternut 0.0012 0.0022 Cherry (Black) 0.0013 0.0025 Fir (Balsam) 0.0001 0.0024 Mahogany 0.0017 0.0024 Maple (Red) 0.0014 0.0029 Maple (Sugar) 0.0017 0.0035 Oak (Red) 0.0016 0.0037 Oak (White) 0.0018 0.0037 Pine (Eastern White) 0.0007 0.0021 Pine (Longleaf) 0.0018 0.0026 46 Table of Contents Notes and Reflections While Shaving Wood Species Pine (Ponderosa) 0.0013 0.0022 Pine (Sugar) 0.0010 0.0019 Poplar (Yellow) 0.0016 0.0029 Sycamore (American) 0.0017 0.0030 Teak 0.0010 0.0019 Walnut (Black) 0.0019 0.0027 Table 2. EMC (%) for outside conditions in several U.S. locations *From Bergman State City AZ CA CO DC FL GA HI ID IL IN IA KS KY Phoenix Los Angeles Denver Washington Miami Atlanta Honolulu Boise Chicago Indianapolis Des Moines Wichita Louisville 12 month Range (MaxMean Min) 7.18 4.9 13.74 3 10.07 1.6 11.88 1.5 13.46 2.2 12.99 2.4 11.59 2.7 10.83 7.9 13.30 2.7 14.08 2.9 13.38 2.5 12.76 2.4 13.27 2.1 Figure 3. Approximate Distribution of Indoor EMC in January LA ME MA MI MN MS MO MT NE NV NM NY NC ND OH OK OR PA SC SD TN TX TX UT VA WA WI WV WY New Orleans Portland Boston Detroit Mpls–St. Paul Jackson St. Louis Missoula Omaha Las Vegas Albuquerque New York Raleigh Fargo Cleveland Oklahoma Portland Philadelphia Charleston Sioux Falls Memphis DFW El Paso Salt Lake City Richmond Seattle–Tacoma Madison Charleston Cheyenne 8-5 CR 14.55 13.15 12.20 13.51 13.21 14.34 13.46 13.10 13.43 6.18 8.53 11.98 13.13 13.81 13.54 12.59 14.20 12.25 13.37 13.64 12.94 12.94 8.11 10.63 12.78 14.44 14.06 13.21 10.28 CT 1.3 1.9 1.5 2.8 2.7 1.4 2.5 7.8 2.7 4.5 4.3 1.6 2.8 3.3 2 1.7 5.7 1.8 2.2 2.7 1.6 2.3 4 7.8 2.5 4.3 3.2 2.9 1.1 *EMC values were determined from the average of 30 or more years of relative humidity and temperature data Figure 4. Approximate Distribution of Indoor EMC in July 47 Table of Contents 8-6 References A useful Wood Movement Reference Guide, with instructions and many more wood species than Table 1, is available from Lee Valley Tools. U.S.D.A. Forest Service. Research Note Fpl - 0226 1973 Moisture Content of Wood in Use Forest Products Laboratory, Forest Service U.S. Department of Agriculture movement. Fine Woodworking Nov-Dec 2006 Bergman R Drying and Control of Moisture Content and Dimensional Changes Ch 13 In Wood handbook—Wood as an engineering material. General Technical Report FPL-GTR-190. Madison, WI: U.S. Forest Products Laboratory. 2010. Hoadley, R. Bruce. Understanding Wood. Newtown, CT. Taunton Press, 2000. Becksvoort, C. Stop guessing at wood 48 Table of Contents Notes and Reflections While Shaving Wood Chapter 9 – Estimating the Strength of a Wood Furniture Design O joint and how well it was made) influence strength for a given set of dimensions. Many experienced woodworkers invoke the eyeball rule, which goes something like this: “You have seen a lot of furniture in your life. If a piece looks strong enough, it probably is.” This is satisfactory if you assume that the conventional wisdom (i.e., whatever everybody else does) is correct and that experience can actually encompass all of the variables that make up strength. I can understand why people would prefer the eyeball rule. Precise engineering of wood construction is complicated far beyond the interest level and mathematical abilities of a one-off custom furniture designer. Detailed formulas are important in architecture and civil engineering, but in furniture construction there is less risk of injury or huge financial losses than, for example, when a floor sags or a bridge collapses. Rough approximations should usually be close enough. Furthermore, overbuilding is often reasonable from both esthetic and cost perspectives. This allows the engineering calculations to be simplified considerably. Another argument in favor of the eyeball rule is that the strength of wood furniture depends on many factors in addition to the dimensions of its parts, for example, joinery, choice of adhesive, the species used, wood moisture content, and grain orientation relative to the stress. For these reasons, even complicated engineering calculations include a large safety factor when applied to architecture or civil engineering. After a brief study of this topic, I concluded that the most useful information for designing durable wooden furniture will come from bending calculations. The small amount ne of my wife’s ancestors evidently was a good woodworker, or perhaps commissioned custom made furniture. We inherited a rocking chair and a table that needed repair when we received them. The chair is so fragile that we really don’t want anyone to sit in it. The table is a bit stronger, but still fragile. We respect these pieces, but do not appreciate having furniture that we are afraid to use. I do not want anything I make to be somebody else’s problem like that. My quality criteria for a piece of furniture are appearance, utility, and durability. Utility and durability both involve the strength of the piece – whether it is strong enough to stand up to its intended use and inadvertent misuse. When I am designing a piece of furniture or checking somebody else’s design, I need to consider how a piece is likely to be used, whether it will be used by children, extremely large adults, etc. Then I need some guidance concerning the strength of the design, especially whether it can stand up to misuse. As I mean the term, misuse may be common. Examples would be an overloaded shelf; a table being stood upon, sat upon, or dragged across a carpet; a table leg kicked or hit with a vacuum cleaner; a chair being leaned backwards; or even a door being deformed by its own weight. The first three examples involve beam strength. The last four involve the strength of joints, mainly against racking or levering forces. Strength and appearance are sometimes at odds, if you prefer a lighter, cleaner look, as I do. Choice of wood and quality of joinery play a role here, because wood species and joint construction (type of 9-1 Table of Contents 49 Strength of Wood Furniture of bending that would be acceptable from the standpoint of appearance or use involves forces much smaller than those that might permanently deform or break the piece. Calculations of breaking point are even more difficult, but might help to provide a check on bending calculations. Finally, data on the breaking strength of joints might provide some additional design guidelines. Remember, however, that these calculations provide rough guidelines for “ideal” straight-grained typical hardwood with no defects. You can consult the Wood Products Handbook for data on specific species, but it’s probably not wise to depend more than that on engineering estimates. It would be safer to design in a safety factor of 50-100%. Please do not be put off by the complicated-looking equations. Even if you do not use them, study the relationships. This will provide a useful understanding of the strength of design options. For example eq 1 tells you that stiffness increases as the cube of leg width in the direction of stress. That will guide where to add wood to incrase stiffness. Table 3 compares stiffness and strenth of common species. Beam Stiffness Two designs of beams are especially important in wood furniture: the built-in beam and the cantilever. A table apron or rail glued into a leg at each end is an example of a builtin beam; so is a bookshelf fitted into dados at each end. A table leg that is supported only where it is glued into the apron or rail is an example of a cantilevered beam. (A third kind, a simply supported beam, would be like a board resting at each end on scaffolding or sawhorses.) Two kinds of beam loadings are 9-2 important: distributed and concentrated. A bookshelf with books all along its length is an example of a distributed loading. Somebody sitting on the side edge of a table would be an example of a concentrated load. A table leg being dragged across a carpet is an example of a cantilevered beam with its load concentrated at the end. The beams that exist within our furniture can be damaged in many ways. For example, they can bend until they split along the grain or even shear across the grain. For practical purposes, however, we normally would wish to keep the amount of bending (deflection) small for the sake of esthetics or usefulness. If we build furniture strong enough not to bend very much, we won’t have to worry about fracture or shear. CALCULATING STIFFNESS The full equations used to calculate beam deflection, given in the appendix to this chapter, are a bit daunting. As I said above, however, they can be simplified without losing their value for furniture making. Furthermore, a program called Sagulator, available on the web, is set up to calculate the sag of a shelf. (See the URL in References.) Sagulator uses the formulas for calculating beam deflection, so it can be used in other situations as well. It does not seem to lend itself to cantilever beam deflections, however. A simplified equation for a typical hardwood beam that is rectangular in crosssection and anchored at both ends, e.g., a table rail mortised into the legs, is D = W•L3 / [(24 x 106)•b•h3] (Eq 1) where L is the span (unsupported length) of the beam in inches, W is the weight in pounds applied to the center of the span, b is the 50 Table of Contents Notes and Reflections While Shaving Wood width of the beam (horizontally) and h is the height of the beam (vertically). For example, the estimated deflection (in inches) of a 36" long hardwood table rail that is 2½" wide and 1¾" thick, with a 200 lb brother-in-law sitting at its center is D = 200 • (36)3 / (24 x 106)•2.5•(1.75)3 = 9331200 / (321.5625 x 106) = 9.3312 / 321.5625 = .029 (about 1/32") For a beam that is circular in cross section, the simplified formula is D = W•L3 / (14.13 x 106)•d4 (Eq. 2) where d is the diameter of the beam. So, for example, the estimated deflection (in inches) of a 36" long by 2" diameter hardwood dowel, with a 200 lb weight applied at its center, is D = 9.3312 / 266.08 = .041" For the deflection resulting from a uniformly distributed load, just divide the centered load result by 2. For the deflection of a cantilevered beam with the load concentrated at the free end, e.g., a chair or table leg, multiply by 64. So, the calculated deflection of a 30" table leg, 2" in diameter, with a 50 lb force applied to the end, would be D = 50•(303)•64 / 266.08 x 106 = 86.4 / 266.08 = .32" (about 5/16") This seems like a significant deflection, but it might not exceed the bending capacity of the leg. That’s another issue, along with the 9-3 strength of the glue joint holding the leg. These are rough estimates. They assume a typical hardwood, with straight grain, no grooves, knots or other defects. e Approximations like these may be useful as rough guidelines for designing furniture. They also suggest some theoretical design strategies for controlling sag: a. D is proportional to the weight of the load and the cube of the span. b. D is inversely proportional to the cube of the vertical thickness of the beam (rather than, say, cross sectional area). So, roughly speaking, a 10% increase in span can be offset by a 10% increase in vertical beam thickness. c. For a given cross-sectional area (weight of beam, etc) stiffness is increased more by increasing vertical dimension h than total area or width b. d. For a given cross sectional area, (i.e., given volume of wood and weight of wood) a beam that is square in cross section is slightly stiffer than one that is round. A final note about vertical compression and buckling of legs. Hoadley (p.80) says that a vertical dowel under a compression load (like a chair leg) may buckle if the ratio of length to the thinnest cross-sectional dimension is greater than 11. So, theoretically, a leg 30" long will not e I don’t recommend that you rely on simplified calculations if you intend to use the minimum of materials or to design a critical application like a ladder. In addition, the weight of the material itself may be significant in such cases: the weight of the wood should be accounted for as a uniformly distributed weight. 51 Table of Contents Strength of Wood Furniture buckle under vertical load if it is 2¾" or more at its thinnest point. Breaking Strength Wood can break in many ways: it can shear across the grain, pull apart under tension, split parallel to the grain as annular rings slide past one another, and so forth. An overloaded shelf might break from bending stresses, while a frame subjected to racking forces might break from some combination of failure modes, e.g., a stile splitting and a rail shearing or tearing apart under tension. The Wood Products Handbook (WPH) gives tables for strength in various failure modes. Grain orientation also affects strength. Obviously, wood is much stronger in compression (parallel to the grain) than it is in bending. Think of a post being driven into the ground. A post being driven by a force distributed across its end can stand up to great force. A concentrated concentric load on the end of a post, however, can easily split the post, which is why we usually use a block of wood as a pad when hammering a post into the ground. Wood is stiffer and stronger against a radial load (perpendicular to the growth rings) than a tangential load, and weakest when the force is at an angle, e.g., as in rift-sawn wood. This is especially true for softwoods. According to the WPH, (Chapters 4 and 8) modulus of rupture reflects the maximum load-carrying capacity of a member in bending. It is proportional to maximum moment borne by the beam. It is an accepted criterion of strength, although it is an extrapolation of a formula beyond its strict validity (beyond the limits of Hooke’s law). The maximum stress due to bending moment for a simply supported pin-ended beam is given by 9-4 f b = M/Z = W•L/Z (Eq.3) where fb is bending stress, M is the bending moment (W•L), W is weight , L is beam length and Z is the beam section modulus. For a rectangular cross section, Z = bh2/6; for a circular cross section, Z = πd3/32). Values of modulus of rupture are tabulated in Chapter 4 of WPH and Appendix Table 2, below. The modulus of rupture for southern red oak is 1.09 x 10E4 lb/in2 (compare to modulus of elasticity 1.49 x 10E6 lb/in2). To calculate the weight that will theoretically bend a beam until it breaks, substitute the appropriate tabled value of modulus of rupture for fb and solve for W. So, in the first example above, what weight, applied to the center of a 36" long rectangular red oak table rail that is 2½" wide and 1¾" thick, will break the rail? W = fb•Z/L =10900 lb/in2 •(2.5• 1.752/6)/18 W = 772 lb (See Bowyer et al) Strength of Joints The strength of a joint depends on its design, the strength of the wood from which it is made, how well it fits together (i.e., how thick and uniform the glued surfaces are) and the type of adhesive used. Obviously, some joints, most notably the dovetail and sliding dovetail, are mechanically stronger, and depend less on the adhesive used, than a joint that could slide apart if it were not glued. The ability of wood to bend surely adds greatly to its ability to absorb transient forces without breaking. On the other hand, changes in wood dimensions with changes in moisture content 52 Table of Contents Notes and Reflections While Shaving Wood might gradually weaken joints over time. McKenna and Moore tested 18 types of joints. They glued the joints with Titebond III, clamped them for one hour, and let the glue cure for five days. All of the joints were made with ¾ x 2½ x 8" cherry pieces. Then they stressed them until they broke. The wood tended to break before the glue failed, but sometimes failed from a combination of wood rupture and glue failure. Many glue joints break that way, because the glue surface is often at least as strong as the wood. The strongest joints were half-lap, bridle, splined miter, and a beefy (d") mortise and tenon. (The M&T appears in the illustrations to be about 2" x 2", for about 4 in2 on each cheek.) These broke at 1600, 1500, and 1400 lb of force, respectively. So, consider the M&T joint, which failed on average at 1440 lb. That’s 1440 lb at a distance of 8", or 960 ft-lb. If the joint connected a table leg, say 30" long, the force required at the end of the leg to break the joint would be given by 8/30 x 1440 = 384 lb, quite a blow. It’s difficult to adapt these calculations to species other than cherry, or to joints of other dimensions or designs, but they give some 9-5 idea of the strength of hardwood furniture made with good joints. Table 1 is taken from Schofield’s article on the strength of adhesives. Titebond III and slow-set epoxy were the two strongest adhesives in the study. These data suggest that -1. the strength of the wood significantly affects the strength of a joint, i.e., adhesive strength may not be the limiting factor (at least for woods weaker than Ipé). 2. tight bridle joints seem to be stronger than snug or loose joints for maple and oak, but not Ipé. This contradicts conventional wisdom. Perhaps tight joints in maple and oak are not “starved” for glue, but Ipé is so stiff or hard that a tight joint actually is somewhat “starved.” 3. Slow set epoxy seems to make loose-fitting joints slightly stronger than Titebobnd III. That makes sense to me because epoxy does not shrink when it cures. By the way, the breaking strength of a bridle joint in cherry wood (from McKenna and Moore) was about 1500 lb.. The joints in the two studies may not have been of comparable sizes, but this is consistent with the idea that wood species is a significant factor in joint strength. 53 Table of Contents Strength of Wood Furniture 9-6 Table 1. Force Required to Break Bridle Joint, by Joint Fit and Wood Species (See Schofield) Adhesive Joint Fit Breaking Force (Lb.) Maple Titebond III Slow-set epoxy % of “Tight” Oak % of “Tight” 1843 Ipé % of “Tight” Tight 1842 Snug 1700 92 1822 98 2733 107 Loose 1593 86 1603 87 2525 99 Tight 1690 Snug 1680 99 1832 96 2712 112 Loose 1635 96 1557 82 2503 103 1908 APPENDIX Equation (1), above, is simplified from the following equation (Eq. A-1). According to the USDA Wood Products Handbook, (WPH) the deflection of a straight beam which has a constant cross section throughout its length is given by D = kb•W•L3 / E•I + ks•W•L / G•A' (Eq. A-1) where D is the beam deflection in inches, W the total load on the beam in pounds (perpendicular to the beam axis), L is beam span in inches, kb and ks are constants dependent upon beam loading, support conditions, and the location where the deflection is to be calculated, I is the beam moment of inertia (in.4), A' is the modified beam area (in2), E is the beam modulus of elasticity (psi), and G is the beam shear modulus (psi). f f 2554 The English (imperial) unit of force given in the WPH tables is lbf, pounds of force (not foot-pound). This is to reconcile physics and engineering, mass vs weight. From my viewpoint here on the Earth, this is an unnecessary fine point of physics. A lbf 2425 For beams having grain direction parallel to their axis, E = EL. A typical EL value for hardwoods is 1.5. (See Table 2 for some values of EL by species.) G is approximately equal to 0.1 EL. (Specific values for the elastic properties E and G are given in Chap. 4, Tables 4–1 and 4–2 of WPH.) For beams with flat-grained vertical faces, G = GLT, and for beams with edgegrained vertical faces, G = GLR). The modified area A' is given by A' = 5bh/6 for a beam of rectangular cross section A’= 9π d2 /40 for a beam of circular cross section The first term on the right side of equation A-1 gives the bending deflection and the second term, the shear deflection. The shear component is usually much smaller than the bending component (an order of magnitude less). That’s the rationale for is a plain old lb. 54 Table of Contents Notes and Reflections While Shaving Wood omitting the shear component from the simplified equations. Values of kb and ks for selected cases 9-7 of loading and support are given in the following Table 2. Table 2. Values of kb and ks for several beam loadings Loading Beam ends Deflection at kb ks Uniformly distributed Both built-in Midspan 1/384 1/8 Concentrated at midspan Both built-in Midspan 1/192 1/4 Concentrated at Free End Cantilever, one end built in, one end free Free end 1/3 1 Extracted from Table 8-1 in WPH The moment of inertia I of the beams is given by I = bh3/12 for a beam with a rectangular cross section and I = π d4 /64 for a beam of circular cross section where b is beam width, h beam depth, and d beam diameter. Therefore, Eq A-1 can be simplified to D = (1/192)•W•L3 / E•(bh3/12) = .0625W•L3 / E•bh3 (Eq. A-2) 6 and if we use an E value of 1.5 x 10 we get Eq 1. Eq. A-2 is useful if you want to use the specific E value for a particular species of wood. (See Table 3) Table of Contents 55 9-8 Table 3. Modulus of Elasticity and Rupture for Selected Woods Species (12% Moisture Content) Modulus of Elasticity EL (x106 lb/in2) Modulus of Rupture (lb/in2) Beech 1.72 14,900 Baldcypress 1.44 10,600 Cherry 1.49 12,300 Mahogany (True) 1.5 11,500 Hard Maple (Sugar Maple) 1.83 15,800 Soft Maple (Red Maple) 1.64 13,400 White Oak 1.78 15,200 So. Red Oak 1.49 10,900 White Pine 1.24 8,600 Yellow Poplar 1.58 10,100 Black Walnut 1.68 14,600 From Table 4.3 of WPH References Laboratory. Chapters 4 and 8. Thomas McKenna, Douglas Moore, Joinery Shootout Fine Woodworking Jan/Feb 2009 Jim L. Bowyer, Rubin Shmulsky, John G. Haygreen. Forest products and wood science: an introduction 2003 WileyBlackwell, Ch 10 Forest Products Laboratory. 1999. Wood handbook—Wood as an engineering material. Gen. Tech. Rep. FPL–GTR–113. Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Table of Contents Mark Schofield, How Strong Is Your Glue?, Fine Woodworking Jul/Aug 2007 http://www.woodbin.com/calcs/sagulator.htm 56 Notes and Reflections While Shaving Wood Chapter 10 – Accurate Measuring and Marking E measurement corresponds to the true or intended dimension. a. Dimensional or linear accuracy, what we usually measure with a straightedge b. Angular accuracy, for example the squareness of a board in all three dimensions c. Trueness -- planar or geometric accuracy -- flatness, straightness, smoothness, etc. Inaccuracies in this category include cup, bow, spring and twist. (See Chapter 4.) verybody would agree that accuracy is essential for making fine furniture. The necessary degree of accuracy, however, depends on the context, such as what part you are measuring and how it fits into the whole piece. For example, it rarely would matter much if an end table designed to be 24" wide turned out to be 24c" wide. However, if the front apron of that table were c" longer than the rear apron, the table would not be acceptable. It would show a gap somewhere, and suffer from weak joints. If one leg were 1/32" wider than the others, nobody would notice, but if it were 1/32" longer than the others, the table would rock, and I don’t mean in a good way. Likewise, if every piece were cut to accurate length but the ends of the aprons were cut at 90.2E instead of 90E, it might be impossible to assemble the joints. If you forced it together, the legs would be crooked and the table would be weak How thin a gap can you see? A fine pencil line is about 1/64" wide, the smallest subdivision on many rulers, yet a butt joint between two boards in a table top that included a gap that wide would spoil the appearance of the table and weaken the glue joint. This has nothing to do with how wide the boards are, and everything to do with how well the surfaces match over their length. The difference between a tight tenon and a sloppy tenon is a few thousandths of an inch, in both width and thickness. We can’t see that difference, and can barely measure it. Theoretically we could adjust and tweak a power tool setup to get that much accuracy, but we seldom if ever do. These examples illustrate four essential concepts. 1. Accuracy is about how closely a Table of Contents 2. Precision, repeatability, agreement. Precision is about how close repeated measurements or operations are to each other. Tolerance is about how much imprecision you can accept and still get the result you want. If accuracy is high enough, then acceptable precision is assured. That’s why we need accurate rulers and why we make mated surfaces dead flat and smooth. Precision is possible, however, even when accuracy is a bit off. In the example of the table, if both the front and back aprons are c" longer than design they were not cut accurately. The table would be c" wider than we intended, but the joints would still be square and strong. In the example of the tenon, the dimensions of the mortise and tenon can be off by much more than a few thousandths and the joint can still be tight if they agree. This leads to seven important points. (Sorry about the do’s and don’ts. These are hard-won pearls for me that I want to pass on to you.) 1. 10-1 Do not not assume that all measurement tools are equally accurate. The 57 10-2 Measuring and Marking different measurement tools needed for different measuring operations (see below) should be checked for agreement with each other. 2. Always measure from the same (reference) edge. Three inches from one edge of a 6" board is not necessarily the same spot as 3" from the opposite edge. They are the same in logic, but not necessarily in wood. 3. Do not assume a factory edge is square or smooth, or that all boards are the same thickness. This sounds obvious on paper, but it is very easy to be fooled when you are working along. 4. Accuracy and precision can be destroyed in some cases by using the wrong marking tool, or using it incorrectly. The tolerance for some cuts is less than the width of a pencil mark. Make a mark that is within the tolerance you need or develop a habit of cutting consistently to one side of the mark, preferably the outside, the side away from the reference edge. If you need greater precision than 1/32" use a marking knife or scratch awl, not a soft pencil. 5. When you mark a dimension, make one crisp mark. Then measure it again, but do not mark it again. If the first mark was wrong, erase it completely and start over. Otherwise, when the time comes to make the cut, you will not remember which mark was the correct one. Cutting between the marks will not increase accuracy. Table of Contents 6. Sometimes, when we cannot achieve satisfactory accuracy, we can achieve satisfactory precision. The repeatability of an operation may therefore be more important than its accuracy. We are taught to measure the wood, not the machine. But we should crosscut cut pieces that are supposed to be the same length with stop blocks. We should rip all pieces that are supposed to be the same width with one setup of the rip fence. We do not measure each cut separately because that would almost always reduce precision. 7. It is easier to fit some joints than to try to cut them perfectly the first time. A tenon that is 1/64" too thick will not fit but can be fitted perfectly with a few strokes of sandpaper or a shoulder plane. A tenon that is 1/64" too thin will be sloppy unless you shim it. Figure 1. Measuring Tools 58 Notes and Reflections While Shaving Wood Tools DIMENSIONAL MEASURING TOOLS The three dimensional measuring tools that are always near my hand are: • 12' Stanley Powerlock II Tape Measure (of course) • 6' Veritas Steel Hook Rule. • 6" hardened stainless steel dial caliper graduated in 1/64" and 0.01" These three agree within .01" at 6". Close enough for my work. I own other tape measures that I do not use in my shop. I use this one exclusively because it agrees with the other tools. In addition, I frequently use a Stanley Zig Zag X-227 6' folding carpenters rule with a brass slider. This is off by less than 1/64" at 7". Its measurement is within 1/32" of the tape measure at 4'. The brass slider is great for measuring the inside of boxes and drawers longer than 7". The rule and slider fit inside grooves and dadoes for drawer bottoms, etc. I use shopmade adjustable story sticks of various lengths ranging from about 4" up to about 30". (Top center of Figure 1.) The longer ones have fancy brass thumbscrews and pins from Veritas. Story sticks do not yield a number, they indicate a distance that you want to repeat or copy, such as the diagonal of a box or cabinet that you are trying to square up. This is much easier and more accurate than measuring the diagonals with a tape measure. I use a 6" combination square to set the blade height on my table saw. 10-3 automotive feeler gauge and shop-made shims at the bottom of the figure. I added a hotel key card to the photo because these are almost always about 1/32" thick and they are great for fitting drawers, etc. Also drill bits make convenient spacers (setup blocks) or you can buy setup blocks. You may need many other highly specialized measuring tools, e.g., dial indicators for setting up or checking a table saw, drill press runout, etc. ANGULAR MEASURING TOOLS The engineers square, second from the right, is an antique Stanley that I inherited. It’s my go-to square, followed by the Starrett combination square. When I use a square, I hold it up between my eye and a light source. The horizontal leg rests on the work by gravity. If the object is square, I do not see any light between the object and the square. The sliding bevel gauge and protractor work together for odd angles. Not shown, but very convenient: shopmade three legged square (for squaring small boxes) and a collection of Veritas saddle square, 45E guide and dovetail guides. I need a stainless steel carpenters’ square less frequently than the others, but it definitely has its place when I am making larger cabinets. . To check the accuracy of a square, find a wide board with a jointed (dead smooth, flat, straight) edge. Use the square to scribe a line perpendicular to the edge. Flip the square over, set it against the first line, and scribe another perpendicular line. They should overlap or be perfectly parallel over their entire length. I measure gaps with spacers such as the Table of Contents 59 10-4 Measuring and Marking DETECTING PLANAR INACCURACIES Cupping is a bend across a board. A straightedge laid across a board’s width will show cupping. Bowing is a bend in the face of a board along its length. Spring or crook is a bend in the edge of a board along its length. A board laid face down on a flat workbench or saw table will show bowing, and will show spring if laid on its edge. Twist is when one corner of a board is in a different plane than another corner. Winding sticks, which are simply pieces of wood with straight parallel sides, will show how much twist is in a board. Lay the winding sticks on the board and sight along the board over the tops of the winding sticks. If the tops of the winding sticks are not parallel the board is twisted. For a procedure to remove cupping, bowing, etc. errors, see Chapter 4, Lumber, Preparing Stock. Marking Tools and easily removable. If I need a more durable mark, I use a soft pencil, e.g., #B. I normally use a mechanical pencil with fine lead for routine marking of measurements. If I need a more precise mark I use a marking knife. I prefer the Veritas marking knife that is flat on one side. Occasionally I mark a long offset with a pencil attached to a combination square. I also have a small compass (shown) that marks with a pencil and a beam compass I can set for any desired radius, that will mark either with pencil or scratch needle. For very fine work that requires precise measurement and cutting, such as marking tenons, dovetails, etc. I use a marking gauge. (Figure 2) This is, in effect, a marking knife with its own adjustable fence. I much prefer the type that cuts with a marking blade, often called a cutting gauge. I have marking gauges that were made with scratch awls. I file off one edge of the scratch awl to make a knife edge. I label parts during layout and after they have been cut with white chalk. It is readily visible Figure 2. Cutting Gauge Table of Contents 60 Notes and Reflections While Shaving Wood Chapter 11 – Sources of Woodworking Information A broad but not very deep orientation to subjects like wood as a material, tools, hardware, finishing, shop math, and adhesives. It does contain lots of useful and interesting reference material that would otherwise be difficult to dig out of the literature, for example, formulas useful in shop math and standard furniture dimensions. ccess to reliable information about woodworking should not be a limiting factor for a woodworker. For many years, locating the book I needed for a woodworking project just never seemed feasible. That has changed. Useful woodworking periodicals, books, blogs and podcasts are now abundant. Videos may soon replace the written word for woodworking information. The issues now are deciding what information you need and verifying its validity. *Flexner, Bob. Understanding Wood Finishing. How to Select and Apply the Right Finish. Pleasantville, NY The Readers Digest Assoc. 1999. The bible of finishing. *Frid, Tage. Tage Frid Teaches Woodworking Book 1: Joinery and Book 2: Shaping, Veneering, Finishing. Newtown, CT. Taunton Press 1993. This is an encyclopedic reference Textbooks & Manuals Printed material remains my preferred medium for learning about woodworking. The exception would be turning and sharpening turning tools. These activities require a “dance” and I need both written and kinetic media to teach me. Every woodworking and finishing book I have bought has been useful in some way. Some were at the wrong level for me, however, or had an emphasis that did not meet my needs at the time. The books that I have found most useful until now are as follows. This list just keeps getting longer as time passes. The books (and video) that I feel are basic essentials are marked with an asterisk. book on woodworking methods. Frid’s Volume 3, is devoted to furniture design. It has lots of useful ideas, but I don’t consider it basic reading. If I were buying Volumes 1 and 2 and could get them in the three-volume edition, I would probably choose it, however. Groz, George. From Gunk to Glow or The Gentle Art of Refinishing Antiques and Other Furniture. New York, NY. Random House, 1973. This wonderful little book was the first book I ever read about woodworking, and got me started in finishing by convincing me that it was not a sacred mystery. I suppose that its out of print, but used copies are available. *Kirby, Ian The Accurate Table Saw. Fresno CA, Linden Publishing 1998. My basic textbook *Bird, Lonnie; Jewitt, J.; Lie-Nielsen, T; et al. Complete Illustrated Guide to Woodworking. Newtown, CT. Taunton Press2005. This is an on using a table saw. I really appreciate Kirby’s philosophy and specific approaches. encyclopedia of woodworking operations. It has similar coverage to Tage Frid’s book but is more modern. It is dedicated to him. Kirby, Ian The Accurate Router, Bethel CT The Cambium Press 1998. This is the book I used to learn how to use a router. It was much more useful than other, more highly touted books on router use. I think Rogowski’s book and video are a bit more useful. Churchill, Jennifer. The Woodworker’s Complete Shop Reference. Cincinnati, Ohio, 2003. Despite its grandiose title, this book offers a brief (140 pp.) survey of woodworking. It provides a Table of Contents 11-1 61 11-2 Sources of Information *Korn, Peter. The Woodworker’s Guide to Hand Tools. Newtown, CT. The Taunton Press 1998. Highly recommended. I very much wish I had read this years before I did. *Korn, Peter. Woodworking Basics: mastering the essentials of craftsmanship. Newtown, CT. The Taunton Press 2003. An excellent survey textbook cum lab manual with woodworking projects that illustrate basic methods. Nagyszalanczy, Sandor. Woodshop Dust Control. A Complete Guide to Setting Up Your Own System. Newtown, CT. Taunton Press, 2002 ½*Nagyszalanczy, Sandor. Setting Up Shop. Newtown, CT. Taunton Press, 2001. The practical guide to designing and building a shop. Many parts are unrealistic for a beginner. Rae, Andy. The Complete Illustrated Guide to Furniture and Cabinet Construction. Newtown, CT The Taunton Press, 2001. Raffan, Richard. Turning Wood With Richard Raffan. Newtown, CT. The Taunton Press, 2001 *Rogowski, G. Router Joinery, Newtown, CT. The Taunton Press 2003 (book and companion video) My favorite book on using a router. Rowley, Keith. Woodturning: A foundation course. New Edition, Lewes, East Sussex, England. Guild of Master Craftsmen Publications, Ltd. 1999 Wearing, Robert. The Essential Woodworker revised ed., Christopher Schwarz, ed. Fort Mitchell, KY Lost Art Press, 2003 Table of Contents Wilson, Aldren A. Hand Tools Their Ways and Workings. New York, W.W. Norton 1982. A beautifully illustrated and well written, complete guide to using hand tools. Reference Books Hoadley, R. Bruce. Understanding Wood. Newtown, CT. Taunton Press, 2000 Hylton, Bill. Rodale’s Illustrated Cabinetmaking. Emmaus, PA, Rodale Press 1998. This is an encyclopedia of practical furniture design elements. Forest Products Laboratory. Wood handbook—Wood as an engineering material. (FPL–GTR–113). Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Laboratory. 1999. Authoritative and technical. Slanted toward commercial wood production, preparation and use. Available online at http://www.fpl.fs.fed.us/documnts/fplgtr/fplgtr113/fpl gtr113.pdf Magazines I recommend magazines with a bit less confidence than recommending books, because the levels and topics presented in magazines are so varied, editors change, etc. Two magazines that I have found useful over the years are: Fine Woodworking, Taunton Press. This is the top end of woodworking magazines, in appearance, skill level and quality. The magazine often publishes comparative tool reviews that are informative (but which may not be completely objective). Popular Woodworking, F+W Publications. Similar to Fine Woodworking, but not as fancy. 62 Notes and Reflections While Shaving Wood Excellent practical articles for an intermediate-level woodworker. Somewhat slanted toward hand tools, but a good balance. When I was starting out, Woodsmith Magazine, August Home Publishing, was very helpful. This is a project-oriented magazine. Doing projects is a great way to learn, but you have to want the resulting product. The articles are profusely illustrated and provide step by step instructions. Woodsmith does not carry advertising. I have not actually made any of the projects but I have learned and used the methods they teach. The Internet The internet has been valuable in my education in woodworking. My first serious attempt to make “fine furniture” was based on plans that I got from the internet, simply by searching with Google for “bedside table.” I found many descriptions, including one article with drawings and a construction narrative that suited my taste and skill level. (See Chapter 52) You-Tube offers a wealth of howto videos on simple and complex operations. I have found lots of useful factual information on Wikipedia, e.g., about types of sandpaper and the size of abrasive grits, the hardness of materials, and how fast various drill bits should turn. Other favorite sites for brief articles, instructional videos, etc. are: Highland Woodworking “Library” http://www.highlandwoodworking.com/index.asp?Pa geAction=Custom&ID=3 Woodcentral, especially the Badger Pond Articles http://www.woodcentral.com/bparticles/bpindex2.sht ml seminars. 11-3 http://www.woodworkingonline.com/ The best tool list I found for a new woodworker (other than Peter Korn’s book) was in a short series of articles by David Savage. These may still be accessible on his site. Fine Woodworking.com is operated by Taunton Press, a book and magazine publisher. It includes a subscription service which is well worth the money because it provides search capability to past articles in the “archive.” The archive includes those published in the Fine Woodworking magazine and many more. I don’t recall what is available to non-subscribers but it would be worth a look. http://www.taunton.com/finewoodworking/index.aspx The Fine Woodworking site has an “ask the experts” section in which questions are answered by selected authorities in addition to the “forum” approach including regular members. This is part of the subscription service. Patrick’s Blood & Gore is a great reference for detailed descriptions and photos of old hand planes. http://www.supertool.com/StanleyBG/stan0a.html Another specialized site that I consulted frequently is Vintage Machinery http://www.VintageMachinery.org/ This is the “go-to”site for old (>20 years) out-of-print user manuals, parts lists, catalogs, and photographs of old tools. I know of no comparable source for this material, which is donated by users of the site. Many more 63 Woodworking Online, especially the video podcasts of Woodsmith woodworking Table of Contents 11-4 Sources of Information useful sites exist, which I usually come across by searching for a specific topic. WOODWORKING FORUMS Internet woodworking discussion forums range widely in quality. Their search engines are rudimentary at best. Most forums have sub-forums for power tools, hand tools, wood turning, etc. Most forums are sponsored by companies with a business interest in woodworking. Most are non-commercial and hosted by intelligent moderators. I have received useful answers to rather obscure and difficult questions, but some wrong (even dangerous) responses, “unencumbered by facts or logic”. Some posts are obviously intended to flog sponsor’s products. It really depends on the moderators. I have found the following general (wide-spectrum) forums to be useful: Woodnet http://www.woodnet.net/forums/ Sawmill Creek http://www.sawmillcreek.org/ Some contributors are just as you would want, knowledgeable and willing to answer questions. The collective expertise is quite impressive. Somebody will know an answer to almost any question, although you might have to wait a while until the expert shows up. I learned a lot from posting questions, but mainly from just “lurking” (reading the questions and answers). Other contributors seem to pretend a depth of knowledge that they probably do not possess. Sometimes, it’s like trying to discuss a problem in a large dysfunctional family, where everybody has an opinion to express regardless of what he/she actually knows, and Table of Contents a few cousins might have been drinking too much. (A Limerick for Woodnet in Reflections.) Nonetheless, despite a vociferous minority and a few actual experts with personality problems, there are a number of serious woodworkers who regularly post useful information. Fact, Opinion, “Common Sense” and Judgement Once, I supposed that choosing a correct woodworking method depended on expert opinion. If I had a woodworking question, I thought that I should ask somebody. I now consider expert opinion as only one kind of information, no more. If a recognized expert believes something, I should consider it, especially if the expert has taught woodworking. Even such expert opinions do not trump all other information, however, because my circumstances are always different than another’s. Some expert opinions depart from the facts as given by studies and are little more than expressions of personal preference, commercial connections, or defensiveness. Whenever possible I demand that an opinion square with known facts. I also need it to suit my skill level and available resources such as time and tools. So, now I weigh expert opinions along with facts and data as I decide what to do. Reality exists whether you believe in it or not. A piece of furniture will be attractive, durable and useful to the extent that it was constructed in accordance with realities such as the mechanical properties of wood and fasteners, the strength of various joint designs and the strength of adhesives. Objective facts are available in magazine articles and books like Hoadley’s Understanding Wood and the Wood Products 64 Notes and Reflections While Shaving Wood Handbook (see above). Some common sense is just information or opinion that a person learned before he was wise enough to demand evidence or to think about it critically. I suspect arguments based only on personal authority (e.g, who the speaker is or how long he has been doing woodworking) and arguments that appeal to common sense. They may be sincerely meant, but common sense is sometimes nothing more than information that a person learned before he was wise enough to demand evidence or to think about it critically. Many people (like me, in 1961) just Table of Contents 11-5 crashed ahead when they started. The way they managed to get along eventually became their common sense, but they might never have reflected much on why they made certain choices, or what they might have done differently. I have revised many pieces of my common sense as I have learned more about woodworking. Ultimately, a wise craftsman will collect as much information as he can within the time available, assess it in light of his own circumstances, and then carry out his own experiment. Every piece of advice about woodworking should come with the following additional advice: “Try this first on a piece of scrap wood.” 65 Notes and Reflections While Shaving Wood The following three chapters set the tone for all of the operations described in this book. Please think about what competence and safety mean to you. They are more than getting the job done. They are fundamental to everything else you will do in your shop. Chapter 12 – Reflections on Competence A lmost everyone wants to be competent, to feel competent, or at the very least, to appear to be competent. Competence is a highly charged issue. The worst insult you could give many professionals would not be that he or she is selfish, nasty, lazy or disorganized. It would be that he is incompetent. You probably have seen people all but admit to negligence, if necessary, in order to protect the impression that they really could have done the right thing if they had bothered, had time, etc. This subject may seem somewhat philosophical. Maybe it is, but it is fundamental for many adult learners’ development as craftsmen. That’s why I am putting it here, early in the book. When I decided to take up furniture making as a serious pursuit, I did not feel competent. I wondered if and how I could become a competent woodworker. I soon realized that I was not sure what competence is, and until I did, I would never achieve it. Once I figured out what I thought it meant, I realized I needed a strategy to improve it. Just reading, watching videos, or spending time in the shop would not succeed. So, my definition of competence in any pursuit is simply the ability to reach a desired objective reliably (repeatably or predictably). I might add, ... with available resources. Therefore, in this view, competence can be judged only with respect to the objective, e.g., doing a specific procedure or making a specific thing. True, we often use the idea globally, but trying to compare oneself to some vague, general notion of competence is pointless and usually disheartening. So, strategy #1 is to build up competence piece by piece. Identify a specific objective or a basic operation, become competent at that, and then move on to another. It may, at first, just be learning to saw to a line. That’s a good fundamental to have. Choose challenging projects, so you can learn something new, but which do not require competence too far afield from your learning objective (and patience). I may choose a subject that I want to learn, say mortise and tenon joints. Then I choose a project that requires M&T joints but is not too demanding of other skills I do not yet possess. I study, I practice my skills, I adjust my attitudes. Its fun and it has worked well for me. Second, according to the sociologist, Richard Stennett, . . . the real difference between able and clumsy hands lies in how each hand is stimulated and trained. . . . The innate abilities on which craftsmanship is based are not exceptional; they are shared . . . by the large majority of human beings Competence in any pursuit is simply the ability to reach a desired objective reliably Table of Contents 12-1 67 12-2 Competence in roughly equal measure.a So, in his view, most people can gain competence to perform many useful woodworking tasks. Strategy #2 is to give attention to developing skill instead of always finding a shortcut that gets the job done. A competent person has to know what he or she needs to do and how to do it; have the actual skills or abilities to do what needs to be done; and adopt the attitudes necessary to master it and to do it right.b Any of these three elements can – and often does – become limiting in the development of competence. We mislead ourselves if we think, for example, that competence is the same as knowledge. You cannot become a competent woodworker by reading about it or watching videos any more than you can become a good dancer by watching others dance. Similarly, increasing skill may increase competence for some woodworkers, but only up to the point that knowledge or attitude becomes limiting. So to speak, a graceful dancer still needs to know the steps and care about doing them well. Knowledge A wealth of information has developed over the centuries, and it offers choices among many ways to accomplish a task. One has to know, however, that an alternative exists in order to make a choice. One has to know the a Richard Stennett, The Craftsman, New Haven: Yale University Press 2008, pp. 276-7 b David Pye describes craftsmanship, the “workmanship of risk,” as depending on judgement, dexterity and care. I agree, although I prefer my terminology, as I will explain. David Pye, The nature and art of workmanship, (rev. ed.) 1995. London: The Herbert Press Table of Contents pros and cons of each alternative in order to make an intelligent choice, that is, choosing the alternative that is likely to produce the desired results within constraints of time, skill level, and available tools. Finding a method appropriate for my situation (especially skill level) is part of my design process. I sometimes have to research ways to make something until I find one that I can do or learn how to do. The more I know, the more likely I am to find a method that suits me. In this sense, knowledge can overcome some deficiencies in skill. That is called working smarter instead of harder.. Valid data and information are fundamental, e.g., the strength of wood, dimensional change with moisture content, the properties of adhesives or coatings, customary methods of work. Woodworking offers choices among many ways to accomplish a task. You have to know that an alternative exists in order to make a choice. You have to know the pros and cons of each in order to make an intelligent choice . A lot of woodworking knowledge can be objectively validated, e.g. the strength of adhesives, the load capacity of a board, the amount of dimensional change due to moisture. Sometimes, however, the best available source is expert opinion, which may actually be based on limited experience. It’s odd that an occupation that is judged on its results would contain so many “truths” that are true only because somebody says they are. It can sometimes be important to take a skeptical approach and to dig out whatever objective evidence may exist to support or rebut expert opinion. A woodworking expert who doggedly 68 Notes and Reflections While Shaving Wood insists on the “right” way to do something, but who cannot explain why it is the right way, may be competent nonetheless, but to whatever extent he is dogmatic, his competence is frozen. A competent woodworker also needs a theoretical foundation, just like a good musician. This may sound a bit pretentious, but valid theories help people to organize knowledge. They simplify both learning and application. For example, to design a joint, it helps me to understand stress and strain, how joints fail under stress, how wood moves with changes in moisture content, etc. Despite my respect for factual knowledge and theory, I don’t consider woodworking to be a “science” or an industrial pursuit. Wood is not consistent among species, individual boards, or even moments in time. The judgement of the craftsman is always necessary. Also, I agree with David Pye’s point that the esthetics of woodworkiug require spontaneity and risk. Skill Skill is the ability to carry out a specific action to obtain an intended result. A more skillful worker can produce a wider variety of results that are closer to his intention, more reliably and quickly. In addition to psychomotor actions like sawing to a line, skill includes the ability to concentrate on a task while remembering why you are doing it, and problem solving: the ability to ask questions and apply facts and judgement to produce solutions. Every woodworking tool, even a ruler, requires some level of skill. (Actually, measuring accurately is a fundamental skill that is often underappreciated by beginning craftsmen.) Every project presents problems that must be solved, some for the first time. Table of Contents 12-3 Skill is distinct from knowledge and attitude, but closely married to them. The prerequisite to developing skill is knowing how you want to perform an action, even if you can’t yet actually do it properly. After that comes practice, discipline, and a desire to improve. You may need hours to learn how to measure, which side of the line to cut, how to prepare and use a handplane correctly, how to hold a chisel, and so on. I am still learning and re-learning. I still have to sharpen myself as carefully as I sharpen my tools. I practice almost every new skill on scrap before I attempt it on the work. (At least that’s what I claim after the piece has become scrap.) I still have to sharpen myself as carefully as I sharpen my tools. Judgement is the integration of multiple pieces of information, so it requires accurate knowledge. It is a skill that needs to be developed like any other skill. Skill can overcome small deficiencies in knowledge or attitude. I may not know all the ways there are to cut a tenon, say, but if I am skillful with a hand saw I can get along quite well despite a lack of information about choice of methods. Likewise, if I am manually and mentally skillful, it is easier for me to get the result I want, and my need for discipline may be less. The skill requirements for hand tool use are mainly dexterity and concentration in performing the operation. The skills required for power tools are no less than for hand tools, but they relate more to choosing the tool and setting up the operation than to performing it. This connects back to knowledge. My personal example of this is setting up a router operation. 69 12-4 Competence Routers were unfamiliar to me, and they are still not my favorite cutting tool. I have to spend much more time setting up a cut than with other tools. I may have routed ten times as many test mortises in scrap than I have in actual work pieces. It’s easy to forget my ideal for the piece or for myself. After all, it will never be perfect. I’ll never become the woodworker that I’d like to be. The hell with it. Get a bigger hammer. Attitude Most craftsman need patience and hard discipline to achieve competent work. I have a persistent impulse just to “get on with it”, no matter that I may not be doing the right thing or may be doing it incorrectly. Most of the poor work I have produced myself or seen produced by others, has resulted because somebody acted on this impulse just to ram it in. My mental cartoon of this is a loose chair with a big old nail sticking in the joint. The caption reads, “If it doesn’t fit, get a bigger hammer.” Sometimes my impatience comes from a lapse of knowledge or skill, when I am unsure of what to do next or how to do it. I don’t like that feeling, so apparently I just want to do something. At other times, time pressure develops or the project gets tiresome. I want to finish or at least move on. Those are the moments in which the quality of my work depends on my attitude and discipline. Fitting tenons is a good example. Even though I cut mortises as precisely as I possibly can with a specialized jig and a router or a table saw, and even though I set it all up carefully, they often do not come out exactly the right size and exactly in the right place. (This mystery is a topic for a separate chapter.) So, like many people, I cut tenons to fit the mortise and cut them just a bit too large so I can tweak the position of the rail. Therefore, I spend time fitting tenons. Table of Contents One project (a double pedestal cabinet base) happened to include 36 mortise & tenon joints, including eight double tenons. The sanding and fitting became tedious. As each tenon got closer to the fit I wanted, I knew that I could drive it home with a mallet. I have a nice mallet. I was often tempted to believe that driving it home would compress the wood and make the joint fit perfectly. Of course, it was more likely that hammering the tenon home would split the mortise either right then or when I tried to take it apart again, or that the wood would return to its original size so that the joint would not fit correctly after I had applied glue to it. I might then sweat bullets trying to get the tenon inserted all the way. I might break the joint trying to do that. In the middle of a glue-up having to make a new part might be a near disaster. But, despite knowing all that, the temptation was strong to just hammer the damn thing home. Its easy at times like that to forget my ideal for the piece or for myself. After all, it will never be perfect. I’ll never become the woodworker that I’d like to be. The hell with it. Get a bigger hammer. But an attitude of patient determination, with a bit of discipline, sometimes allows me to produce work that is up to my level of knowledge and skill, however modest those may be. I’m not bragging here, on the contrary – I can’t afford to give any bit of competence away by being 70 Notes and Reflections While Shaving Wood careless or impatient. If I plan my work and think about what I am doing I usually produce better work than when I do not. If I am patient and disciplined it feels as if I am working above my level. I like that feeling. This is when the underappreciated value of procrastination becomes evident. It is better to gaze at the plans just one more time than to jump into a project I do not fully understand yet. It is better to take a break from fitting tenons than to pick up that hammer. Another essential attitude is self criticism. By this, I do not mean depreciating my work. I don’t let myself do that. I mean trying to see how my method of work and my output realistically could be better next time, even as I try to accept them it for what they are. This is difficult when I have tried hard and feel that I have accomplished something worthwhile. So I reflect about each project and make notes to increase the chance that I will do slightly better next time. Expanding my taste also seems to be important, to fire my imagination with possibilities. Someone who avoids a certain method of work or style of product because he doesn’t like it may literally not know what he is missing, and may believe he is missing nothing of value. I see woodworkers with broader tastes than mine, but many others whose tastes are much narrower. I am still expanding my repertoire. Finally, there is the role of performance in developing competence. Since I am aware of every piece’s shortcomings, I am reluctant to show it off. Making workshop furniture is a comfortable arena in which to practice. Making a piece for somebody else, however, for sale or as a gift, is like a musical recital. Table of Contents 12-5 Making a piece for somebody else creates a healthy and stimulating expectation that it will be critically appraised, however kindly or unkindly, according to others’ esthetics and values. Furthermore, the piece is likely to be on display in somebody’s home. They may even tell others that you made it. Especially if you sign your name to the piece, this can be a goad to learn just a bit more, practice more, try harder. Conclusion So, that’s my framework for becoming a competent woodworker. I think of competence as goal specific. I try to remember that the range of acceptable competence is very broad, something like the range of acceptable competence in music. First, few musicians can play more than a few kinds of instruments. Second, anyone who can get beyond a certain threshold can make pleasant music under the right circumstances. Some musicians can even make a very nice living without much competence at all. Many succeed based on authenticity and originality – think of the classic Delta blues musicians. A non-virtuoso has to choose his pieces carefully and rehearse a lot, but he (she) can produce pleasing work and enjoy producing it. A musical instrument is made to sound beautiful, if you will let it. Wood is naturally beautiful, too. Any community arts and crafts fair will demonstrate that many people can learn to make acceptable, or better, pieces if they practice enough and choose their pieces carefully. Competence in any art or craft is openended, unlimited. Most people need years (most of a lifetime) to become broadly virtuosic in any field. That’s why it is so important to think in terms of competence for a given task or a project. If I select a project 71 12-6 or accept a commission, all I need is the competence to produce that product, i.e., if it meets the criteria of attractiveness, utility and durability set by myself and my client. No matter how much I accomplish, however, I may still make beginners’ mistakes because I am actually a beginner in that particular subject or because I forgot to use competence I already possessed. I see work in magazines that surpasses mine by so much that it seems to have been made by magic. On the other hand, I see expensive commercial furniture in high-end stores that is inferior to my work. So it goes. Table of Contents Competence is a journey, not a destination. I wish to be competent enough to reach my objectives, brave enough to push my limits, modest enough not to over-reach, and wise enough to know the difference. Each project is what it is, usually the best I could do and, I hope, each a little better than the last. I said, 'It's certain there is no fine thing Since Adam's fall but needs much labouring.’ -- WB Yeats 72 Notes and Reflections While Shaving Wood Chapter 13 – Reflections on Workshop Safety “There are a lot of lucky people in the world but it sure is hard to predict who they will be.” B. Benjamin T strategy. A “case study” may help to set this up. his quotation puts great wisdom in a simple sentence. All custom woodworkers are playing roulette with their fingers and other body parts, and many have lost at least one spin of the wheel. Fortunately, the odds are more in your favor than in a casino. The risk of injury, even from unsafe behavior, is low. This is a good thing and a bad thing. The good part is obvious: most people have long periods where nothing bad happens. That bad part is that runs of good luck can reinforce unsafe behavior. Some people decide that certain safety practices are unnecessary, because they get away with breaking the rules (or see others get away with breaking the rules). They may decide that certain safety equipment is not really necessary, that they are too lucky, too careful or too smart to be injured. An example is someone like Nimrod in the fable below who says that he always knows where his hands are. They are whistling in the dark. Novices should not be mislead by people who espouse a “cowboy” workshop philosophy. In the last decade of my professional career I studied patient injuries from medications use. I learned something about error and accident. Accidents are predictable, but never expected -- that's why we call them accidents. Accidents happen when rare events coincide, when equipment and procedures fail at the same time. There is no reason why woodworking accidents should be an exception. Workshop safety is not simple. It requires specific procedures for specific operations, carried out within a general Table of Contents Nimrod – A Fable Nimrod’s first large project was some shop cabinets of plywood and MDF. His table saw had come with a blade guard assembly and a riving knife. He kept them on the saw for a while, but they did not seem to do much except get in his way. He had to remove them to cut dadoes and eventually he did not bother to replace them. When he replaced the dado stack with his crosscut blade, he did not bother to locate the insert that had come with the saw, but just put the dado throat plate back on. Nimrod was careful when he ripped plywood on his table saw. He ripped plywood with the rip fence in place. He held the work down with one hand and pushed it through with the other. He used push sticks whenever his hands might get too close to the blade. He was always careful to push the board all the way past the blade before releasing the pressure. He had heard that he should “trust his gut.” He felt safe. He cut a lot of plywood with no mishap, for that project and later ones. He decided that it was safe and even bragged occasionally that he did not have time to fool with the blade guard, etc. He said that he always paid close attention, and always knew where his hands were. He felt that he was more efficient without the guard. Then, one day, Nimrod needed to rip a glued-up oak panel. There was a bead of dried squeeze out along the glue line, but he did not notice it. Even if he had known it was there, he would not have thought anything of it. The area in front of the table saw was dimly lit. Because his customer was impatient, Nimrod was in a hurry and had not changed a light bulb over the saw table. Also, he had not swept the sawdust off the floor in a while. His mind was focused on getting the job done and the customer off his back. He started the cut, as always. Then, in an instant, the piece lifted up and started to twist. Nimrod tried to control it but his foot slipped on the floor. Faster than his eye could blink, the board struck his face. He was sitting on the floor and the panel was lying behind him. His left hand was bleeding, but he could not yet feel it or see it. A Strategy of Safety Despite what some woodworkers may say, nobody "always knows where his hands are.” There is always the unexpected moment when 13-1 73 13-2 Workshop Safety the workpiece rocks on a speck of debris, the cut closes on the blade, the cat runs between your legs, the lights flicker, whatever, and your attention is diverted. A strategy of accident prevention reduces the risk of accident by (1) excluding unsafe conditions and (2) adding layers of protection, so if one layer fails there is another to save you. You won't be injured except in the very unlikely event that all of them fail at the same time. That’s why I use the “roulette” metaphor. Injury remains possible, but the odds against it can be greatly improved. The operator has to decide how safe he wants to be. EXCLUDING HAZARDOUS CONDITIONS Hazardous conditions include obstructions on the floor like lumber, extension cords, sawdust on a glazed concrete garage floor; dim or glaring lighting; measuring tools left on the saw table; and animals, kids or gabby adults in or near the shop. Most of these can be excluded easily from the shop, except maybe the last one. Your general mental state can also be a hazard, e.g., working while angry, tired, rushed, or distracted; and working in too tight a space for you and the task at hand. Avoiding unsafe conditions is a form of hygiene. Like eating well, quitting smoking, etc., it can feel like a real pain, at first, and may feel strange. You may ask why you are bothering, when you don’t have the accident that your good hygiene has prevented. Frankly, that’s how I feel, sometimes. But, I have been injured in my shop. I have worked my way gradually into an expectation that I would have a clean, neat, well-lit shop. It took about a year before I lost most of my tolerance for messiness in my shop. Also, I Table of Contents can treat a two-minute pause for clean up and put-away as a mini-break in my routine, to clear my mind and remember what I am really in the shop to do. And then the next tool I need is usually where it’s supposed to be, so I don’t waste time searching for it. Individual hazards can interact. Suppose that you have a 1% chance of injury from tripping or stumbling while operating a power tool; a 1% chance of being injured because of poor lighting, and a 1% chance of injury while being distracted. If they didn’t interact, these risks could add up to something like a 3% chance of injury. This is bad enough. But if they do interact, the risks may be higher – in fact they may go instantly to almost 100% , where only luck can save you. In the Fable, Nimrod was operating in a simple “world” of ripping plywood, originally with a sharp blade, in a neat well-lit shop. He believed that this world was permanent. The fact that he had ripped plywood in that manner so many times had proved to him that it was a safe procedure. He probably thought that he could ignore recommended safety measures because he was especially careful in how he pushed the wood through the cut. He thought he was, in effect, a special case in technique, luck, or both. In reality, his technique was not special, only his experience was special. He had been operating in a small corner of the table saw risk universe. Perhaps be believed the wisdom of others who inhabited the same small corner. His experience mislead him.c c In his book, The Black Swan, Nassim Taleb describes the experience of a farm turkey. The turkey experiences months of kind treatment and regular feeding. His experience would teach him, if he were as smart as Nimrod, that his world is benign and safe. Even if he heard stories about Thanksgiving, he might not believe them, or think that he was immune, based on his experience. 74 Notes and Reflections While Shaving Wood Just two seemingly irrelevant details were enough to cast him into an entirely unfamiliar universe. The details were (a) the dried glue squeeze out, which had formed a bead on the bottom of the panel; and (b) the fact that the panel was solid oak, not plywood. Actually, either one might have been sufficient to pop him into the larger and more dangerous world. Also, his shop had degraded to “dim and messy.” The saw table had a gap in it caused by his neglect to replace the throat plate. He was not in his accustomed risk environment. When Nimrod started his rip cut, the panel he was ripping rocked a little on the seemingly insignificant little ball of dried glue. That was enough to allow the leading edge of the panel to catch in the wide throat plate opening. Nimrod’s forward pressure twisted the board a bit, and the hardwood closed slightly on the blade. The board bound the saw blade, which lifted it and threw it back into his face. Meanwhile, the horizontal resistance of the board disappeared, and his hands jumped toward the spinning blade. The board, on its way to Nimrod’s face, knocked the push stick out of his right hand, and luckily saved it. LAYERS OF PROTECTION The second part of a strategy of safety is to add layers of protection. This basic idea is the complement of hygiene – it is about safety equipment and appliances you add to the workplace to make it safer, such as guards, push sticks, jigs, protective eye wear, dust collectors, and such. This involves proper technique, which is discussed further below, but you need to have the right equipment in place in order to practice proper technique. So, this is about Table of Contents 13-3 more than keeping guards in place. For a simple example, some rip cuts on a table saw should be made with fingerboards to hold the work against the rip fence, and a long push shoe that holds the work down as it guides it through the cut. If you don’t have or don’t bother to find a fingerboard and push shoe, you can’t practice proper technique. Most people will just go ahead and make the cut with a push stick or their fingers, which is why this aspect of strategy deserves attention in itself. It is not hard to keep the proper instruments within reach of each tool, but you have to decide to do that. Tactics: Safe and Effective Technique Safe and effective woodworking technique should take place within a strategy so that the risk of injury is minimized. This sounds a bit fancy, I know, but it is very powerful, as the following example will show. Suppose you always try to rip using correct technique, like Nimrod did. Suppose you manage actually to do this 99.9% of the time. (That would actually be more consistency than most of us could actually achieve.) So you would have a 0.1% chance of a serious kickback, assuming everything else stays the same. That’s 1 kickback in 1000 cuts over the long run. You are likely to have a serious kickback eventually if you use your saw a lot, you just don't know when – it could be your next cut. Now suppose just having a a riving knife installed, or a splitter and anti-kickback pawls, also prevent 99.9% of serious kickbacks. If you use both good technique and have a splitter installed, your chance of kickback drops to 1 in a million. Maybe you will never have a kickback. If you add a fingerboard to hold the work down, then maybe the chance of a kickback would be 1 in 75 13-4 Workshop Safety a hundred million, and so forth. These odds improve geometrically because each layer of protection has to fail simultaneously or in sequence before the kickback can occur. The laws of probability give us the leverage we need to be safe. This advantage is literally too good to pass up. So workshop safety requires three elements: (a) exclude hazards that can pop you into an unfamiliar and dangerous world, (b) add layers of safeguards like a splitter, riving knife, anti-kickback pawl, blade guards, and (c) use proper technique. So, what about technique? Admittedly, there are many unsafe practices that should be avoided. I’ll mention quite a few later on. But the fact is, nobody ever goes into a wood shop for the purpose of being safe. We go into the shop to accomplish some objective. If we want to be safe, we sit in a chair and read a book. Therefore, it is a basic misunderstanding to separate safety from effectiveness. When we try to separate them, they can become antagonistic opposites, and we think that we have to balance safety against productivity. Most people who are injured chose to carry out an unsafe operation. The reason? They wanted to accomplish something and did not know how to do it safely. But they did it anyway, unsafely. Therefore, the best way to understand safety is to think of it as inseparable from competence. The three elements required to be a competent craftsman are knowledge skill and attitude. The same three elements are required for safe technique. SAFETY AND COMPETENCE ARE INSEPARABLE As I review my history as a woodworker, I can see many occasions where a lack of knowledge, skill or attitude was the Table of Contents explanation for an unsafe practice. Knowledge. My first serious injury was from my jointer, soon after I got it. What I did seems absurdly, obviously, stupid today, but I did not understand how wrong it was at the time. Else, I would not have done it. I had not learned the use and misuse of the jointer, or the likely consequences of misuse. Today, I feel that I had no business using it, but I was ignorant of my ignorance and felt entitled to use the tool anyway. My second major injury, much worse this time, was from the same jointer, years later. I was trying to plane a workpiece that was too small. I was using push sticks, but holding them incorrectly. Again, I did not know that this is a recognized misuse of the tool and a common cause of accidents. The other, indirect, cause was lack of skill with a hand plane, which I should have been using instead of a jointer. I probably still had no business using it. On other occasions, I have been rapped hard on the knuckles by work spinning out of control on the end of my drill press, luckily not sharp enough to cut off my fingers. It had not occurred to me that accepted safe practice is to place the work against the left side of the post, use an accessory fence, or clamp the work to the drill press table. I once thought dust was part of woodworking and dust collection was for tool freaks. But then I read a few articles in medical journals about pulmonary complications of wood dust. Now I use use dust collection equipment down to 1 micron and a dust mask when necessary. So, I did (and do) dangerous operations because I cannot think of an alternative. For example, I work alone. I carry 76 Notes and Reflections While Shaving Wood 4 x 8 sheets of ¾" plywood from the truck into my shop. These are very heavy. Each would make a great sail in a gust of wind. Each time I did this, I “hoped” that a gust of wind would not put me and my back in bed for a week. Hope is not a safety strategy. Eventually, I made a simple little cart (a skateboard for plywood sheets) and no longer carry them across my driveway. Skill Skill is the purposeful application of knowledge. Skill with a power tool, except perhaps for wood turning, is largely a matter of deciding which tool to use and setting it up correctly. Making the cut is often the shortest step in the sequence, if the others were done correctly. Skill includes choosing the correct tool, solving problems, concentration and discipline. Accidents from lack of attention, distraction, and poor setup belong here, although inattention could be a separate category. For example, over my career, I have - been hit in the chest by a flying chuck key left on a drill press chuck. - caught the power cord in my new portable belt sander. - drilled through a board into my left hand with a hand drill. - cut into a metal sawhorse with my circular saw. - had a router start up when I plugged it in. (I left the toggle switch ON when I removed it from the router table, which has its own switch). - had a router bit fly out of a table mounted router after I forgot to tighten the collet. - ruined an expensive workpiece by drilling into it with a loose router bit Table of Contents 13-5 - had foreign bodies in my eye from wire wheels when I forgot about my goggles. Luckily, I had no permanent injury from any of these. The common thread running through all of these is that I lacked the skill to carry out an intended operation, often because I was trying to think about too many things at once. I have since tried to develop solid techniques and routine setups. A well-set-up operation using jigs, clamps, etc, may remove distractions when the machine is turned on and the actual cut is being made. I try to stop before each operation to run a mental checklist. By listing all of these misadventures, I may appear accident prone or a real klutz. I am not accident prone, perhaps the opposite. I am a bit of a klutz sometimes. But I am writing this because I feel that almost anyone is subject to these kinds of mishaps, or worse, if he is not careful to develop skillful, safe, methods of work. Attitude. Two important attitudes are (a) a sense of risk and (b) acceptance of personal fallibility. Continuing to work when tired or annoyed is a prime example of denial of personal fallibility. I am not fooled by people who try to justify this as saving time, or “getting the job done”, especially not the ones who seem proud of that attitude, as if it made them more professional or manly. If they injure themselves, the downtime and lost income will greatly outweigh the time saved by unsafe practices. Sorry to belabor the point, but here it is again: unsafe practice and incompetence look like brothers. Familiarity and experience can create 77 13-6 Workshop Safety an illusion of safety. Probably the most dangerous activity for most people is hurtling down a crowded freeway at 60-80 miles per hour in a fragile sheet metal box called an atomobile. If you sneeze while driving at that speed, the car will travel about 90 feet while your eyes are closed. Yet we drive with one hand, eat, drink, dial a cell phone, etc., with no sense of danger. Because we are familiar with riding in a car, it does not feel as dangerous as it actually is. The point is, you have to be able to recognize danger with your intelligence as well as your feelings. So, I can use my familiarity with power tools to make myself more vigilant, not more complacent. The tool may not sound right, the work may not be going through as easily as usual. My familiarity can be valuable if I notice these things, stop, and investigate, and if I use it to build up knowledge, skill and good habits. Finally, there is the attitude of “blame the victim”. Many people who injured themselves will say, afterwards, that they were stupid. I said it about myself, a few times. I certainly felt that way. But, I’m not stupid. (At least, I wasn’t stupid back then.) I’m a smart person who did a stupid act. That literally means that I did not think about it. More often, however, my action was ignorant, meaning that I may have thought about it but lacked the necessary knowledge; or inept, meaning that I lacked the skill to do it correctly. It’s a small step to say that someone was injured because he was stupid; therefore stupid people are the ones who have workshop accidents. The logic seems to be, “Since I am smart, not stupid, I am safe.” Of course, this distancing from the victim is a form of denial, whistling past the graveyard. Research shows that anyone can Table of Contents make an uncorrectable error. If anybody can be at risk then I should not feel too safe, and I should be constantly vigilant and careful. WHAT ABOUT THE “DON’TS”? Safety is a part of competence. Because you are trying to accomplish some objective with your tools, no number of “don’t do this” warnings can make you into a safe woodworker. Only learning how to accomplish your objective safely, acquiring the skill to do it, and remaining careful and alert can do that. For example, to make a strip of molding, it is both easier and safer to rout the molded edge and then cut the molding off the board, than to cut the board and then rout the narrow piece. A bit of planning will effortlessly increase safety. That sounds a whole lot different than the safety rule not to rout a narrow board. The former is a safe way to accomplish an objective. The second is a “don’t,” that deemphasizes the reason for doing the operation in the first place. Of course I value the “safety rules” that come with new tools. I just don’t think they go nearly far enough. Data from the Consumer Product Safety Commission (CPSC) and the few published reports that I have been able to find show that most workshop accidents occur when rules were violated and/or safety equipment was removed. What I have tried to bring out here is not that safety rules are useless, but rather that they miss the point of why people use tools. Conclusion. Recognize that woodworking is inherently more dangerous than other possible pursuits. Since the risk can never be zero, the objective 78 Notes and Reflections While Shaving Wood should be to beat the odds in our game of workshop roulette. The foundation of safe practice is to decide that competence and safety are two sides of the same coin. Both comprise knowledge, skill and attitude. One side of competence, what we usually mean by the term, leads to quality. The other side leads to safety. They are inseparable aspects of the same issue. Yet safety and effectiveness are often presented as separate concepts. They should be taught (learned) together. As I have learned new techniques in woodworking, I have consciously chosen to learn how to accomplish my objective effectively and safely, instead of using the first idea that I can think of. Like most woodworkers, I have pride in my work. I want to be competent and to be recognized as such. If safety is part of competence, I have to be safe to consider myself competent. I now see unsafe behavior in the same way that I see an inability to square up a board – as incompetence. When the OSHA rules were first enacted, somebody drew a cartoon of an OSHA cowboy, wearing a helmet, steel-toed Table of Contents 13-7 boots and a face mask, sitting on a horse equipped with a roll bar. I often feel that way about wearing hearing protection, a dust mask and goggles. But norms can change. When I was a child, the only child restraints in our ‘48 Chevy were the back of the front seat or an arm flung out across my chest in a quick stop. Now, most of us automatically buckle up ourselves and our children. We would not go back to the simpler but more hazardous time. Our norms have changed because of education, because the safety devices have become familiar, and because the mass production of safely equipped cars eventually lead to user-friendly, non-intrusive equipment that enhanced our competence as drivers. We can develop safe habits, get used to them and not want to go back to the old ways. We can quiet the voice that says, “Hurry up.” and amplify the voice that says, “That’s not how you do that.” “Don’t ever put your hands there.” “Put the guard back on.” Perhaps many accidents would be prevented if more woodworkers took as much pride in safe practices as they do in cool tools. 79 Notes and Reflections While Shaving Wood Chapter 14 – Workshop Dust Control W workers from breathing dust: dust masks and respirators; point-source dust collectors; and air cleaners, including forced high-velocity ventilation. Each has the same basic issues, although they appear in different ways. The issues in dust control equipment are the size of the particles passed by the filter (bag or cannister); throughput (cubic feet per minute, CFM); static pressure, the amount that the impeller or fan can pull against resistance; and noise. ood dust control is definitely desirable, possibly even necessary, not only for my safety in my garage workshop but also for the safety of my family in the attached house. Woodworking increases the concentration of fine particles in the air – not just sawdust, but molds, fungi, and bacteria that lived in the wood. Some woods are toxic, some are irritants and allergens. Microorganisms that accompany wood dust can be pathogenic as well as allergenic. Treated wood and manufactured products like MDF and some plywoods may release toxins or irritants. Long term exposure to these contaminants is associated with poor health and increased incidence of nasal cancer. Fungal lung colonization or infection sometimes are difficult to cure and are occasionally fatal in people with compromised immune systems. These facts changed my formerly relaxed attitude about workshop dust control. I do not mean to make woodworking sound like mining arsenic in your underwear. Most of the toxicities and allergies are rare, but they do happen. The science on the dangers of wood dust is complicated by the variety of wood species, microbial species, working conditions, and individual differences in allergies; exposure to other pollutants, e.g., tobacco smoke; pre-existing pulmonary function, e.g., asthma, etc. Some exotic woods have long been recognized as very dangerous. Lists of toxic woods are available on the internet, but the scientific justification for specifically including each species is seldom documented. DUST MASK The advantage of a dust mask is that it directly affects the air you breathe. Effective masks are inexpensive and can easily be available whenever you need them. The effectiveness of a face mask is limited by the fit of the mask and how difficult it is to breathe through the mask. A good mask will have a flexible margin to fit well against the face, especially over the bridge of the nose, and be adjustable to fit different sizes. It should have a one-way valve to allow easy exhalation and avoid fogging eyewear. It should carry a NIOSH N90 or N95 rating. DUST COLLECTOR The best strategy of workshop dust control is to capture dust at the source, before it gets into ambient air. This can be done with a large vacuum cleaner (shop vac) and a dust collector, a sort of high-capacity central vacuum cleaner. Strategies Types of Dust Collectors Three strategies are used to protect wood- Dust collectors are connected by ductwork to Table of Contents 14-1 81 14-2 Workshop Dust Control one or more sources of dust, e.g. a saw or router table. Sometimes they are connected to a dust hood or shroud that confines the dust so that the collector can suck it away. A single stage system basically maintains air velocity at a more or less constant rate from the source to the filter. That means that the filter receives more larger particles and needs to be cleaned more often, A two-stage system uses a cyclone or baffle to slow down the air stream, which allows larger particles suspended in the air stream to fall into a collection barrel. Cyclone or baffle systems are more effective than single-stage systems because the filters in a two-stage system may retain their throughput longer, so that finer filters are more practical. On the other hand, a partially clogged filter removes finer particles, as long as it maintains sufficient airflow. A 10-inch table saw, 6" planer, band saw, etc., each require about 300-500 cubic feet per minute (CFM) of air flow in order to maintain the air velocity needed to suck dust away from the tool and into the filter. A 1 HP portable dust collector connected through about 6' of flexible 4" hose should be adequate if connected to one tool at a time, two at the most. (See the article by Burkin in References) The woven polyester bags that come with some smaller portable dust collectors (DC) are claimed to trap particles down to 30 microns in size. That sounds good, but it’s not. In fact, a 30 micron filter could increase the danger. It is very possible to make shop dust into an even worse health hazard if you do not remove fine particles. A shop vac or DC that blows air through a coarse (e.g., 30 micron) filter will just blow invisible fine dust around the shop, possibly keeping it suspended longer so that you can Table of Contents breathe more of it. I think shop-made air cleaners made with furnace filters would fall into the same category. They may trap visible dust but not the dust that matters to your health. The smaller particles are the most dangerous, because they float in the air longer, penetrate more deeply into the lungs, and have more reactive surface area per unit weight. Woven felt collection bags and pleated filter canisters are available that are rated to remove particles as small as 1 micron. These are definitely a wise aftermarket addition to a portable DC that shipped with a 30 micron bag. The finer pore size reduces air flow per square inch, but increased surface area may compensate somewhat for the reduced air flow. Also, in use a filter will clog somewhat. This reduces air flow but it also decreases the particle size that can pass through the bag or cannister. The industrial standard for maximum particle size is less than a micron. DC systems for home shops that can reliably remove particles below 1 micron would be very expensive. Shop vacs, however, are now commonly available with HEPA filters. These are definitely preferable, but more expensive because the HEPA filter requires a stronger motor and better design to maintain CFM and static pressure. AIR CLEANER An air cleaner is not connected to a source of dust. Air cleaners are usually hung near the ceiling and against a wall. They continually pump air through a set of filters and, if installed in the right location, set up a circular air flow around the shop. You can estimate the capacity you need by calculating the volume of the 82 workshop space and then dividing that by the CFM rating of the air cleaner. That will give you a rough idea of how long the air cleaner will take to filter the volume of air in the shop. This cannot be taken too literally, of course, because the air flow around the shop and through the air cleaner is not likely to be regular. My Choices 4 I feel that the following arrangements are a reasonable compromise. I do not claim this as any standard that others necessarily should emulate. : 1. I originally used a 600 cfm portable DC unit with a large felt bag rated at 1 micron. I replaced that with a 1100 CFM unit. I hook my DC to whatever tool I am using, one at a time, with quick-connects. 2. In my larger shop I had an air cleaner hanging near the ceiling. This was rated to remove 1 micron particles. It had a crude flow meter to show when the filter needs to be cleaned. In my smaller one-car garage shop I do without an air cleaner but I still recommend one as a part of a dust control system. 3 A shop vac connected to the router, chop saw, and hand tools. A shop vac Table of Contents hooked to a chop saw is very noisy and only partially effective. I built a plywood hood to enclose the chop saw, to limit the spread of dust. Since I don't think the vac removes fine dust, I recommend running the air cleaner with it as a backup. A shop vac with a HEPA air filter would be much better. I can hook up my shop vac to my random orbit sander (although it is inconvenient). I do really dusty jobs, like sanding, outdoors whenever possible and almost always wear a paper dust mask. I have to empty the felt dust collection bag and the shop vac every few (6?) months. This is a chore. I cannot empty them without significant amounts of dust billowing up around me. I do it outdoors, when a mild breeze is blowing. I wear a dust mask, goggles and a hat. I empty them into double black plastic garbage bags. Further Reading Burkin, A. Dust Collection For the One-man Shop How to Get the Most Performance From 1-hp to 2-hp Dust Collectors. Fine Woodworking, March/April 2000 pp 82-87 Nagyszalanczy, S. Woodshop Dust Control 14-3 Workshop Dust Control 14-4 Possible Toxicity of Selected Wood and Wood Dust (See Notes) SPECIES C L A S S REACTION CATEGORY POTENCY DUST OR WOOD INCI- SPECIES CLASS DENCE REACTION CATE- POTEN -CY DUST OR WOO D INCI- R G D,W C nasal G D R GORY Alder I E&SK,R Myrtle Arbor vitae I R Oak, red Ash I R Olivewood S R,E&SK G D,W C Baldcypres s S R SM D R Padauk I R,E&SK ,N EX D,W C Balsam fir S E&SK SM D R Redwood S R, nasal cancer SM D R Beech S R G D R Rosewoods I,S R,E& SK EX D, W C Birch S R,N G D,W R Satinwood I R,E&S K EX D, W C Black locust I N G R Spruce S R SM D,W R Blackwood S E&SK G D,W C Teak S E&SK EX D C Boxwood S R SM D,W R Walnut, black S E&SK G leav es & bark U Cocobolo I R,E&SK G D,W C Wenge S R,E&SK G D C Ebony I , S R,E&SK G D,W C Willow S nasal cancer G D U Elm S E&SK SM D R W. red cedar S R, nasal cancer G D C Fir I E&SK SM R Yew, Europe I E&SK G D C Ipe I R,E&SK Zebrawood S E&SK G D R Mahogany, Amer. S R,E&SK SM D S DENC E R Table of Contents 84 Notes and Reflections While Shaving Wood 14-5 Possible Toxicity of Selected Wood and Wood Dust (See Notes) SPECIES C L A S S REACTION CATEGORY POTENCY DUST OR WOOD INCIDENCE SPECIES CLASS REACTION CATEGORY Mahogany, Africa S R G D R Maple S R G D,W R POTEN -CY DUST OR WOO D INCIDENC E Table Notes. The information in this table is from the Musical Instrument Makers’ Forum http://www.mimf.com/archives/toxic.htm. I cannot verify the information in the table, and offer it as examples of possible wood toxicities. I have omitted many less common species. * Class: Woods are either an irritant (I) which cause a reaction fairly rapidly after exposure and will cause a similar reaction repeatedly, or sensitizers (S) which may have a latency period of hours or months and may require repeated handling before an allergic reaction occurs. Sensitizers are the more severe, because once you're sensitized, you're sensitized for life and the reactions only get more dramatic. * Reaction Category: o Eye (E) and skin (SK) irritation (hives, itching, redness). o Respiratory [R] problems. o Nausea (N), headache, or general malaise, possibly even liver or kidney malfunction. o Cancer of nose and sinus. Statistics show that woodworkers have a 40 per cent greater chance of nasal cancer than the general population, but the majority of statistics on nasal cancer are based on data from 1920-1960 when the furniture industry became highly mechanized with little or no dust control methods. So don't freak. * Potency: Small(SM), great (G), extreme (EX). This is the potential of the wood or sawdust doing harm and would vary with the individual i.e., those who are allergy prone might think twice about working with wood classed as extremely potent. * Source: Dust (D) or wood (W), or both * Incidence: Unknown (U), rare (R), or common (C). 85 Table of Contents Notes and Reflections While Shaving Wood Chapter 15 – Safe and Accurate Ripping on a Table Saw made to stabilize it during the cut. If it rocks even a little bit during the cut it may jam and the saw may hurl it back at you in an instant. At best you will burn the work and maybe you will injure yourself or damage the saw. A board that is not four square can be clamped to a flat carrier board like a piece of ¾" plywood or MDF. General Tools has a product called an EZ Jointer Clamp that is useful for ripping boards that have waney or spring edges. For a twisted board, you need a ripping sled that will prevent the board from rocking. (Figure 2). To make a ripping sled, take a flat piece of ¾" plywood or MDF as the base of the sled, wide and long enough to hold the board while you are cutting it. One edge of the plywood must be straight, to run along the rip fence. Drill and countersink holes for 1 1/2 2" flat-head machine screws (about 1/4-5/16") in the base, a bit further back from the edge of the base as the board is wide. Take some 1 x 1x6" pieces of wood as hold-downs. Drill them about an inch from one end to accept the screws. Run the screws up from the bottom of the base, through the hold-downs, and use them to clamp the board to the base. Then run the edge of the plywood along the rip fence to R ipping is sawing with the grain of a board. A table saw is ideally suited for this task, especially if it is fitted with an outfeed table or a stand to support long boards as they come off the back of the saw table.a Safe and accurate ripping requires skill and has many prerequisites. This chapter will provide a brief summary of the major points of ripping on a table saw. I highly recommend that you read a book on the use of a table saw, such as Ian Kirby’s The Accurate Table Saw. You can rip light stock with a combination saw blade or even a crosscut blade. A combination blade may clog when ripping some woods. Heavy, hard stock may require a blade whose teeth are shaped and set specifically for ripping. Other than the design and sharpness of the blade, the first concern in ripping is the condition of the board. Ideally, a board should be four-square. (See Figure 1. EZ Jointer Chapter 4 on Clamp preparing rough cut stock.) A board that rocks on the table for any reason (e.g., debris, surface imperfections, twisting) is dangerous and should not be ripped on a table saw unless special arrangements are a Reasonable alterative methods are cutting with a larger band saw (14" or larger with a ½" blade or larger), a circular saw using a saw guide, and hand sawing. The mechanics of using a circular saw or hand saw are a bit complicated to explain, and are outside the scope of this chapter. Figure 2. Ripping Sled 15-1 Table of Contents 87 15-2 Safe and Accurate Ripping on a Table Saw cut the board. Of course, if you don’t mind holes in the board, you can just screw it to the base. Or, on the other hand, you can make an adjustable sled. Figure 2 shows a fancier version that has slots instead of holes for the screws and commercial hold downs with knobs. The second issue in ripping is the use of a splitter or riving knife and, for narrow pieces, a zero-clearance insert. (See Figure 2). It is essential that the board being ripped not be able to contact the rear teeth of the blade. This can happen if the board closes up because of internal stresses; if it rocks, twists or tilts during cutting; or if it jams between the blade and the rip fence because the fence is not parallel to the blade. A splitter or riving knife is installed behind the blade to hold the board away from the back of the saw blade. The difference is, a splitter is fixed in position, while a riving knife moves up and down and tilts with the saw blade. A splitter can be installed after the saw is built, but a riving knife is built into the saw itself. Many older saws shipped with a splitter and antikickback pawls, which act like ratchets to keep a board from moving backward (toward you). A zero-clearance insert is important to keep narrow cutoffs from catching on the throat plate, or even falling into the throat of the saw, which might make the board twist during cutting. Usually, you have to make or buy these as after-market accessories. The third issue is the location and adjustment of the rip fence. According to Kirby, it is not safe to rip a board with a rip fence that extends much past the blade, because of the danger that the board will bind between the fence and the blade, and kick back. He recommends a secondary fence, attached to the long fence, that extends only as far toward the rear of the saw as the rear of the blade. Also, a secondary fence that extends out over the front of the saw would aid in lining up the cut. My rip fence locks in position, as parallel as I initially set it to be. I have not felt the need to use a secondary fence for normal ripping. Some fences do not lock parallel to the blade, and a secondary fence is surely an important safety feature. Regarding the adjustment of the fence, opinions differ. Some authorities think that the fence should be as parallel to the blade as you can make it. That’s how my saw is adjusted. Others think it should toe out a few thousandths of an inch at the back of the table. If the saw is set up correctly, so that the blade is parallel to the miter gauge guides, then the easiest and most accurate way to check the adjustment of the rip fence is to put a long block of wood into the right hand miter gauge slot and check the fence against that. The fourth issue is the manner of propelling the board and holding it against the fence. First of all, I place a push shoe and push stick on the saw table to the left side. I remove everything made of metal from the table, except for aluminum push sticks. My ruler, tape measure, etc are usually laying there where I put them down. They might make dandy projectiles. At least they will ruin my saw blade if they touch it while it is turning. They can go to the right side of the fence. Figure 3. Plywood Push Shoe (½" plywood, about 6" high by 11"long. 1/4-1/2" notch at heel) 88 Table of Contents Notes and Reflections While Shaving Wood I stand to the left of the saw, usually touching the left front rail with my left hip. This positions me away from the path of most kickbacks and gives me a good view of the registration of the board against the fence. I may guide wider pieces with my left hand, but usually I use finger boards or a push stick. I do not let my fingers get within 6" of any powered cutter. I push the work into the cut with my hands, and as soon as the end of the board is within a foot of the front of the saw table I pick up the push shoe to finish the cut. I push the work all the way past the blade so that it cannot touch the rear of the blade. I turn off the saw and do not reach for anything on the table until the blade has stopped spinning and I can see the individual teeth on the blade. I never reach over the blade to pick up the cut piece. A push shoe pushes the workpiece forward and down at the same time (Figure 3). It is safer than a push stick for propelling the board through a rip cut. I do not use any old stick laying around as a push stick. I make them out of sound hardwood. They are at least 10" long and have a birdsmouth notch cut in the end. One of them is a commercially made thin aluminum push stick with a handle. They get chewed up eventually and I make new ones. I have made finger boards but I much prefer commercial ones. Finally, if you are ripping light stock or soft wood, a combination saw blade may work just fine. If, however, you want to cut thick and hard wood, e.g., 2" thick maple, your saw may well bog down unless you use a sharp rip blade. The cutting action of a rip blade allows it to plow though a cut that would clog and bog down a crosscut or combination blade. Take it easy. Don’t try to feed the work too fast. Some mild scorching may occur, but you can clean it up later. Serious scorching suggests that something is out of adjustment, Table of Contents 15-3 Overheating the blade will not help, however, and may ruin it. Cutting Narrow Strips Cutting narrow strips on a table saw can be difficult, even dangerous if it tempts you to put a push stick where it does not belong or, worse, your fingers too close to the blade. If you try to cut a thin strip between the blade and the fence, without being able to push it through the cut, it is likely to kick back. The small strip may not be dangerous, but it is likely to break. An accurate and safe way to rip narrow strips is to let them fall to the left side of the blade, that is, the side opposite the rip fence. Do not trap them between the rip fence and the blade. Mark the board to cut it on the left side and then use the rip fence to guide the cut. You definitely need a zero clearance insert to do this. For repeated cuts, you can use a simple jig that runs in the left hand miter slot. Commercial versions are available, but this jig is easy to make in the shop with a miter slot Figure 4. Jig for Cutting Thin Strips runner, e.g., borrowed from a featherboard. Attach a thin strip of wood to the miter slot runner, to act as a spacer. To make it 89 15-4 Safe and Accurate Ripping on a Table Saw adjustable, cut a slot instead of a hole for the attaching screw. Tighten it somewhat, slide it back so that it is closer to you than the blade, and tighten further. Then adjust the rip fence so that the supply piece is snug against the jig, and make the cut. For successive cuts of the same thickness, you leave the jig where it is and move the rip fence and supply piece. Tapers and Bevels I bought an aluminum taper jig when I was first starting out. It is moderately useful, but it does not accommodate some taper cuts and it is awkward to hold some work pieces down flat on the table while using it. I often prefer to make my own from a flat piece of plywood. I set the angle as rise over run (see the chapter on crosscuts) and screw a secondary fence to the plywood. I can easily add a hold-down clamp like the ones shown in Figure 2. I am reluctant to tilt the arbor on my table saw to cut bevels. I want it to be as close to 90E as I can get it, and I do not enjoy resetting it to 90E after I have tilted it. About the only time I tilt the arbor is to make compound miters, e.g., for segmented turning blanks. So I use sleds to cut the two angles that come up most often – about 20E for raised panel doors, etc and 45E for long bevels for boxes, etc. See Chapter 36 on cutting raised panels and Chapter 46 on cutting staves. A sled for cutting bevels is easy enough to make. These directions are for a 45E sled. You can cut the buttresses at other angles, also, e.g., 70E for a 20Epanel raising jig. (See Figure 5.) 1. Cut a piece about 6" wide and 20" long, preferably from ½” Baltic Birch plywood. This will be the base of the sled. 2. Cut 3-4 pieces of ½" or ¾" plywood to 45E exactly. The narrow leg of the Table of Contents 3. 4. 5. 6 7. 8. triangle should be 6" wide, to match the base. (See Chapter 16, Crosscutting on a Table Saw). These are the buttresses. Mark layout lines equally spaced along the length of the base and square to the edges. The outside buttresses should be about 2" in from the ends. Drill pilot holes along the lines, offset about d" so the screws will be centered in the buttresses.. Countersink both sides of these holes so that the buttresses will sit flat on the base. Glue and screw the buttresses to the base so that the square side of the buttress is flush with the edge of the base. Make sure that the buttresses are the same height and that they are square to the base. Cut a piece of ½" plywood to form the 45E jig face. It should be about 10" wide to span from the top of the buttresses to the table top and about 20" long to match the base. Glue and screw a ¾" square by 9" long piece of hardwood along the front of the jig face (the side toward you when you are using the sled). This will be a lip to help you push the workpiece past the saw blade. Place the face so that it overhangs about 2" from the buttress. This will give you space to clamp the workpiece to the 45E fence. Optionally, cut a piece of plywood to fit the right side of the jig. This will allow the jig to double as a 90E tall fence. If you do that, cut some largish holes in the plywood to allow you to put clamps through the right side. Or, you can install T track along the 45E face. Optionally, glue a sheet of coarse 90 Notes and Reflections While Shaving Wood sandpaper to the 45E face to help keep the workpiece from slipping down. Cut some test pieces. In most cases you must clamp the workpiece to the sled to get accurate, clean results. It is not unusual that the bevel will be off a little bit. To adjust the angle, you can glue strips of tape, wood or 15-5 sandpaper to either the 45E face or the bottom of the base. If the saw burns the cut, it may be that the wood is slipping down as the cut progresses or that the work is not sliding parallel to the blade. You might not be holding the sled firmly against the rip fence, or the base may not be parallel to the blade. Figure 5. Panel-Raising Jig (Example of a Bevel Jig) with Workpiece Clamped in Place 91 Table of Contents 92 Table of Contents Notes and Reflections While Shaving Wood Chapter 16 – Safe and Accurate Crosscutting on a Table Saw T crosscut pieces of the same length, however, a stop block attached to the miter gauge extension is useful. Never use the rip fence for this purpose because the cutoff workpiece may become trapped between the rip fence and the blade and be launched back at you in an instant. However, a reasonable second choice is to attach a spacer block on the rip fence in front of the blade. (The front of the saw is the side nearest you.) The workpiece must not be touching the spacer or the rip fence when it is being cross-cut. Then the cutoff piece will be clear of the spacer block and the rip fence when it is cut and cannot jam. Normally, I trim the factory end of all boards before I crosscut them, to eliminate checks, dye marks, etc. and to make sure that the end is square. All of the supply stock has to be trimmed before the stop block is installed, or a flip stop can be used. Then the boards are cut with the trimmed edge against the stop block. his chapter will provide a brief summary of the major points of crosscutting with a table saw. I highly recommend that you read a book on the use of a table saw, such as Ian Kirby’s The Accurate Table Saw. (See Chapter 11) The miter gauge is used for crosscutting.a It must be properly adjusted to provide accurate cuts, and you cannot necessarily trust the angle gauge as the saw arrived in your shop. You should calibrate it for accurate work. See Chapter 18 for ways to do that. To cut longer pieces accurately, I use a miter gauge extension. One came with my table saw. Many fancier (and more expensive) ones are commercial aftermarket accessories, but they need be no more complicated than a flat piece of ¾" plywood bolted to the miter gauge. Assuming that the miter gauge runs in the left slot, the right end of the miter gauge extension should reach at least to the plane of the saw blade. If the extension is wooden, and wide (tall) enough it can be longer. The first cut will leave a kerf in the extension that will be very useful in lining up future cuts. The extension also makes it easier to attach stop blocks for repeated cuts of the same length. I prefer to keep my miter gauge locked at 90E and to use jigs to crosscut angles. (See section below.) Repeated Crosscuts The rule is to measure and mark the workpiece, not the saw. If you need many Figure 1 Crosscut Sled a With very few exceptions the rip fence should be out of the way when you are crosscutting on a table saw. 16-1 Table of Contents 93 Crosscutting On a Table Saw Crosscut Sled A miter gauge, even with an extension, may be inadequate for cross-cutting especially wide, long or heavy boards. For example, one of my first projects required me to cut table legs from long piece of mahogany that was more than 8" wide and 2" thick. The first step was to cut off a piece 22 ½" long so that the legs would all be the same length after I ripped them. It was very unlikely that I could hold that plank in position as I slid it along using my miter gauge, even with an extension. There were other ways I could have cut it, but I chose to use a crosscut sled. A crosscut sled will greatly improve accuracy in crosscutting wide boards, say for casework. Because it is a sled, and is guided by two runners, it carries the workpiece past the saw blade without much worry that the workpiece will rotate during the cut. Depending on how large you make the sled, its base can be made from a variety of materials. A smaller sled base could be made of ¼" Masonite. A larger sled base could be ½" plywood or ¾" MDF. The main issue is that the base be flat. For my large sled, I used ½" Baltic Birch plywood. The size of the sled is usually a compromise between versatility and manageability. Obviously, you can cut wider pieces with a bigger sled, but it may be so heavy that it is hard to manage and so big it is hard to store when not in use. Because of this, it is reasonable to have more than one crosscut sled. I made my first crosscut sled approximately the same size as my saw table: 42" wide x 27" deep. I can cut boards up to 24" wide with it. It is a bit of a pain to lift and to store. This sled will be used for years and is well worth careful design. 1. The front of the sled, where you put your Table of Contents 16-2 hands, is a danger zone. If you push the sled too far, the blade may cut through the front fence. If your hands are there you may be badly injured. a. Include a large extra piece of wood at the front of the sled to act as a blade tunnel. I used four layers of Baltic Birch plywood glued together. Include a tail on the base to support it. b. Design a handle that will keep your hands away from the blade tunnel. c. Include a stop block to prevent the sled from traveling forward too much. A stop block is surprisingly difficult to include in some sled designs. Mine is just a bolt through the left side of the base that hits the rear rail of the saw to limit sled travel. d. I bought a polycarbonate blade guard but I never used it. 2. The fences are bridges to keep the sled from falling apart when the first cut is made. They have to be tall enough and strong enough to do that even when the blade is set to cut the thickest piece. A 10" table saw will cut through a maximum thickness of about 3½". That requires a pretty tall fence, so I made mine with two levels. 3. If the sled will double as a miter sled (see below) the rear fence must be short enough to allow you to cut miters at the ends of long pieces. 4. Runners must fit the miter guide slots of the top of the table saw with no lateral play, yet slide easily. They should be about 1/32" thinner that the depth of the slot, because you 94 Notes and Reflections While Shaving Wood will not want them to bottom out in use. They can be made of quartersawn hardwood (oak or maple), phenolic resin, or metal. I have used shop-made wooden runners as well as purchased phenolic runners. It is much easier to buy commercial ones. Construction is fairly straightforward and is described in many references. Here are a few tips that may help. 1. Prepare the runners by drilling countersunk holes in the bottoms of each runner near each end and near the center. 2. To align the runners and the base, a. apply a coat of wax to the table top and the miter gauge slots and buff it off. This is to protect the top from glue. b. Lay the runners in the slots and apply a smooth yet generous coating of PVA II glue along the top of each runner. c. Using the rip fence to square the base, lay the base on top of the runners. Mark the midline of each slot on the top of the base. Using short screws or tacks, fasten each end and the middle of each runner to the base from the top. (These are temporary, just to apply clamp pressure to the joint between the base and the runners.) Let the glue set for at least an hour. d. Slide the sled backward and screw the front of each runner to the base from the bottom through the holes you drilled in step 2. Slide the sled forward and repeat at the rear. Remove the sled from the saw and screw the runners in the middle. 3. The front fence must be accurately square to the blade. After the runners, the rear fence and the stop block are installed on the base, cautiously cut a saw kerf through the rear fence and part way across the sled. Then you can use Table of Contents 16-3 a framing square registered to the saw blade or saw kerf to square the front fence to the blade. The length of the legs of the square you use matters, as does the accuracy of the square. If you do not have a framing square, of if you are not sure it is accurate, you can use geometry. (Using geometry to lay out angles often allows me to mark guidelines over longer distances, which increases accuracy, and relieves me of having to rely on protractors, etc.) There are two ways to use geometry to orient the front fence at 90E to the blade. One is based on the 3-4-5 rule: a triangle whose sides are in the ratio of 3 to 4 to 5 must be a right triangle. So, just construct a 3-4-5 triangle with one vertex at point A. Figure 2. Layout of Right Angle to Kerf a. b. Use a straightedge to extend the left side of the saw kerf toward the front of the base. This is line AB in Figure 2. Put the front fence on the front of the sled where you want it to be. Mark point A on the sled where the rear of the fence would intersect a line extended from the saw kerf (line AB in Figure 2). 95 Crosscutting On a Table Saw c. d. Measure along the kerf 18" to establish point B. Draw an arc with radius 30" from point B. The point on the arc that is exactly 24" from A is point C . Draw line AC on the sled and fix the front fence along this line using only two screws. Don’t glue it in place yet 16-4 accuracy is necessary. (See Chapter 18 for how to check this.) If the fence isn’t square to the blade, mark a corrected point C and repeat the process. Possibly, you can enlarge one hole slightly, so that you can pivot the fence. When the fence is dead-on, loosen the screws somewhat and squirt some PVA II between the fence and the base. Screw the base to the fence using at least three long screws. Miter Sled The second way is shown in Figure 2. a. Use a straightedge to extend the left side of the saw kerf toward the front of the base. This is line AB in Figure 2. b. Put the front fence on the sled, where you want it to be. (Represented by line C .) Mark where the rear of the fence will intersect line AB. This is point A. c. Set a compass for a radius of about 8-10" (the larger the better). You want the arc to pass through point A, so put the pencil side on point A and use the needle side to choose an arbitrary point D. Mark point D so you can find it again. d. Draw an arc that intersects line AB and that would intersect line AC. e. Take a straightedge and draw a diameter line from B through D to where it intersects the second arc. That is point C . Draw line AC on the sled and fix the front fence along this line, using only two screws. Don’t glue it in place yet 4. Check the sled for accuracy. Since you will use it to cut wide boards, a high degree of Table of Contents MITERED CROSSCUT A miter is cut at an angle across the face or width of a board, as in a picture frame. Miter joints usually need to be extremely accurate, for example miter joints for boxes or picture frames. At least, each corner has to be square because otherwise there will be visible gaps. I often attach ½" - 1" wide edge banding to plywood drawer fronts. I prefer to miter the corners of the edge banding because it looks much more finished. It is demanding, however. As I stated earlier, I prefer to keep the miter gauge on my table saw locked at 90E. Even if I were willing to adjust it to 45E, it would take many test cuts to get it exact. Then, when I was done, I’d have to go through the same process to return it to 90E. (Unless you have tried this, you may not recognize that the precision needed is much greater than the resolution of the angle graduations on a miter gauge.) I cut very accurate 45E miters with a simple shop-made jig that screws to my crosscut sled. The strategy is to make a fence that is exactly 90E and then bisect it to cut 45Emiters. Measuring 90E is easier and more accurate than measuring 45E. Then it’s fairly easy to bisect it exactly, to get precise 45E miters. Even if you miss 45E by a few seconds of arc, the joint will still add up to 90E and the 96 Notes and Reflections While Shaving Wood box, picture frame, etc will close with no gaps. If you have a crosscut sled, this secondary jig takes just a few minutes to make. Take a good, flat piece of ¾" plywood. Assuming that your crosscut sled is built to cut perfectly square, use it to trim two adjacent edges of the plywood so that they are 90E to each other. Then cut the plywood into a perfectly square shape by running those adjacent edges against the rip fence. A 10" square is a good size. Turn the square piece of plywood so that one corner lines up with the saw kerf at the front of the crosscut sled and the opposite corner lines up (fits into) the saw kerf in the front fence of the sled (the one nearest you.) Make the cut. Figure 3. Miter Jig Resting on Crosscut Sled. One point of the plywood square in sticking into the saw kerf in the fence of the crosscut sled. It is concealed by the accessory “positioning board”. Screw the resulting plywood triangle to the crosscut sled as shown in Figure 3. The base of the triangle should fit against or be parallel to the front fence of the sled. The two leading edges of the square are now 45E to the direction of cut. They will be the “fences” for the work to rest against. That is basically the whole jig. If you like, you can add secondary fences to the edges of the plywood square. I did not bother. A ¾" high “fence” is high Table of Contents 16-5 enough for cutting edge banding. You can also attach another piece of plywood or ¼" Masonite to the plywood to help make sure that you can position it correctly when you reattach it later. If you do not have a crosscut sled, you can use the same idea with an accurately adjusted miter gauge. In that case, I would carefully cut the plywood square into two pieces along the diagonal and then attach the resulting triangular piece to an extension on the miter gauge. Here’s the trick: cut one leg of the mitered joint from the left side of the jig and the mating side from the right. Probably, the two leading edges of the plywood will be exactly 45E to the saw blade, anyway. But even if they are very slightly off, the error in one miter will cancel the error in the other so the miter joint will be square. As you work along, it is not really necessary to keep track of mating pieces. Just mark one side of each piece as the “top” as the joint is laid out. Always keep that side up. That way each side of each miter will be cut on the correct side of the jig. Make test cuts, press the mitered edges together, and check the joint for square. After the first miter is cut, there will be a saw kerf in the plywood. This is perfect for lining up the cut to a pencil mark. When I am making edge banding, frames, etc. I start at a corner, say the lower left corner. I cut miters on the adjacent legs without measuring, near the end of each piece. Press the mitered pieces against the corner, hold the joint firmly together, and mark the length of each leg. If you have a good eye, you can cut each leg to length using the pencil marks. Since I sometimes do not have a good eye, I often cut the miters about 1/32- 1/16" long. I then use a shooting board with a 45E fence and a low angle block plane to trim the legs to perfect length. (See the article by Ed 97 Crosscutting On a Table Saw Speas). Repeat for the opposite corner with two other pieces. One practical limitation of this jig is the length of the legs that can be cut. The front fence of a wide crosscut sled will limit the length you can cut. For example, if the apex of the jig is 12" from the front fence, the geometry of the sled might limit the length of a leg to about 17". For longer pieces, it would be worthwhile to make a dedicated miter sled with a narrower front fence. BEVELED CROSSCUT A bevel is cut across the thickness of a board. Beveled crosscuts are used to make mitered joints for table legs, long boxes, etc. They are cut on a table saw with a sled run along the rip fence. How to make the sled, cut the bevel, and assemble the box is included in the chapter on box making. They can also be cut with a sliding compound miter saw. Miscellaneous Miter Angles I cut a lot of pieces with odd angles in the process of making segmented turning blanks. It can take a great deal of trial and error to get the crosscut angle exactly right so that a segmented ring closes with no gaps. Therefore, if I expect to make many cuts at the same angle, as I do for segmented turning blanks, I make a specialized sled for each angle and leave the fence fixed in place. These topics are covered in other chapters. Sometimes, however, I may need to crosscut odd angles for a project that I do not expect to repeat. I do this by screwing a temporary fence to my big crosscut sled. If there is any chance I may want to cut that angle again, I mark a guideline on the fence. 16-6 You can use a protractor to set an odd angle, or you can use a sliding bevel gauge to copy one. I prefer to use trigonometry to lay out odd angles. This is actually much simpler than the procedure described above for 90E angles. All you do is lay out rise over run on the sled and attach a secondary fence corresponding to the line between them. The distance parallel to the saw kerf is rise, and the distance parallel to the front fence is run. For example, if I need a piece that tapers 1" over its 6" width I would mark a point on the saw kerf and, using a square, I would mark another point 6" over and 1" below the first one. For greater accuracy I might use multiples, e.g., 18" over and 3" down. Draw a guide line between the two points and attach a piece of scrap wood as a secondary fence. If I know the angle, say 10E and need the rise and run I use the fact that, for a right triangle, the ratio of rise over run is called the tangent of the angle. So, if the wood is 6" wide rise = tangent (α) x run rise = 0.176 x 6" = 1.05" Finally, if I know the rise over run and want to know the angle in degrees, I find the arc tangent of rise over run, for example, 1/6, which is 9.46E. Many calculators offer a tangent and arc tangent function. Reference Kirby, Ian The Accurate Table Saw. Fresno CA, Linden Publishing 1998. Speas, Ed. Shooting Board Aims for Accuracy. Fine Woodworking May/June 1994 98 Table of Contents Notes and Reflections While Shaving Wood Chapter 17 – Other Saws Useful in Furniture Making A workpiece until you can actually see individual teeth. The “footprint” of a band saw is usually much smaller than a table saw, making them a good choice for a smaller shop. In some workshops with limited space (and money) a bandsaw may be the first choice. The major limitation of a bandsaw as your “go-to” saw is the distance from the blade to the frame. (rip capacity). Cutting up plywood sheets is much easier on a table saw, but there are workarounds, as I will describe below. Some people would say the main limitation is the maximum distance from the table to the top blade guide (resaw capacity), but that’s an issue in choosing among bandsaws. There is no comparison in resaw capacity between a table saw and even a smaller bandsaw. Also, bandsaw blades are either much more expensive than table saw blades for equivalent quality (e.g., carbide teeth) or they need to be replaced more often. s mentioned earlier, many furniture makers consider a table saw to be the center of their workshop. A table saw is a very versatile tool, especially when you consider the availability of dado stacks and shaper cutters (the latter less common nowadays). Other kinds of saws also have an important place in a woodshop, either as specialized complements to a table saw or even as perfectly acceptable replacements. First among these is a bandsaw. Bandsaw Bandsaws excel at two things: cutting curves and resawing thick planks into thinner ones, including making veneer. But a larger bandsaw can be used for straight-line ripping and even crosscutting for joinery work, although you may need to spend more time keeping the bandsaw adjusted than you would a table saw, and you may need to finish the cut edge with a hand plane. You can think of a bandsaw in many ways as a powered hand saw. The blade moves continuously, whereas a hand saw moves back and forth. But a bandsaw cuts perpendicular to the surface of the work like a hand saw and will cut where you send the workpiece. So, in addition to ripping, crosscutting, cutting curves and resawing, a bandsaw will cut joints like dovetails and tenons. A bandsaw has no hazard of kickback. In my opinion it is much safer to use than a table saw, with one exception. Most bandsaws do not have a blade brake. Those heavy wheels that propel the blade usually continue to spin silently long after the motor has been shut off, and often the upper guide is concealing the blade. You have to remember this and not let your fingers near the blade to remove the CHOOSING A BANDSAW In addition to resaw capacity and rip capacity, issues in choosing a bandsaw are frame strength, blade widths, type of blade guide, ease of changing blades, and the power of the motor. Blade width and frame strength are related. Some band saw frames are not strong enough to properly tension a wider (3/4" or 1") blade. Necessary blade tension is somewhat debatable, as is the advantage of a blade wider than ½". Also, some guide systems will not accept very narrow blades, which would limit your ability to cut arcs with a very small diameter. Blade guides can either use ball bearings or ceramic blocks. Ball bearings may get dirty and stick. They eventually wear out. 17-1 Table of Contents 99 17-2 Other Saws Used in Furniture Making Ceramic guides also eventually wear out. Most people prefer ceramic guides. Ceramic guides are available for some saws as aftermarket accessories. The horsepower needed depends on the size of the saw, but see the views of Michael Fortune, discussed in the next section. Among “convenience” features, some bandsaws have a blade brake, and some have a “de-tensioner,” a lever that lets you conveniently take the tension off the blade for blade changes or if the saw will not be in use for a while. (You can always loosen the tensioning knob.) Some band saws have two (or more) speeds. That feature might be interesting if you want to cut metal, but that’s beyond our scope here. Bandsaws are available in a wide range of sizes, from small, inexpensive benchtop models to behemoth industrial models that cost many thousands of dollars and require exotic power supply. Most have two large wheels, one above and one below the work table. Some smaller saws have three wheels. All band saws have system for adjusting the plane of the top wheel, which moves the blade back and forth; blade guides; and an adjustable spring blade tensioner. These work together to prevent the blade from twisting while you are cutting. Bandsaws are designated by the diameter of their wheels, which determines the rip capacity. A 14" bandsaw makes the most sense in a small shop. A good one should have enough power and resaw capacity for most purposes. These saws have a rip capacities of about 13-1/2". and typically can resaw boards about 6 in. wide. Some models can be modified with a “riser block” to increase the resaw capacity. Fourteen inch saws with 12" resaw capacity are available. Motors are generally in the 1 hp to 1-½ hp range. These saws all will take up to a ½" band, while the best can accept up to a ¾" band. Benchtop saws have 9" to 10" wheels. Table of Contents They are the lightest and most economical, but they also are limited in what they can do. Motors are typically rated at about ½ hp or less, and the saws have small tables and limited resaw capacity. They are useful for model making, fine scroll work, and other work on thinner boards. I have found them very convenient for cutting tail boards for dovetail joints. For greater resaw and rip capacity, consider an 18" bandsaw. Motor size is typically in the 2-hp to 3-hp range. These typically require on 220 volt power. COMMENTS ON USING A BANDSAW Instructions on how to set up and use a bandsaw usually ship with a new saw and are plentiful elsewhere, but a number of points are controversial, or, as Michael Fortune says, misunderstood. Much of the controversy stems from differences in how the saw will be used. Fortune uses his band saw (as do I) as his main power saw and is willing to provide the necessary attention to detail. (Bandsaws may require a bit more attention than do table saws.) Others use a band saw as a secondary tool used mainly for cutting curves or for resawing. Also, in my (admittedly cynical) opinion, some manufacturers of saws or accessories prefer to accept or even exploit consumer ignorance instead of teaching how to use a saw correctly. This is especially true when it comes to the subject of drift. Drift is the tendency of a band saw blade to wander off a straight line parallel to the rip fence. The truth is that a properly set up and operated band saw with a sharp blade does not drift. But, if you want to sell bandsaws to customers who believe that drift is unavoidable, or to sell a fence whose main selling point is that it can be adjusted for drift, why would you teach consumers that they 100 Notes and Reflections While Shaving Wood need a new blade or a saw adjustment instead? According to Fortune, the three main issues in the performance of a bandsaw are blade type, blade alignment, and moderate tension. His views contradict some of the common wisdom for setting up bandsaws. He feels that a well-functioning bandsaw does not require high blade tension, special blades, high horsepower, or continual fence adjustments to correct for blade drift. He says that a 14" band saw with a half-inch blade is adequate for cutting joints, resawing wide laminates in exotic woods, and making curved patterns in thin or thick stock. First, he recommends a sharp (recently replaced) ½" wide, 3-tpi (teeth per inch), carbon steel skip-tooth blade for all of the above tasks. A finer blade (more tpi) tends to clog with sawdust and overheat. The blade weld must be smooth. Second, the gullets of the blade should be centered on the upper wheel. Since most wheels are crowned, if the blade tracks too far back, it will tend to twist and drift to the left. If it runs too far to the front, it will tend to drift to the right. Also, the tires should not be grooved. Fortune does not mention this in his article, but I believe that it is easier to track the blade correctly if the top and bottom wheels are in the same plane, i.e., lined up with each other front to back. At least, if the wheels are too far from co-planarity it may be impossible to track the blade at the center of the top wheel. Third, he recommends slightly undertensioning the blade. The example he gives is to set the tension for a ½" blade to the setting for a d" blade. Fourth, he puts more emphasis on adjusting the upper guides than the lower ones. He recommends .001 clearance for the top side guides and about 1/64"-1/32" clearance for the Table of Contents 17-3 top rear guide. In other words, a dollar bill is too thick to set side guide clearance. Fifth, the saw table should be squared to the blade (I think both side to side and fore/aft) and the fence should be squared both to the table and the miter slot on the table. Feed the stock slowly. Only finger pressure should be needed if the saw is set up correctly. If more pressure is needed, the blade is binding or clogging with sawdust (and therefore may be overheating). If the blade tends to wobble you are pushing too hard and/or the blade is dull. The saw should cut straight at this point. If it begins to drift after some hours of cutting straight, probably the blade is dulling unevenly or the set of the teeth on one side has worn more than the set on the other. It is best to replace the blade. The fence can be adjusted for drift as a temporary measure. To adjust the fence for drift, draw a straight line on a piece of scrap and cut freehand to the line. Cut about 10-12". Then set the fence to the angle of the stock. There is more to setting up and operating a band saw. This discussion has just expressed Fortune’s and my opinions concerning some possibly “misunderstood” points. As always, a good book on the subject will provide more detail and a wider set of operations. Miter Saws MANUAL MITER SAW The most basic miter saw is operated by hand. It is a frame saw or backsaw that slides in a guide called a miter box, which controls the horizontal angle of the saw and holds the blade vertical. A good quality manual miter saw is capable of very accurate crosscuts and miter cuts. While the most basic miter box has 101 17-4 Other Saws Used in Furniture Making very few horizontal angle pre-sets (e.g., grooves cut in the fences) a good quality miter box is almost infinitely variable from -45E to +45E or more. Many miter boxes also have grooves or detents in the table and fence that allow the workpiece to be angled vertically, so that compound angles can be cut. capacity may be to one side only (single bevel) or both sides (double bevel) A sliding compound miter saw (SCMS) is a compound miter saw with a mechanism that allows the blade and motor assembly to slide forward and back. This increases the crosscut capacity of the saw. POWER MITER SAW Usefulness A power miter saw, also known as a chop saw, is an electric saw used to make a crosscut in a workpiece at a selected angle. Most of these saws are smaller and more portable than a table saw. Common blade sizes range from eight to twelve inches. The workpiece is typically held against a fence, and the spinning blade is pivoted down into the work. The cutting action forces the workpieces against the fence. This essentially eliminates the hazard of kickback that is presented by a table saw, but the workpiece still can lift up and flip sideways if not well secured with your hand or a holddown.. A major feature of a miter saw is the miter index that allows the horizontal angle of the blade to be changed relative to the fence. Many miter saws enable precise one-degree incremental changes to the miter index, with "stops" that allow the miter index to be quickly set to common angles (such as 15°, 22.5°, 30°, and 45°). In a shop without a table saw, a powered miter saw is especially useful as a complement to a band saw. It complements the band saw’s limitations in the length of a crosscut board. In a shop with a table saw, a chop saw is a convenience tool. It allows the table saw to be dedicated to ripping, and is often much easier to set up for angled cuts. The more features a miter saw has, the more difficult the engineering becomes, or should I say, the less accurate the saw may become at a given price point. This is especially true for the sliding feature. You can work around inaccuracies in angle settings, but sloppy bearings and sliders affect the precision (consistency, repeatability) of performance. Some craftsmen say that a SCMS has no place in cutting parts for fine furniture. Others pay as much for a SCMS as they would for a table saw and get excellent accuracy and precision. Types There are several types of power miter saws. A standard saw allows horizontally angled (or mitered) cuts while the blade always remains vertical. In addition, a compound miter saw (CMS) allows the blade to be tilted vertically (beveled) This allows vertical, horizontal and compound angled cuts. The bevel cutting Table of Contents Circular Saw A circular saw definitely has a place in a small to medium-sized furniture shop. In a shop with a table saw, a circular saw is useful for initial cutting of stock that is too large for the table saw, such as 4' x 8' sheet goods and large boards as they come from the lumberyard. It is easier and safer to cut such stock with a circular saw on sawhorses or a low-to-theground cutting frame than on most table saws unless the table saw has large infeed, outfeed and side tables. 102 Notes and Reflections While Shaving Wood In a small shop, a circular saw can provide an alternative to a table saw if careful arrangements are made to use it. Originally, the term circular saw was used to distinguish a saw shaped like a disk from band saws and hand saws. This section refers to a hand-held, electric circular saw designed for cutting wood, also sometimes called a carpenter’s saw. Circular saws can be either left or right-handed. A right-handed saw puts the main handle in your right hand, and the auxiliary handle in your left. The blade is to the right of the motor. The wider side of the baseplate sits over your workpiece to the left of the cutline. The cut-off falls to the right. When a right-handed saw is held in the right hand, the operator has to lean over the saw, because the motor will be in the line of sight. However, if you hold a workpiece down on a sawhorse with your left hand while you cut off the right end of the board, your left hand is out of danger. The weight of the saw will be over the cutoff (waste) part that will remain supported. The saw base can be adjusted for depth of cut and can tilt up to 45° and sometimes 50° in relation to the blade. Adjusting the depth of cut helps minimize kickback and is essential if both sides of the cut will be supported. The blade may mount directly to the motor's drive shaft (sidewinder) or be driven indirectly through worm gears, which yield higher torque. USING A CIRCULAR SAW IN A FURNITURE SHOP The accessory that makes a circular saw into a substitute for a table saw is a simple edge guide, what I prefer to call a “zero clearance” edge guide. It can simply consist of two boards, attached one on top the other. The top board is narrower than the bottom board (say, 1" wide) and jointed as straight as can be. It is Table of Contents 17-5 offset about an inch from the left edge of the bottom board and offset from the right edge of the bottom board by the exact distance between the edge of the saw base and the inside of the blade. When the guide is first used, the left edge of the saw base rides along the right edge of the top board and cuts the right edge of the bottom board to the exact offset. After that, you simply line up the right edge of the bottom board with the layout line and cut. Some details matter. The specific design of the guide has to allow the work and guide to be clamped to a table that will support the work on both sides of the cut. So, the bottom board should extend beyond the top (guide board) an inch or so on the left side to allow room for clamping. The clamps cannot interfere with the travel of the saw. In particular, the clamps have to be low enough that the saw motor will clear them when the blade depth is set. Half-inch thick Baltic Birch plywood is a good choice for a shop-made guide. Commercial guides are available, usually made of aluminum. These are usually thinner, lighter and more stable than shop-made guides. Some are made to clamp to the edges of the workpiece. Some are made in sections that bolt together for long cuts. I usually have a “story stick” (actually a rectangle of 1/4" Masonite or plywood) available to check that the guide is positioned set exactly where I want it and parallel with the cut line. Note that this guide arrangement has to be set over the supply or work piece, not over the piece you are cutting off. The saw kerf is to the right of the guide. The waste has to be to the right of the blade or the part you cut out will be short by the width of the saw kerf. The cutoff has to be supported or it may tear the work at the end of the cut. I usually clamp the stock and the guide on top 103 17-6 Other Saws Used in Furniture Making of my workbench or an accessory table, with sacrificial pieces of scrap ¾" plywood underneath them to prevent the saw from cutting the bench top. I set the blade depth just c" - ¼" deeper than the workpiece, so the sacrificial plywood will be slightly scored, but I’m confident that the workpiece will be cut through, even if I happen to lift the saw slightly during the cut. I also use a shop-made “T” square for shorter crosscuts with the circular saw. Hand Saws The two main types of saws are crosscut saws and rip saws. The difference lies in the shape of the teeth. Rip saw teeth are like chisels. They cut in one direction. Crosscut saws have teeth like little knives. The angle across the tooth is called fleam. The more fleam, the sharper the tooth will be as a knife but the weaker it will be. They cut in both directions (although usually they cut more in one direction than the other.) Both crosscut and rip saws are available in a variety of points per inch (ppi or tpi). Coarser saws cut faster but leave a rougher surface. Finer saws are the opposite. Finer saws are better for harder materials and finer cuts like tenon cheeks and dovetails. The finest saws may have variable tpi so that the beginning of the cut is finer and the end of the cut is faster. In addition to ppi, saws are available in a variety of tooth sets, plate thicknesses, lengths, widths, shapes, and handle designs. Tooth set refers to how much the teeth are bent out to form the kerf. If the saw teeth have too little set, the plate will soon bind in the kerf and become difficult or impossible to use in most applications. An exception is a flush cutting or dowel saw. This saw is meant to cut off dowels or tenon ends flush with the carcass without leaving scratches. They Table of Contents usually have very thin, flexible plates and no set. They can make shallow cuts without binding. Too much set, on the other hand, leaves a ragged cut and may actually make the saw difficult to use because the teeth are scratching the side of the cut. The finest saws have a plate that tapers away from the teeth so that saw set can be minimized without binding the plate. A backsaw has a stiffener along its back. On a saw that cuts on the push stroke, this allows a thinner plate and narrower set, and therefore a finer kerf, without risk of the saw’s buckling under load. The thinner kerf removes less wood and cuts more efficiently. A carcass saw is usually a backsaw with either rip or crosscut teeth. A frame saw uses a frame of steel to place the saw blade under tension to accomplish the same objective of a thinner kerf and finer cut. Bucksaws, bow saws, hacksaws and coping saws are all example of frame saws. Pull saws cut on the pull stroke rather than the push stroke used by a European-style saw. The blade of a pull-saw is under tension when the teeth are cutting and therefore does not need to be as stiff as a European-style saw. Saws are often named for their traditional function in the days before power saws. Small differences in form and function affected productivity and mattered greatly to a woodworker for whom time was money. The names and differences in design remain today, but are somewhat blurred and overlapping. Likewise, fine details of ppi, fleam, set, rake angle, plate thickness, etc., while important, matter more to frequent hand saw users. In an era of power tools, small differences in productivity no longer matter to most woodworkers as much as they once did. The main types of hand saws in a furniture shop are dovetail (fine rip meant for 104 Notes and Reflections While Shaving Wood endgrain work); tenon (fine crosscut) saw; frame or backsaw used in a miter box (crosscut, various tpi available); veneer saw (basically a very small crosscut saw with razor-sharp teeth); flush cutting saw (crosscut with no set), compass (keyhole) saw; and coping or fret saw. I cannot justify the expense of handmade Western style says. My everyday and detail hand saws are Japanese pull saws. The “utility” saw has a .03" thick, 3" x 12" plate, 14 tpi. My everyday fine saw has a.01 thick, 1 !/4" x 6 1/2" plate and 26 tpi. I have a few finer Japanese saws. I use a good quality miter box equipped with a frame saw with interchangeable blades. My other saws are pretty much mid-quality saws. USING HAND SAWS Using hand saws is a psychomotor skill, which means that you have to know what you are supposed to do and then find a way to get your body to do it. An adequate description just of what you should know about using the various types of hand saws would take a chapter in itself, and I may not be the one to write that chapter. Please refer to the hand tool books by Peter Korn and Aldren Wilson. Christopher Schwarz also has written extensively about this skill. Very briefly, here are the basics for freehand sawing. For critical cuts, especially repeated cuts, use a good miter box. 1. Secure the work. Holding it with one hand while you cut with the other will produce rough and usually frustrating results. Chances are, you will saw with your stronger hand so the contest between saw hand and hold hand is unfair. The workpiece will move unless it is fixed in place. Also, if the holding force is too far from the cut the board may chatter, which is annoying and makes an accurate cut even more difficult. Table of Contents 17-7 2. Mark the layout line on three sides and check your progress on the back of the cut frequently. Checking the back of the board is especially important when using a coping saw. 3. You saw with your whole body, whether you mean to or not. The position of your eyes, shoulders, even your stance will affect the accuracy of the cut. You need to line up properly with the saw or it will cut at an angle. 4. Start as you mean to carry on. Every stroke will influence the one after it, but the first 2" are most important. Start the cut as accurately as you can. There is no shame in using a saw guide, block of wood, etc. to start the cut correctly. At least, use your left thumb against the saw to control it as you start the cut. There are some tricks to starting a cut. Sometimes you can notch the starting place with a chisel so that the saw will not wander before you have established a kerf. (Be very careful if you try this with rip cuts on narrower pieces because you may split the wood.) If the saw plate is shiny, look at the reflection of the board in the saw. If the reflection shows the board heading off at an angle, the saw is not straight. Begin cutting at an acute angle so that the saw is making a shallow cut along the layout line. This will help to guide the saw across the wide part of the cut. Start the cut on whichever side provides wood to back up the cut. That is, be careful you don’t chip off an unsupported piece when you are starting the cut. 5. Don’t push the saw into the wood. Slide the saw back and forth. For example, if the wood is held vertically in a vise, the main (only?) downward force should be the weight of the saw. There is an optimal relationship between the speed of cutting and downward force. If you push down too much relative to the rate at which the teeth are passing through 105 17-8 Other Saws Used in Furniture Making the cut, you will clog the teeth. Shorter strokes feel easier to control, but probably are not. Longer strokes will cut faster. 6. Relax your grip. The saw “wants” to cut in a straight line (except for a coping saw) and you should let it. Twisting and canting the saw to correct a cut seldom works. If you are off the line on the waste side of the cut, it might be best to start over from the opposite side. The cut will be crude, but maybe you can clean it up with a plane. 7. Cut to the correct (waste) side of the layout line, and leave a sixteenth or so extra so that you can plane the cut to smooth it, square it up, etc. 8. Support the cutoff so that it won’t break off at the end of the cut. Practice on scrap. JAPANESE-STYLE HAND SAWS Japanese-style saws cut on the pull stroke rather than the push stroke used by a European-style saw. This allows a Japanesestyle saw to be made with a thinner blade, which removes less wood and cuts more efficiently. Many Japanese saws have impulse (laser) hardened teeth. They are very sharp and stay sharp longer. They are impractical to sharpen, however. Instead, some saws (kaeba) have removable handles and replaceable blades. Top-of-the-line Japanese saw blades are made of very hard (and brittle) steel. They can be re-sharpened but can easily be broken by a person who uses too much force on the push stroke. (It is natural for most of us Westerners to cut on the push stroke, and so you may automatically push too hard until you have become accustomed to a pull-saw.) For this reason, it may be best not to buy a very expensive Japanese saw as your first purchase of this type. Types of Saws The Japanese word for saw is nokogiri. Apparently the type of saw is properly an adjective, so for example a Dozuki saw may be called a Dozuki-nokogiri. Azebiki – The Japanese version of a stair saw or pattern makers saw, used for cutting into the flat surface of a board. The blade has a convex curve which can begin the cut anywhere on the surface. Usually has a long handle with a double-edged (ryoba) blade. Dōzuki – a backsaw. (A saw with a stiffening strip attached to the back.) Kaeba – A saw with a replaceable/disposable blade. Kataba – a single sided saw without a back. The blade of a kataba saw is thicker than the blade of a Dozuki. The thicker blade does not need a rigid back and therefore this saw can be used for deep and long cuts. Good general purpose saw and usually inexpensive. Kugihiki – A Japanese flush-cut saw used for cutting off dowels, etc. Some (all?) kugihiki saws have the set on one side only. Mawashibiki – The Japanese version of a keyhole saw. Ryōba – Saw with two cutting edges, a cross-cutting (yokobiki) blade on one side and a ripping (tatebiki) blade on the other. The Japanese means "double blade". References Michael Fortune. Five Tips for Better Bandsawing. Fine Woodworking November / December 2004. 106 Table of Contents Notes and Reflections While Shaving Wood Chapter 18 – Three Ways to Check for Square Crosscuts 2. That is also why I use a bench hook/shooting board for close fitting A ccurate crosscuts, most commonly 90E crosscuts, are an essential part of fine woodworking. These methods are for checking the 90E setting on a table saw miter gauge, or chop saw. a One-Cut Method 1. Rip a piece of flat plywood, ideally 24" long and 12" wide. The “long” edges (perpendicular to the cut line) must be straight and parallel. The dimensions are not at all critical but should be large enough to allow good test resolution. How Accurate is Accurate Enough? The short answer is, accurate within less than 0.1E The effect of an angular error depends on the deviation from 90E and the length of the cut.b So, the wider the board, the more accurate the angle has to be to maintain a given tolerance. Note that a very significant angular error is possible well below the resolution of the dial on a miter gauge or cutoff saw. Angular Error (E) α Cut error at 12" (12 * tan(α) Max length for 1/32" cut error 1/(32 * tan (α)) 0.1 3/128" (a hair more than 1/64") 18" 0.25 3/32 7" 0.5 7/64 3 ¼" 2. If you are setting up the miter gauge or chop saw for the first time, set it as square as possible with a try square. 3. Mark both ends of the cut line so that you can identify the once-mated parts after the cut and keep track of which end was against the miter gauge or chop saw fence. 1. That is why I keep my miter gauge locked at 90Eand use a jig to cut angles. a I did not invent any of these. They are part of woodworking lore. 4. Cut the test board 5. Flip the offcut over, keeping the cut sides adjacent to each other. (Flip offcut “F” to be adjacent to stock “B”) b Cut error is the tangent of the angular error multiplied by the length of the cut. 18-1 Table of Contents 107 18-2 Three Ways to Check for Square Crosscuts 6. Place both pieces against a straight surface, e.g., the miter gauge or saw table. Push the pieces together along the cut line. The boards will touch at one end of the cut. 7. The width of any gap equals twice the error. A gap next to “B” on the stock (i.e., at the bottom) means that the miter gauge should be moved counter clockwise, and vice versa. For example, a c" gap is a 1/16" error. If the board is one foot wide (front to back), that corresponds to an error of approximately 0.3E, significant but less than you can read on most miter gauges. Adjust by trial and error. Two-Cut Method 1. Prepare a piece of plywood so that one edge is straight (mark this edge). The piece should be wide enough to register well against the miter gauge (or fence on a chop saw). On a table saw the piece should be as long as the distance from the front of the saw to the front of the blade. For checking a chop saw the piece should be short enough that the saw can cut it through, e.g., 6-7". 2. Place the prepared (straight) edge against the miter gauge or chop saw fence and cut about c-¼" off one side of the piece. (Make the offcut large enough so that the blade is stable and that both sides of the blade are in wood to prevent deflecting the blade. Unequal resistance on one side could possibly deflect the blade enough to invalidate the test.) Flip it over so that the straight edge is still against the miter gauge. Cut about c - ¼" off the opposite side in the same manner. 3. Now measure the width of the board at both the front and back. If the gauge is square, the widths of the board will be equal at both ends. If the gauge is not square, the widths will differ. The measured difference in widths is twice the actual error. For example, an actual error of 1/64" will produce a 1/32" Table of Contents difference. As described above, the angular error corresponding to this example would be minuscule. 1/32" error over, say, 12" corresponds to less than ¼E of arc. You would have to adjust the miter gauge by trial and error. I doubt that using the scale would be much help. Five Cut Method This method multiplies the error by five, so may be much more sensitive than the one-cut and two-cut methods. It is a good method for a chop saw because the width of cut is constrained. If you cut a 4" square test piece, you get, in effect, the equivalent of the single cut method on a 8" wide piece (more or less). 1. Select a piece of flat stock about 16" to 24" square (smaller for a chop saw, e.g., 4"). Again, the size does not really matter but you want a piece large enough to give good test resolution. Plywood or MDF is a good choice. You should have vernier calipers to measure small variations in the final cutoff. 2. If you are setting up the miter gauge or chop saw fence for the first time, set it as square as possible to the blade with a try square. 3. Mark the side that is touching the miter gauge or chop saw fence. Place the stock against the miter gauge and cut off about ¼" to ½" from the stock. Make the offcut large enough so that the blade is stable and has equal resistance on both sides. Unequal resistance on one side could possibly deflect the blade enough to invalidate the test. 4. Rotate the stock 90E by placing the fresh cut edge against the gauge/fence and make another cut. Repeat until you have rotated the stock back to the original position. While 108 Notes and Reflections While Shaving Wood cutting, make sure that there is nothing between the stock and the gauge/fence. Even a little splinter of wood or bit of sawdust may invalidate the test. 5. Make another cut in this position (this is the fifth cut). The offcut should be large enough to be strong (stable) and narrow enough to measure with your calipers. Save this offcut. 6. Mark one end F for forward (this edge was cut first by the blade or was against the saw fence), and B for back. 7. Measure the width of each end, F and B. The difference in width between the F and B ends corresponds to five times the angular error. 8. Use the difference B-F to determine how much to move the gauge or saw to bring the 18-3 cut more into square. If B > F, the miter gauge must be rotated counter-clockwise. (On a chop saw the saw blade must be rotated clockwise.) For example if one end is 1/8" thicker than the other, and the final cutoff piece is 12" long, then the fence is out of square by 1/5 of the arctangent of (c" divided by 12"). That happens to be about 0.12E . 9. Once the new reference is set, repeat the five sided cut test to see how close to square the system is with a new correction. There may always be some difference between “F” and “B” by this method. Eventually, further adjustments will become pointless. For example, an error of 1/16" over a 12" test board using the five-cut method corresponds to 0.06Eangular error. This would yield about 3/128" of cut error when crosscutting a two foot wide board. 109 Table of Contents 110 Table of Contents Notes and Reflections While Shaving Wood Chapter 19 – Cutting Tenons With a Saw M joint between the shoulders of the tenon and the stile contribute negligibly to the strength of the joint.) A properly made M&T joint is very strong. A M&T joint in ¾ in. thick by 2½ in. wide by 8 in. long in cherry can withstand over 1000 lb. of racking force. When over-stressed, a properly made M&T joint fails when the tenon fractures across the grain and/or the stile fractures along the grain. A poorly made M&T joint, however, may fail at much less stress if the tenon comes loose and separates from the mortise. Failure of the glue bond is more likely if the tenon cheeks are loose in the mortise and/or the shoulders are not tight against the edge of the stile. The difference between tight and loose for most M&T joints is a few thousandths of an inch (based on the amount of resistance to being assembled and disassembled dry). Few adhesives are strong across a gap of more than a few thousandths of an inch. Measuring and cutting to this tolerance is a challenge, to say the least, so I often cut tenon cheeks just a bit over-size and then trim them with a shoulder plane until they fit well. The question is, how much over-size must we cut them? Trimming tenons with a shoulder plane, rasp or sandpaper stick is a bit tedious and time consuming if you have to remove much wood. In addition to the fit of the tenon into the mortise, the angle of the joint must be correct (usually 90E) and the faces of the boards must have the correct offset. Think of a frame made with M&T joints at the corners. It will not be rectangular if the angles are wrong and it will not be flat if the angles are wrong or if the rails and stiles are not flush with each other. The same goes for the rectangle formed by table aprons and legs. I am belaboring these somewhat ortise and tenon (M&T) is a fundamental type of joinery used for many kinds of furniture. A mortise is a hole cut into one member of a joint, such as a leg or the stile of a frame. A tenon is a projection from another Figure 1. Mortise in a Stile and Tenon on a Rail member of a joint (e.g., a table apron or rail) that fits into the mortise. M&T joints come in a fascinating number of variations. (Hylton has an excellent catalog of joints used in woodworking.) This chapter covers some basic methods for cutting rectangular tenons into the ends of a rail. (Figure 1). Cutting mortises is included in Chapters 20 and 21. Cutting tenons with a router is covered in Chapter 21. A Good Joint A good M&T joint has to be strong and geometrically accurate. The strength of an M&T joint requires that the shoulders fit evenly and tightly against the stile and that they be held in position by a strong glue joint between the tenon cheeks and the walls of the mortise. (The glue joint at the end of the tenon against the bottom of the mortise and the glue 19-1 Table of Contents 111 19-2 Cutting Tenons obvious facts because I want to make two points. First, you can’t expect uniformly cut tenons to fit correctly unless the mortises are uniform in every respect, and they seldom are. I cut most mortises with a plunge router. I use a jig to make them all uniform. (Chapters 20 and 21) The widths of the mortises vary little, but the position of each mortise relative to the reference surface may vary somewhat, as may the length of the mortise. For this reason, I cut mortises first, and then lay out tenons just as if I were going to cut them by hand, especially the lines for the tenon cheeks. Second, cutting tenons horizontally on a table saw with a dado stack or vertically, by flipping the board around in a carriage jig ignores the principles that we should measure the work and not the tool, and that we should measure cuts from the reference edge or face. This flipping method is recommended in some books, magazine articles, and popular TV woodworking shows. Many woodworkers say they use it. So, obviously it can be made to work, but both of these methods will produce ill-fitting tenons if the stock is not all of uniform thickness and the mortises are not all offset exactly from the show face. Theoretically, you can keep tweaking the height of the dado stack or the position of the fence to accommodate variation in stock thickness, but this is never very precise, takes a lot of fiddling, and does not address the issue of alignment. (If you are not yet convinced, see the sidebar, Don’t Cut By Subtraction below.) I did not understand this problem until I appreciated how to cut tenons by hand, i.e., lay out and mark the tenon on the workpiece. (“Measure the work and not the tool.”) The workpiece is marked to correspond exactly to the mortise that it will mate with. Then, the tenon will compensate for any irregularities in the mortise. The tenon will fit well and the Table of Contents adjacent surfaces will align just the way you want them to. Dimensions The strength of a M&T joint depends mainly on long-grain to long-grain glue surface, i.e., the area of the cheeks. This is reinforced somewhat by the mechanical contact of the shoulders around the mouth of the mortise. Therefore, we wish to maximize the area of the tenon cheeks. The classical size of a M&T is half the thickness of the stock by twice the thickness of the stock. So, for a ¾" board, the tenon would be d" thick and about 1½" long. The tenon could be shorter if necessary to fit within the stile. Sometimes, a tenon will be mitered at the end to maximize its length in a narrow stile. Cut the Size You Want Trimming tenon cheeks is difficult to do accurately. The more you have to trim, the more likely that you will change the plane of the cheeks so that they don’t mate as well with the walls of the mortise. It is worth effort to cut the tenon to size, perhaps just 1/64" tight, so that it will need little if any trimming to fit snugly into the mortise and line up correctly with the rest of the joint. There are many ways Figure 2. Tenon Marked Out on Rail. (Lines are thick and dark for illustrative purposes) 112 Notes and Reflections While Shaving Wood to cut a tenon directly to the size you want. You can cut them by hand, of course, or use a table saw, a band saw or a router. I prefer to cut tenons on a table saw, but a router and jig can cut perfect tenons as well. On a table saw, you can use a tenon jig and a spacer, or use two blades with a spacer between them. After they have been set up for the first time, these methods will always cut tenons of exactly the same thickness. The band saw method is the most flexible. It is like hand cutting except that the band saw replaces the tenon saw. You can also make a spacer for the band saw method. For each method, you should lay out the lines for the tenon cheeks and shoulders as if you were going to cut them by hand. Carry the lines around to the side of the workpiece.(Figure 2.) The width of a pencil line or the width of a saw kerf is greater than the accuracy that you need to achieve for wellfitting tenons. Therefore, it is best to use a marking gauge or marking knife. (Chapter 10) If you have trouble seeing the scratch made by one of these tools, you can use a sharp pencil, if you remember which side of the line to cut on. For cutting tenon cheeks by hand, use a sharp rip saw with a high number of ppi (nine or so). I prefer a Japanese pull saw. I use a miter box to cut tenon shoulders. For more detail on cutting tenons with a hand saw, see Wearing. There are two basic kinds of tenon jig for a table saw: one runs along the rip fence of the table saw, and the workpiece is clamped to it. In effect, it is a high fence. You can find plans to make one in Kirby’s The Accurate Table Saw or on the web. The other kind is a sled – a heavy right-angle piece of iron with a clamp to hold the workpiece vertical and a bar that fits in the miter gauge slot. (See Figure 3) 19-3 Figure 3. Delta Universal Tenon Jig. (The rear workpiece support has been removed.) TENON JIG AND SPACER This method takes a while to set up the first time – getting the spacer just right. After that, it cuts tenon cheeks accurately and efficiently. It is flexible to the extent that the first cut can be adjusted to one of the layout lines. It always cuts tenons very close to the same thickness. 1. 2. 3. Prepare a block of wood that is exactly as thick as the tenon you want, plus the width of the kerf cut by the blade you will use. For example, if you want a d" tenon and will use a blade that cuts a 3/32" kerf, the block should be 15/32" thick. (Cut some test pieces until you get the thickness of the block exactly right.) Put the block between the jig and the workpiece. Set the jig so that the blade will cut on the layout line for the cheek nearest the jig. Make the cut. Remove the block and make the second cut for the other cheek. When you cut tenons this way, variation in the thickness of the workpieces does not matter. 113 Table of Contents 19-4 Cutting Tenons TENON JIG WITH TWO BLADES AND SPACER This method requires the most initial setup time, but once the spacer and zero-clearance insert have been prepared, it is the fastest and most accurate of the three. If you have a lot of tenons to cut, consider this method. 1. Make a spacer that will fit over the saw arbor, between two blades. In theory, it should be exactly the same thickness as the tenon you want to cut, Don’t Cut By Subtraction Suppose you needed a 6" long piece, and you had a 10" long piece. Would you rather measure the 6" workpiece or the 4" cutoff? Most people would choose to measure the length that they want, from the reference edge, rather than the length that they will cut off, measured from a non-reference edge. When you cut tenons by cutting one cheek, spinning the workpiece around, and cutting the opposite cheek, you are measuring the waste but not the part you want to use. The tenon is what remains after you subtract the waste from the original thickness of the board. Therefore, the thickness of the tenon and the quality of the fit depend on the thickness of the stock. If the boards vary a bit in thickness, the tenons will vary a bit, also. Even S4S lumber may not be close enough to produce consistent results by this method. If you have a thickness planer, it’s probably best to plane each board to equal thickness all in one setup. Even then, you should probably set up to cut the tenons a bit thick and trim them to fit with a shoulder plane. Re-setting the jig or dado accurately for each board may not be enough, no matter what the calipers say. You have to make test cuts on scrap, but if your stock is not uniform, scrap of the exact thickness may be in short supply. Finally, because you remove the same amount of waste from each side, the tenon will be cut in the center of the stock rather than being cut to the offset you need for that particular mortise. The surfaces of the stile and rail may not line up correctly. e.g., d", but some fine tuning will be necessary to accommodate the set of the saw teeth. You can make it out of hard wood, fiber washers, steel washers, etc. If you make it of wood, the grain should run parallel to the saw arbor so that it will be incompressible. Finegrained wood like maple would be OK. 2. Put the spacer on the saw between two identical saw blades. I use the two outside cutters from my dado stack. 3. Test the spacer by cutting grooves in a horizontal piece of scrap. Use the throat plate that you use for cutting dados. Do not make a test tenon on the end of a board because it may fall down into the throat of the saw. 4. After you get the spacer trimmed to the desired thickness, Figure 4. make a zero-clearance insert with two openings in it for the two saw blades. This is mandatory for safe operation, to keep the workpiece from tilting or falling into the throat of the saw. 5. Put the workpiece in a tenon jig, line up the layout line and cut the tenon. BANDSAW METHOD This is the motorized version of cutting tenons by hand. Or, you can make a spacer, similar to the method above. The steps are the same. You set the fence with the block in place, cut the inside cheek, remove the block, and cut the outside cheek. (The block will be a different thickness from the one used on a table saw, because the kerf cut by the bandsaw will be narrower.) Cutting Shoulders and Edge Cheeks Tenon shoulders are best cut on the table saw after the side cheeks are cut. Use the cheek 114 Table of Contents Notes and Reflections While Shaving Wood cuts to set up the fence and blade height. That way, you can get nice crisp uniform shoulders. Avoid “stepped” shoulders like the one shown in Figure 4. The edge cheeks are best cut on a band saw using a spacer block and a stop block. It may be necessary to adjust the fence repeatedly to cut to the layout lines. If you cut the mortises with a router or drill press, you will now have to decide whether to square the round ends of the mortises or round over the corners of the tenons. I prefer the latter approach. I use a fine wood rasp or sanding stick to quickly knock off the corners. A shoulder plane is great if you need to do much trimming. It must be very sharp and the iron must cut right up against the edge of the plane to avoid the little steps shown in Figure 4. If you get those steps, shave them off with a sharp chisel pressed in against the bottom of the cheek. You can also use some 100 grit sandpaper glued to a stiff flat stick or a wood rasp. The edges of the sanding stick or wood rasp should not be allowed to cut into the shoulders. Also, pay attention not to trim a taper into the tenon. If you remove too much wood from a 19-5 tenon cheek, glue a shaving back onto the cheek and trim the tenon again after the glue has set. A properly-fit tenon should slide into the mortise with some effort but should not need to be hammered in. References Duginske, Mark. The New Complete Guide to the Band Saw. East Petersburg, PA Fox Chapel Publishing 2007 Hylton, Bill. Rodale’s Illustrated Cabinetmaking. Emmaus, PA, Rodale Press 1998. This is an encyclopedia of furniture design elements. Kirby, Ian The Accurate Table Saw. Fresno CA, Linden Publishing 1998. Miller, Jeff. Perfect Mortise and Tenon Joints. Fine Woodworking Sept/Oct 2004 Wearing, Robert. The Essential Woodworker revised ed., Christopher Schwarz, ed. Fort Mitchell, KY Lost Art Press, 2003 115 Table of Contents 116 Table of Contents Notes and Reflections While Shaving Wood Chapter 20 – Safe and Accurate Routing R The Accurate Router. Much later, I discovered a book and DVD called Router Joinery, featuring Gary Rogowski. They helped me a lot, not only in their demonstration of techniques, but also because they demonstrated Rogowski’s casual no-frills style. I highly recommend them and the other references listed below. The following notes document some of my personal experiences and some points that I feel need to be highlighted. outers (the electric kind) are nearly indispensable for cutting stopped dadoes, stopped grooves and edge treatments (moldings). Hand tool alternatives include router planes, which excel for fine work, especially for cutting narrow grooves and dadoes and for cutting inlets for inlay. Plow planes will cut rabbets, grooves and dadoes, and molding planes are useful for edge treatments. But a router is less expensive, more versatile and more convenient (and noisier, dustier and more dangerous). I prefer to use a router to cut mortises and sliding dovetails. I often cut rabbets on the router table, because it allows me to leave my table saw in its normal set up. The router is useful for cutting half-blind dovetail pins, especially on a lipped drawer front. I have used a router to cut out a drawer opening in the center of a table apron, with acceptable results on my second try. (My first try included an accidental climb cut that shattered the 13/16" thick red oak workpiece.) Routers are about as simple as a power tool can be – a motor, a collet, and a bit, held in a base. Despite their simplicity, routers are very versatile. Some people prefer to use a router where I would use a jointer, table saw or hand tool. To take full advantage of their usefulness, I highly recommend instructional books or DVD’s. At first, routers seemed awkward and foreign to me. I had been an on-and-off woodworker for about 30 years before I used a router for the first time. The first two books I stumbled across did not help – perhaps they were not elementary enough for me. I started to make some progress when I read Kirby’s Choosing a Router A small fixed-base router or trim router, in the 1-hp to 1-1/2-hp range might be easiest to use until you are familiar with basic operations. A ¼ “ collet is OK for light work. As you gain experience, consider a larger fixed-based or plunge router in the 1¾- to 3½-hp range. You will want a half-inch collet for serious routing. When choosing a full-size router, consider these features: soft-start motor, smooth-operating and rock-solid depth-adjustment and plunge mechanism, comfortable handles, and an easily accessible on/off switch. If the router has a push-button arbor lock (for one-wrench bit changes) make sure that you can press it when the router is in the base. Some motors have to be removed from the base to access the button. Built-in dust collection is a useful feature but sometimes makes the router a bit more awkward to use. Large routers meant for table mounting and spinning large bits, especially, should have variable speed control 20-1 Table of Contents 117 Safe and Accurate Routing Hand-held and Table Mounted The short of it is, you probably will want both. Many routing tasks can be done safely and effectively with a hand-held router, especially mortising, cleaning out inlay recesses, cutting grooves and dadoes with a straight bit and straight edge treatments using a bearing-guided bit. A plunge base is needed for mortising, and most grooves and dadoes, especially if they do not extend to the edge of the workpiece. A hand-held router is a necessity for work that is too large to rest securely on a router table. A table-mounted router is easier to use for workpieces that are inconvenient to secure to a workbench, curved edge treatments, template work, especially large bits, and jointing edges (I use a handplane for this). Some specialized joint-cutting bits cannot be used with a hand-held router. These include chamfering bits, tongue-and groove, drawer lock and lock miter bits. Setup is much quicker for cutting grooves, e.g., for plywood panels. A table-routed router can be heavier (stronger) than would be comfortable to use hand-held. Fundamentals I consider the router to be the most dangerous power tool in my shop. This opinion is based on the nature of the tool: a sharp bit spinning at 12,000- 22,000 RPM, without cutter guards or anti-kickback accessories. That spinning bit can cut in any direction (and will, if you don’t guide it properly). A router can tip over while still running. A table mounted router can throw the workpiece across the room if you’re not careful. My opinion is also based, in part, on the memory of some near disasters. Table of Contents 20-2 1. Always unplug the router before touching the bit or collet. Sounds trite, I know, like safety boilerplate. You might be shocked someday to find out how easily a router can turn on accidentally, especially if it has a hand-grip “safety” switch. A. When you remove a router from a router table, remember to switch it off at the router itself. Otherwise, you will get a big surprise the next time you plug it in. B. I have accidentally turned on the switch on my router table by hooking it with my pocket ruler. Bad switch design, maybe. But you can’t predict when the unexpected may happen. 2. Wear eye protection especially, and ear protection. 3. Bottom the bit in the collet all the way and then pull it out about 1/16" before tightening the collet. Periodically clean out the collet with mineral spirits. 4. Control the router and the work piece. When hand routing, use an edge guide, jig or bearing-guided bit. Secure the workpiece with clamps. Always hand rout with the router in a vertical position, gripped in both hands, in a comfortable stance. 5. As a rule of thumb (there are exceptions, such as drawer joints) router tables are for cuts with the grain. Rout across the grain with a hand-held router and an edge guide or jig. 118 Notes and Reflections While Shaving Wood 6. Listen to the sound of the router. The motor should sound about the same cutting as when it is spinning with no load. If not, you may be forcing it. 7. Be especially careful with bits that have a ¼" shank. Avoid large cutters on a ¼" shank. To gauge how strong a bit may be, look at how much metal remains between the flutes. 8. With a straight bit, make a series of shallow (c- ¼" deep) passes rather than one cut if a lot of material has to be removed. 9. Use a straight bit to remove stock before cutting with a dovetail bit. 10. A “zinging” high-pitched buzz or rapid, intermittent chit-chit-chit sound when the bit is cutting may indicate that the bit is vibrating, probably because the bit is dull and/or you are trying to cut too much in one pass. It may vibrate loose from the collet or it may break. If the router “zings” when the bit is not cutting, the bit may be bent or damaged. 11. Cutter resistance increases as the square of the radius, e.g., a ½" diameter cutter has 4 times the resistance of a ¼" cutter. 12. Viewed from the top, the router bit is turning clockwise. Rout against the direction that the cutter is turning. When pushing the router away from you the guide should be on the left. If the router is between you and the guide, move the router from left to right. Avoid climb cutting, i.e., don’t Table of Contents 20-3 feed the work in the same direction as the cutting edge is turning. This is the same principle as not feeding work into a table saw from behind, i.e., in the same direction that the teeth are moving. The only exceptions would be extremely light (1/128") cuts to finish a surface, and then only if you are familiar with using the router. A. When cutting with a hand-held router, cut counterclockwise around the outside of the work and clockwise around the inside of the work. When using a router table, the router is upside down, so you should reverse this rule. B. On a router table, push the work from right to left, when the workpiece is between you and the cutter. Normally, you would not cut with the work trapped between the cutter and the fence. If that were necessary for some reason, the work should be fed from left to right. This is awkward and dangerous. Look for a better way to make the cut. 13. Avoid lowering work onto a spinning bit in a router table, even if you first drilled a pilot hole. That’s what plunge routers are for. If the spinning cutter touches the pilot hole on the wrong side, the work may go flying. I admit that I do not always obey this rule when starting a shallow and narrow (¼" x ¼") stopped groove. When I have pressed a board down on a spinning cutter I always have the right end of the board up against the push shoe (see Figure 1). 119 Safe and Accurate Routing 14. 15. 16. Rout small workpieces on the router table using jigs, push shoes, featherboards, etc. etc to hold them. (Figure 1) Beware of the hidden bit, e.g., in a groove being cut on the router table. If it becomes trapped in the work it may burn, break, or cause unexpected climb cuts. It may appear suddenly at the end of the cut, at the end you are pushing against. You need variable speed control in order to use oversized bits like panel raising bits. The larger the diameter of the cutter the faster the cutting edges travel. Full speed may be too fast for a large bit. This increases vibration, may burn the work and is not safe. Recommended Bit Speed by Bit Diameter Bit Diameter Maximum RPM < 1" 22,000 1 - 2" 18,000 2 - 2.5" 16,000 2.5 - 3.5" 12,000 At 22,000 RPM the cutting edge of a ¾" bit travels at 49 MPH, but the outside edge of a 3½" bit would be traveling over 220 MPH. 17. The following seven items are on a checklist taped to my router table. a. Collet tight? b. Fence adjusted correctly and tight? c. Vacuum connected and d. e. f. g. 20-4 plugged in? Height adjusted correctly and locked? Speed correct? Extraneous material removed from table, especially metal objects (rulers, collet wrenches, etc)? Work holder, push shoes, etc at hand? Need persuading? If you forget to tighten the collet, the bit may fly out of the router. At best it will land on the floor and be damaged. It may hit you. If you forget to tighten the fence, it may slide backward as you make the cut and might put your fingers into the cutter. If you forget to tighten the height adjustment in the router table, depending on how that is arranged, the router may rotate itself downward out of the router table. (As the saying goes, please don’t ask me how I happen to know these things so well. As I said, I have had some close calls.) Mistakes like this happen when you are distracted, focusing too much on one part of the operation, which lead you to forget another essential part like tightening the depth adjustment. The most difficult aspect of router use for me now is awareness of when a climb cut might happen. I understand the principle of feeding the work against the rotation of the cutter, but I don’t always recognize the occasions when an inadvertent climb cut might occur. For example, if I am trying to widen a groove slightly and use the “wrong” side of the cutter. Obviously, I am still leaning how to use it well and I still make mistakes. 120 Table of Contents Notes and Reflections While Shaving Wood 20-5 Cutting Mortises In Router Joinery (book and DVD) Gary Rogowski says that a plunge router is the ideal tool for mortising, and I accept that. So, I invested some time in learning how to do that well. I noticed, also, that experienced craftsman do not make a big deal out of mortising with a router. Tage Frid’s mortise jig is just a trough made out of wood. Apparently I, too, could learn this skill. Mortising was one of the first operations that I attempted with a router. At first, I spent hours – actually, days – trying one thing and another. For a long time, I usually managed to get some essential detail wrong, like how to position the work, clamp everything in place, position the stop blocks, etc., and still leave space for the router to actually cut. I found myself going in circles, and the day would end with the mortises still uncut. I am recounting this sad tale because I see now that part of my problem was my general lack of experience in the kinds of joinery I was attempting. Somehow, I expected the tool to substitute for knowledge and skill. I should have cut a few mortises by hand to really understand what I was doing before I tried to use a router to do it. My unfamiliarity with routers did make it worse, however. Part of my confusion was caused by not knowing what is important and what is not. Now that I have cut some mortises with hand tools, I see that the length and position of a mortise for an M&T joint does not have to be perfect, because you should always lay out and cut the tenon to fit it. It’s much easier if all the mortises are uniform because that will simplify cutting the tenons, but you can accommodate small variations in length or even position. Table of Contents Figure 1. Small Parts Push Shoe The mortise does have to be aligned exactly, e.g., parallel or square to the reference surface. Otherwise, the joint will be cocked. It’s a real pain to cut a twisted tenon (intentionally, that is). The sides have to be straight and parallel. You can drill out mortises on a drill press, using a fence and one end stop, and finish them with a chisel. This works fine. The great advantage of this method is that I can see where the drill is cutting. The disadvantage is that the cleanup takes more time and it is not as accurate as a router. Figure 2. Bracket Jig With Guide Plate I tried an upcut router bit in the drill press. That did not work very well. Also, lateral force is supposed to be quite hard on drill press bearings. For cutting mortises, I prefer to use a router fitted with a collar (guide bushing). The 121 Safe and Accurate Routing collar runs in a jig as shown in Figure 2. The jig rests on the bench top and is clamped in a vise. The work is clamped between the support block (the body of the jig) and a secondary support. (This is a modified bracket jig as described by Kirby. (See also Chapter 21.) Stop blocks can be attached to the guide plate. (Make them thin enough that they stop the base plate rather than the motor housing.) The “zero” stop block and the position of the router bit can be permanently marked on the plate, but of course it will be different for different bit diameters. A slightly more elaborate version of this approach is described by Bernie Maas in his article, Router Joinery (Figure 3). This jig has an adjustable fence, and an opening (“viewing window”) in the guide plate that helps to line up the cut. Figure 3. Maas Mortise Jig (from Fine Woodworking) I prefer to lay out each mortise on the work so that I can see it through the slot in the guide plate. Also, It is necessary to mark the centerline of the mortise where it can be seen through the viewing window. Both of these jigs give me the predictable precision I want. Each has the disadvantage that the size of the guide collar limits the size of the bit. For rougher mortising, e.g., nibbling out recesses for a half-lap joint, a bracket jig (Figure 2 without the guide plate) is sufficient. Fit the router with an edge guide instead of a guide collar. Table of Contents 20-6 THE JIG You can buy useful mortise jigs and save yourself the trouble of making your own. (See Chapter 21 for a review of a commercial mortise jig.) If you want to make one, see the original articles listed in References. Here is my summary of how to make a bracket jig or Maas jig. Prepare a piece of hardwood or MDF, say 6" W x 2" thick by 12" long, by exactly squaring two adjoining surfaces. Two 6 x 12" blocks of 3/4" MDF glued together to form a 1½" thick piece might be more stable than hardwood. This support block provides a strong and wide surface for the top to rest on (to make and keep a right angle). The 6" width will allow you to clamp the jig in a vise, with the workpiece resting on top of the vise or on the benchtop. You will need to use clamps and a secondary support to hold the workpiece in the jig. Or, you can make the jig narrower, e.g., 3", and clamp both the jig and the workpiece in the vise. The guide plate is a flat piece of plywood with a slot cut for a guide bushing. The slot is to guide a ½" or ¾" O.D. guide bushing installed in the base of the plunge router. The slot must be exactly parallel to the edge of the guide plate (actually, parallel to the support board or vise jaw). Cut it on the router table or with the router edge guide. Any good flat and stiff hardwood plywood should do. Baltic birch plywood might be best. The guide plate must be thicker than the length of the guide bushing but not so thick that the collet bottoms out on the collar before achieving the desired depth. (That is, the length of the cutter and the length of the guide collar determine the thickness of the guide plate.) For cutting ¼" mortises use a piece of good d” or ½" plywood with a slot exactly 122 Notes and Reflections While Shaving Wood ½" wide cut in it. A ½" OD guide collar will accept a ¼" or d" bit. For ½" mortises, the slot should be ¾" wide, and the wood should be at least 9/16" thick. The guide plate should be at least 5-6" wide to give good support to the router base. It can be wider, depending on how much offset may be necessary in the future. The slot can be any length desired, but I prefer that the guide plate be at least 3" longer than the slot at the right end. Screw the guide plate to the support block with wood screws or with hanger bolts in threaded inserts (Maas jig). For the basic jig, the plate must be screwed with the exact offset desired from the edge of the work. For example, to center a ¼" mortise horizontally on workpiece 1-5/8" wide, the offset would be 11/16" (half of 1-5/8 minus 1/8"). Draw a line along the bottom of the guide plate exactly 11/16" from the slot and use it to line up the plate on the support. The support block in the Maas jig is adjustable. Even if the mortises are centered, always place corresponding reference surfaces, e.g., the outside edge, against the side of the jig. SETUP The most difficult part of setup is marking the exact zero point inside the slot. It’s easy to calculate distances, but test cuts are essential. Clamp a piece of scrap in the jig, and make some test cuts. Establish start and stop points. 1. Let the left end of the guide groove be the zero point. Alternatively, choose a desired “zero” point near the left end of the slot and screw a stop block to the plywood, perpendicular to the edge of the jig (and the slot). 2. Place the edge of the router base Table of Contents 3. 20-7 against the 0" stop and cut a hole in the scrap. Mark the inside of the slot precisely with the 0" point for the left edge of the hole. Make a crisp clean line with a triangular file, coping saw blade, etc. (The distance from the edge of a ¼" bit to the edge of my Porter Cable (PC) router base is 3". The offset between a ¾" OD collar and a ½" bit is about 3/32" as nearly as I can measure.) Install a stop block for the other end of the mortise. The stop blocks for a f" long ¼" mortise are actually 6-3/8" apart. The stop block should initially be clamped but can later be tacked in place. It can easily be moved for different length mortises. Theoretically, the distance from the zero point (edge of the slot, when the router is all the way to the left) to the edge of the base, when the router has finished its cut to the right, is given by M - 1.5 Dc+.5 Db. Where M is the intended length of the mortise, Dc is the diameter of the cutter and Db is the diameter of the base, i.e., 5¾ “ (Note that the cutter travels M-Dc to produce a mortise of length M.) Cutting Edge Treatments and Precise Curves A bearing-guided router bit is pretty easy to use. Normally, I cut edges on large pieces with a hand-held router. The bearing rides along the part of the edge that will not be cut or along a template stuck to the work with double-sided tape. It is very important that the edge be as flat or as fair (smooth) as you will want the final cut to be. The cutter will magnify tiny chips or imperfections in the 123 Safe and Accurate Routing edge as the bearing rolls over them. When cutting straight edges on a router table, I use a “zero-clearance” (or nearly so) accessory fence to keep the work from being pushed into the bit. If possible, I set the fence so that the bearing is exactly in the plane of the fence. When cutting curved edges on the router table I use a guide pin to help me guide the work until it touches the bearing. I have made very close-fitting curved edging this way. I used a router to cut rule joints for the Dropleaf End Table. Whenever I cut across the grain on the router table I use an accessory support and a backer block. I use the miter gauge for horizontal cuts and an “L” shaped push shoe if the board is vertical. See the discussion of dovetails, below. Stopped Grooves, Stopped Dadoes and Rabbets I cut grooves and dadoes for panel doors, etc., on the table saw (when I had one). I use a router only for stopped grooves or dadoes. For finer work, narrower grooves, etc I use a router plane. Almost any sort of guide will work with an electric router for dadoes. For cutting grooves, I have two edge guides for my PC plunge router. One is finely adjustable with a thumb screw (PC 42690) and the other (PC 42160) is much simpler. It just slides along two long steel rods and is held in position by set screws. I threaded the rods on the model 160 guide so that they will screw into the model 690 if I want a guide on both sides of the workpiece. You can also get the effect of a guide on both sides of the work piece if you attach a wide secondary fence to the edge guide and run its bottom edge in a groove. I am starting to appreciate the router as a tool for cutting rabbets, e.g., on a lipped 20-8 door or drawer. I use a bearing-guided bit in a router table. Normally, I set the table depth exactly even with the bearing. Dovetails and Sliding Dovetails I sometimes cut the pin board for half-blind dovetails (DT) on my router table, using a stop block. I can space them however I like, and I can vary the size of the pins if I like, by making a second cut. The main point of cutting them with a router bit is to get pins of the same shape as the bit. (I cut the tails by hand or with a band saw). I use only DT bits with a ½" shank. I don’t trust ¼" shanks for a bit that has to take that much wood. I cut sliding DT mortises, e.g., in table legs, on the router table after hogging out some wood with a straight bit. I cut the tenons for sliding DT on the router table with the board held vertically in an “L”-shaped push shoe, stabilized with a clamp that runs along the top of the fence. See Chapter 59, Dining Table for details. I think that perhaps I should cut the tenons with a hand-held router in a jig. See Chapter 39 Boxes and Drawers. References Anonymous. Tage Frid’s Mortising Jig. Highland Hardware Library (on-line) _________ Routers Go Left ibid Kirby, Ian The Accurate Router, Bethel CT The Cambium Press 1998. Maas, Bernie. Router Joinery Fine Woodworking, March/April 1986. Purdy, S. In search of the right mortising technique. Fine Woodworking, . May/June 1998 Rogowski, G. Mortising with a router. Fine Woodworking Nov/Dec 1996 Rogowski, G. Router Joinery, Taunton Press 2003. 124 Table of Contents Notes and Reflections While Shaving Wood Chapter 21 – Cutting Mortises and Tenons With a Commercial Jig W The instructions in the manual are useful to get you started, and you should follow them. They are, however, seriously misleading in two respects. They suggest that you cut the tenons first and they hint that you cut from the jig setup rather than marks on the workpiece. This may work “on paper” but is not a reliable procedure if the position of the rail (the piece with the tenon) is critical. The jig assembly described in the manual has clamps only at the right side of the jig. With this setup, it may be tempting to flip the workpiece end for end to cut mortises at the opposite ends of the workpiece (post or stile). It is usually quicker and more accurate to cut all mortises with the reference edge (RE) against the jig. You will need a clamp at the left side of the jig. Either buy another clamp or switch the clamps. Use the “optional” center clamp to hold the work when cutting mortises near the end of a workpiece. hen I had easy access to a table saw, I preferred to cut tenons with a table saw tenon jig (See Chapter 19). The alternative to cutting tenons on a table saw is cutting them with a router and jig. You can make a bracket jig out of wood, as described in Chapter 20 to cut mortises, and you can figure out a way to use it to cut tenons as well, cut them by hand, or use a band saw. Or, you can buy a factory-made jig that will cut matching mortises and tenons. A number of products are available, for example the General Tools Jig, the Trend M/T jig and the Leigh FMT jig. These vary in price, versatility and convenience. I chose the Trend M/T jig. The Trend M&T jig is a steel bracket jig used to cut mortises and tenons with a plunge router. This jig is compact and somewhat versatile. It can cut mortises and tenons in five widths from ¼" - e" as it ships. Three more sizes are available. It has adjustable angles in two planes. The guide plates, guide bushings and setup bar are very useful. The jig can cut accurate and tight joints. It cuts tenons with rounded ends, which fit the rounded ends of the mortises. (It will also cut rectangular mortises and tenons.) One limitation is the maximum crosssection size of a workpiece: about 2" square for mortising. (See table below.) The maximum rail width for cutting a centered tenon in the end is 2", but you can remove the middle clamp to cut a wider piece. Check whether the workpieces will fit in the jig before getting to work. If you remove all clamps and such this would still be a useful bracket jig and you possibly could fit larger pieces. Best Way to Use this Jig Best practice is to measure the work rather than the machine. It is best to mark the work almost as if you were going to cut the mortises and tenons by hand and then use the jig to increase speed and accuracy. Mark out and cut the mortises first, and then mark out the tenons based on where the mortises actually were cut. That way, any small errors in the location of the mortise can be compensated for when you cut the tenons. (I always have a few mortises that were not cut exactly where I intended them.) It is nearly impossible to work accurately the other way – to mark out mortises from tenons. This approach differs from the almost industrial attitude conveyed by the manual. 21-1 Table of Contents 125 Trend Mortise and Tenon Jig Before you start, check the jig for the desired forward/back angle, usually 90E. Use a square. The zero degree mark may not be accurate. STOCK PREPARATION Stock preparation is fundamental to using this jig, as it is for almost any other woodworking operation. Establish and mark two adjoining perpendicular surfaces. The reference side of the workpiece is registered against the jig. If the face is not perpendicular to the reference side, the mortise will not be perpendicular to the face. (If you always set the reference edge against the jig, however, any small angle will be consistent for all mortises.) 4. 5. 21-2 on it is upside down (see figure). That will establish the center line relative to the jig body and the RE. Position the workpiece so that the midline of the mortise coincides with the edge of the setup bar. Set the left and right guide plates (“X”) individually to establish the length of the mortise. If necessary, use the “extra” clamp to hold the workpiece when cutting a CUT MORTISES FIRST 1. 2. 3. With this jig, all you need to mark out is the centerline and the ends of the mortises. The front/back position of the guide plate may remain unchanged for cutting an entire set of mortises and tenons. For initial setup, it’s a good idea to make the centerline of one mortise about 5" long so that you can align the jig accurately parallel to the RE. Install the 2c" bushing and the desired router bit.* Tighten the bushing lock ring well because it can loosen in use, which will ruin the cut. Set up the jig to cut mortises using the setup bar. A. Put a workpiece in the jig with the reference edge (RE) facing the jig. B. First set the vertical position of the workpiece and then the forward-back (“Y”) position of the metal guide plate. Hold the setup bar so that the printing 6. 7. 8. mortise near the end of the workpiece, e.g., at the top of a table leg. Use a vertical piece of wood to register and support the vertical position of the workpiece when cutting mortises. It’s a good idea to just dimple the ends of each mortise with the router bit to check position against the layout lines before cutting the whole mortise. Then plunge the bit all the way at each end of the mortise. Then cut out the remaining stock in ¼" deep stages. For subsequent pieces or cuts, keep the RE of the workpiece toward the jig. If 126 Table of Contents Notes and Reflections While Shaving Wood you can’t, you may need to re-set the guide plate (“Y”) coordinate. Position the workpiece (“X” coordinate) to line up with the setup bar. CUT TENONS TO FIT MORTISES 1. 2. 3. 4. 5. Mark out the centerline and length for tenons to correspond to the mortises. You can set both reference edges of the workpiece against the jig at one end, but when you flip it over end for end you may have re-adjust the position of the cut. You need layout lines to position the workpiece correctly. Install the proper bushing and a e" router bit.** Tighten the bushing lock ring well, because it can loosen in use, which will ruin the cut. Best practice: put a piece of scrap into the jig and set its height with the setup bar. Use this to set “0” on the router, then set router depth. Cut a test tenon and check the tenon length. Be careful to mount the workpiece into the jig against the vertical guide and to clamp it evenly. (It is too easy to mount it at a slight angle, so the the tenon is cut crooked.) If the RE of the rail and stile will be flush, you will not need to re-set the center line for the tenon. Otherwise, set it with the setup bar upside down so that it measures from the jig side. (As in step 3B above.) Then if you put the reference face of the rail toward the jig every tenon will be cut the same distance from the reference face. Unless you want a centered tenon, you may need to adjust the side-to-side 6. 21-3 position of the guide plates. Cut the tenon in ¼" stages - It takes a bit of practice to keep the guide bushing against the guide plate when cutting tenons. If you try to cut too much depth in each pass, the router may tip or wander into the center. A spiral upcut cutter cuts much cleaner tenons with less tearout. The manufacturer recommends a clockwise cut. This is a climb cut, which may be smoother but harder to control, If you get tearout at the tenon shoulder, deeply score the shoulder location before cutting the tenons. ___________________ *¼" mortise uses a ¼" spiral upcut cutter and 2c" guide bushing. All mortises require the 2c" bushing. **¼” tenon uses 1¼ ” guide bushing and e" cutter. A 5/16" tenon uses 1 3/16" guide bushing. See p. 9 of the manual for other bushing sizes. All tenons use the e" cutter nominally, except that different combinations of cutter and bushing are possible. A spiral upcut bit makes a much smoother cut and reduces tearout. A cutter with a e" shank, however, won’t fit in the normal ½" collet. So, consider a e" center cut end mill bit with a ½" shank or another combination of cutter and bushing. For example, for a d" thick tenon, use a ½" spiral upcut bit with a 1¼" bushing. The ¼" guide marks on the setup bar should be used to set the length of the tenon when using this setup, because the guide bar assumes that the 1¼" bushing is used to cut ¼" tenons. 127 Table of Contents Trend Mortise and Tenon Jig Trend Routing Technology Inc 7351 West Friendly Avenue, Suite A Greensboro, NC 27410 Toll Free 877 918‐7363  Tel 336 292‐5051 Fax 336 292‐5061 Email sales@trend‐usa.com or  technical@trend‐usa.com Material thickness ½" - 1 f" Tenon thickness 3/16" Tenon width max. using vertical guide a 3½" Mortise length max. using vertical guide b 3½" Tenon length max. Angle tilt compound -10Eto 45E 21-4 Spare parts from Ace Tool www.acetoolonline.com 2201 Wantaugh Ave Wantaugh, NY 11793 877-783-8899 -e Approx. 1" a. on end of 4" wide material. b. unlimited if vertical guide removed and wood repositioned. 128 Table of Contents Notes and Reflections While Shaving Wood Chapter 22: Lock Miter Joint on a Router Table T he lock miter joint is a form of splined miter joint with the splines running lengthwise along the joint. (Figure 1) It h a s s o m e advantages over a Figure 1. Lock Miter standard miter joint. Joint The interlocking joint does not slip under clamp pressure, and the increased gluing surface area makes a stronger joint. As with a plain miter joint, a lock miter joint hides end grain and is well suited for plywood construction if the plywood is high quality. For a discussion of other types of miter joints and other joints used for boxes, please see Chapter 39, Boxes and Drawers. A lock miter joint can be cut on a table saw but is quicker (perhaps) to cut with a specialized lock miter router bit. There are three prominent issues in cutting lock miter joints on a router table: setting up the cut, feeding the workpiece, and tearout. Figure 2. Set-up Jig for Horizontal Cut Many approaches to setting up the cut are possible. The fundamental objective in setting up the router table is aligning the center of the workpiece with the center of the router bit. This is more difficult than it sounds. The underside of the middle cutter should be precisely aligned with the center of the workpiece. By far the easiest way to do this is to use the Infinity Tools setup jig (Figure 2). Or, you could visually align the underside of the middle cutter to the midline of the workpiece. The setup instructions for the Veritas bit are very helpful. The most efficient way to check bit height is to cut two pieces of scrap the same thickness as the workpieces. Cut them as they lay on the router table (horizontally). Then flip one piece over and fit them together. (Figure 3) If the surfaces align, the bit is set to the correct height. If they do not, adjust the bit height by half of the protrusion. Before you do this, however, read the section on feeding the workpiece, below. Figure 3. Checking The next step Bit Height is to set the position of the fence. It is better to have a shaped opening in the fence, Setting Up The Cut Setting up the cut requires great precision in order for the edges of the joint to form a sharp corner with little or no overlap. (A small overlap, e.g. 1/32", can be planed or sanded off.) Because both pieces are cut with the same setup, a setup error for the horizontal piece may be repeated in the vertical piece, thus magnifying the error. It is usually necessary to make many test cuts. Also, the workpiece must be held in a specialized sled to prevent the bit from cutting too deeply at the end of the cut. 22-1 Table of Contents 129 Lock Miter Joint on a Router Table 22-2 remember that the setup depends on the midline of the workpieces. The stock for the next project may not be precisely the same thickness, especially if it is plywood. (This is another reason to use this jig with solid wood or Baltic birch). Feeding the work Figure 4. . Infinity Tools Setup Jig for Vertical Cut so that there is not too much space around the half of the cutter that is buried in the fence. The easiest way by far to set the fence position is to use the Infinity Tools jig. Align the centerline of the workpiece with the vertical line on the jig. (Figure 4) Otherwise, use trial and error. Start with the front of the fence aligned with the top corner of the top cutter. If the vertical test piece extends beyond the horizontal piece, the fence should be moved back approximately the same distance as it protrudes, exposing more of the cutter. If the horizontal piece protrudes beyond the vertical piece, the fence should be moved forward, exposing less of the cutter. After you have made a satisfactory setup, you can make your own setup jigs, but The pieces that will comprise the front and back of a drawer should be cut horizontally so that the pressure of opening and closing the drawer will be resisted by the splines. (See Fig 1.) By the same token, note that during assembly the sides will slide into the front and back and may need to be clamped only in that direction. Horizontal Cut Because the router bit has no bearing, and because firm pressure is necessary to keep the cutter from pushing the horizontal workpiece away from the fence, the workpiece may tend to “dive” into the cutter at the end of the cut, especially if the workpiece is narrow. This is dangerous to your fingers and may ruin the workpiece. It is good practice to have as narrow an opening as possible in the router table fence, but that may not be sufficient to prevent the problem with narrow boards. Two ways to avoid this problem are (1) to cut a much wider piece than needed and then trim it to the desired width and (2) to make a specialized carrier jig. The carrier jig should be tall enough so that the horizontal part (part 2 in Figure 4) rides along the fence above the level of the bit. The workpiece is pressed against the fence and clamped to the carrier jig between parts 1 and 3. Parts 1 and 3 can have rabbets cut in them to further secure the workpiece or to accommodate workpieces of different thicknesses. Sandpaper glued to the bottom of part 2 will help to keep the workpiece secure in the carrier. Figure 5. Carrier Jig for Horizontal Cut 130 Table of Contents Notes and Reflections While Shaving Wood Vertical Cut. The vertical cut may have a tendency to tip, either away from the fence, or sideways along the fence, or both. A tall fence and an “L” shaped push stick that rides along the top and back of the fence will alleviate this problem. The workpiece can be clamped to part 1 for extra security. Also, use a fingerboard at the bottom so that the cutter cannot push the workpiece away from the fence. 22-3 basically left with a simple miter joint that has empty grooves running along its length. Some tearout may still occur, especially on the inside surface of the side (vertically cut) pieces. You can ameliorate this problem by scoring through the top ply before cutting the side pieces. Use a sharp knife about one board thickness from the end that will be cut. Tearout and Shattering Tearout and shattering can degrade both the appearance and the integrity of the joint, especially with cheap plywood. I feel that cutting a lock miter joint with a router bit can work well only on solid hardwood or high quality plywood, e.g., Baltic birch. A minimum of nine plies with few or no voids is needed for good results. With fewer plies the spline may shatter while being cut and you are Figure 6. Carrier for Vertical Cut 131 Table of Contents Notes and Reflections While Shaving Wood Chapter 23 Notes About a PC4216 Dovetail Jig T his jig is made to cut dovetail (DT) and box (finger) joints with equally spaced pins and tails. In practical terms, the only variety in spacing you can get with this jig is by skipping one or more positions. This is offset by the fact that this jig is less expensive than other DT jigs and very easy to set up. Once set up, it is repeatable. The manual that ships with the jig is very useful. It is complete, although it takes some getting used to because it repeats information. This chapter will address a few issues I did not feel were complete: tearout, especially with plywood, and positioning the work to get half-pins and rabetted half-blind DT. Also note, there is an Advanced Instruction Manual available on the internetc that describes, among other things, how to use the templates without the jig, e.g., on a router table or with a template mounted on a clamping board. Use of a clamping board allows the template to be brought to the work and allows joints to be cut in boards of unlimited length. Tearout The first project I made with this jig used cheap birch plywood. Tearout was severe, actually chipping off layers. The DT were not only unsightly, but actually difficult to assemble because the wood was so chewed up. Opinions vary about this. The quality of the plywood matters greatly. Probably it is best to use solid wood pieces to get the cleanest DT but some tearout might still be possible. Of course, the cutters must be sharp. (Mine were c www.portercable.com/jigs/dovetail/ SupplementalManual.pdf Table of Contents very sharp.) In a subsequent project I made box joints in a better grade of birch plywood (9-ply ½" plywood, but not the quality level of Baltic birch plywood). I also used a solid carbide upcut bit. I cut at a moderate rate. I had no tearout) Baltic birch has much less tearout than cheaper plywood if the backer board is solidly against the work piece. The next issue after wood quality and cutter sharpness is the mechanics of routing. The vertical workpiece is supported on the back side (toward the jig) by the waste board. Make sure that it is tight against the workpiece. The clockwise rotation of the cutter means that the left side of the cut is supported, but the right side is not. Therefore, most tearout will be on the right outside as the cutter exits the wood. This has to be supported somehow to reduce tearout. If you are cutting half-inch thick drawer sides, it should be possible to add a waste board in front of the workpiece and clamp them both in the jig. Sandwich the workpiece between waste boards – one horizontal on the top of the jig and one vertical in front of the workpiece. Possibly ¼" Masonite would be stiff enough. Otherwise, you would have to clamp the waste board with separate clamps placed along the sides of the jig. You could cut the DT into a wider board and then cut the board to size after cutting. That way, there would be space on each side to screw on a waste board. You could use an upcut spiral bit to to pre-cut the DT and then follow with the DT cutter. You could scribe the wood with a knife at the point where the cutter would exit. from the wood. You could do a shallow climb cut (right to left) on the right side of each tail 23-1 PC Dovetail Jig (where the cutter exits each tail). Positioning the Workpiece The DT cutter has diameter d = ½" and the template guide collar has d = ¾" therefore the distance from the template finger to the cutter is c". To get a half-pin (¼") at the reference edge (RE), set the RE of the workpiece c" from the edge of the finger. Alternatively, mark the exact center of the workpiece and position it so that it is centered between two fingers. Box Joints The jig makes excellent box joints. The instructions on pp 22-23 of the user manual are fine. Let the drawer sides be the first board (corresponding to the tail board in a DT joint). The sides will be cut against the left offset guide. To set the left vertical offset guide, note that the cutter is ½". The ¾" collar on the router sets the distance from the inside of a finger on the template to the cutter at c". The fingers on the template are ¼" wide. So - to get a full socket at the reference edge(RE), make a mark f” from the RE and align that mark with the right side of the first finger. (½" + ¼" + c"= f") To get a full finger at the RE, make a mark c" from the RE and set it flush with the left side of the first finger on the template. (c + ¼ + c = ½”, a full finger.) If you want both edges of the board to match, i.e., to be symmetrical, mark the center of the board and align it with the center of a space on the template. You probably will want both ends of the board to match, e.g., both ends to have a pin at the RE. You can follow the instruction to put the outside surface against the jig base 23-2 (step 3 on p 22) for only one end, but it does not matter for box joints. Just flip the board end for end and keep the RE against the right offset guide. Let the drawer front and back be the second board (corresponding to the pin board in a DT joint). The second board will be cut against the right vertical offset guide. You can use the above logic to set the right offset guide but the instructions on p 23 are probably better. Loosen the right offset guide. With the RE of the first board facing the right offset guide, center the fingers cut in the first board in the template spaces. Each template finger should be c" from the edge of the fingers cut in the board. It is more accurate to use a c" spacer. Clamp the board in place, push the right offset guide against the RE of the board. Tighten the right offset guide in place. Insert the second board and cut the sockets. To cut the other end of the second board, simply flip the board end for end, keeping the RE against the right offset guide. Through DT Through DT are much like box joints, except that tails are cut on the sides of a drawer to take advantage of the mechanics of a DT joint and the orientation of the pin board matters. The instructions on pp 16-17 of the manual are fine, as far as they go. Position the workpiece as described above. It does not matter whether the outside of the tail board is facing toward or away from the jig. New Step 9: After you cut the first end of the tails board, flip it end for end so that the RE is still against the left offset guide. Pins are cut on the front and back pieces of a drawer. The outside of the pin board must face away from the jig. Turn the template around and position 134 Table of Contents Notes and Reflections While Shaving Wood the board with the RE against the left offset guide per step 1 in the manual. Make sure the pin board is square and snug. Cut the pins at one end of the pin board. Now, to position the other end of the board, you cannot flip the board end for end against the left offset guide as you did in cutting tails because the outside of the board would then be facing toward the jig. Instead . . . a. make a mark on the end you just cut 1c" from the RE. b. loosen the left offset guide Figure 1. How to position the right offset guide. c. Align the mark against the right side of the rightmost finger, where the workpiece meets the backer board.(See Figure 1). Clamp the board in position. d. Set the right offset guide against the right 23-3 side of the workpiece and tighten it in position. This allows you to cut the second end of the pin board relative to the reference edge. e. Flip the board over, making sure that the outside of the board is facing away from the jig, that the RE is against the right offset guide, and snug against the template. Make sure the backer board is snug against the work piece and clamped in position. f. Cut the pins Rabetted Half-Blind DT The manual is a bit confusing here. The spacer (as wide as the rabbet) refers to the rabbet at the top of the pinboard (actually, whichever long edge is the reference edge). For a drawer front that has rabbets only at the ends, no spacer is needed. The manual speaks of having the rabbet as deep as the depth gauge on the halfblind template. Rather than adjusting the depth gauge on the template or choosing a depth for the rabbet that you may not want, it is much easier to adjust the depth of cut on the router. SO-1. Cut the rabbet on the workpiece to whatever dimensions you prefer. 2. Put the workpiece in the jig horizontally with the rabbet up, under the fingers of the half-blind template. 3. Adjust the depth of cut on the router so that the cutter touches the bottom of the rabbet. 4. Replace the pin board with a piece of waste of the same thickness 5. Put the tail board in the jig and proceed according to the manual. 135 Table of Contents 136 Table of Contents Notes and Reflections While Shaving Wood Chapter 24 – Drilling – Using a Drill Press T press for planing segmented blanks flat, but frankly the idea seems uncontrollable and dangerous to me. I do not have a mortising attachment. I cut mortises with a router or drill them out on a drill press and clean them up with a chisel. he drill press is an essential piece of equipment for precise drilling. Most of the holes I drill are pilot holes for screws or clearance holes for bolts. I drill most of these with a cordless hand drill. If the hole I need is through metal, if it has to be perfectly square to the surface, or if I want to drill a series of holes in a straight line, for example, for shelf supports or to hog out a mortise, I use a drill press. A drill press is a very versatile machine. In addition to drilling, it can be used as a light-duty router or shaper, surface planer, drum sander and disk sander. With an attachment, it can become a mortising machine. I do not exploit all these possibilities. I use my drill press to spin small sanding drums that fit in the chuck. Sanding this way is great for cleaning up a band saw cut that is rough and just a bit outside the line. I have a somewhat rudimentary dust collection scheme for my drill press, so I tend to use fairly slow speeds and coarse grit. (A shop vac nozzle laying on the DP table.) That way, most (?) of the sanding dust falls on the table and is sucked into the nozzle. I do heavier drum sanding on my lathe, which is set up for serious dust collection. (Smoothing bandsaw cuts with a spokeshave is faster and does not create dust.) Occasionally, I use a drill press to turn a very small router bit (less than ¼" diameter and less than ¼" depth of cut. The advantage is that I can see the cut because the cutter is above it. The disadvantage is that the work may be harder to control and therefore more dangerous. Routing with a drill press is done at the highest speed. Routing, shaping, and heavy sanding are hard on the bearings and may be very dangerous. I have thought about using a drill Safety Eye protection is a must because debris may be thrown out and up. It is very easy to leave the key in the chuck (if the chuck has a key, of course). When you turn on the drill press the key becomes a missile. (Also, it is very easy to misplace the key.) Luckily almost every drill press has a magnet, clip or recess where you can store the key, so I am just suggesting that you form the habit of seeing the key in its holder every time you reach for the on/off switch. This sounds trivial and obvious, unless, like me, you have forgotten a few times and taken a shot to the chest from a flying key. An adjustable fence is essential because it is a convenient means of controlling the rotation of the workpiece if the drill grabs it. Otherwise, you risk a few really hard raps on the knuckles before you can get your hand out of the way. This is especially likely when you are drilling metal with a twist drill. As the bit exits the hole, a flute may catch. If the metal has a sharp edge it may cut you. Hold downs are also necessary for some operations. In other words, make an accessory drill press table to fit your drill press. It should have a fence that slides forward and backward and locks in place easily, “T” tracks that will accept hold-downs and a removable table insert that you can replace after too many holes have been drilled in it. An insert also allows a sanding drum attachment to reach below the surface of the accessory table. Such a table can 24-1 Table of Contents 137 Drilling be made of 2 pieces of ¾” Baltic Birch or other plywood and some hardwood cleats to fasten it to the primary table. See the design described by Roland Johnson (References). I spent years messing around with partial solutions, and wish I had made a drill press table much sooner. Size Matters The horizontal distance from the post to the chuck is the nominal size of a drill press, and is the limit of how wide a piece you can drill. The capacity of the chuck limits the diameter of drill shanks you can use. The vertical distance from the chuck to the table at its lowest position determines the maximum thickness of work you can handle. The distance the quill travels determines the maximum depth of hole that can be drilled without shifting the worktable. Really deep holes, such as the relief hole through a lamp standard, should be drilled on a lathe. For drilling precise holes in work that will not fit on the drill press table, you can buy portable drill guides that will accept up to a 3/8" shank. These are turned by an electric hand drill and help you to control the angle and depth of the hole. I find that one is useful (essential, really) for drilling dog holes in a workbench. Runout Runout is wobble. Imagine the point of a drill bit held in the chuck. When the chuck rotates, the point of the drill bit should not move through a circle. If it does, the bit will not cut smoothly or accurately. You may even be able to feel some side-to side vibration. Worst, the holes will be larger than the diameter of the bit so dowels will not fit tighly in their clearance holes.. 24-2 Runout could be the result of a bent bit, but if many bits show the same problem, it probably indicates that the chuck is not properly aligned on the taper. In his excellent article (see References) Roland Johnson explains how to check runout and how to correct it. Briefly, you use a piece of straight rod and a dial indicator on a magnetic base to check it. Drill rod is available from a machine shop or on the internet. I used a new drill bit. Put the rod in the chuck, set the tip of the dial indicator against it, and slowly turn the rod. It should not wobble more than .005" at a distance of an inch or so below the chuck. To correct runout, tap the chuck bottom with a rubber or leather mallet until you get the correct tolerance. If that does not work with a few tries, remove the chuck. Remove the chuck by tapping the top of the chuck around its circumference with a rubber mallet until it comes off. Clean off the taper and the mating surface inside the chuck with acetone or mineral spirits, and re-install the chuck. When the tolerance is acceptable, press the chuck against a piece of wood to register it in place. Types of Drill Bits Types of drill bits used in woodworking include twist bits, brad point bits, Forstner bits, auger bits, expansive bits, tapered bits, countersinks, spoon bits, hole saws and circle cutters. (Figure 1) Twist drills are by far the most commonly found, most versatile and least expensive. They can be used to drill wood and metal. Use self-starting bits (auger bits with a small screw thread at the end) only with a bit brace or eggbeater, not with a power drill. Brad point bits are usually able to drill a cleaner hole and may be preferable for drilling clearance holes for dowels, counterbore holes, etc. It is necessary to start them 138 Table of Contents Notes and Reflections While Shaving Wood slowly, however, to give the flutes a chance to start in the wood. Also, remember the starting spur that protrudes beyond the flutes. It may pierce the opposite side of the board if you mean to drill a hole “almost” as deep as the board is thick. Forstner bits are similar to brad point bits and are often used to drill larger diameter holes. They clog easily and can overheat. They have to be repeatedly raised out of the hole to clear the chips. The larger diameter Forstner bits often have a saw-tooth edge. Most wood screws are tapered. A tapered drill with a countersink is best for drilling pilot holes for wood screws, especially in hard wood. The distance from tip to countersink is adjustable. It is important to adjust this distance appropriately (perhaps c" or so short) because if the pilot hole is drilled too deep with a tapered drill bit, the diameter of the hole may be too wide for the end of the screw and the screw will not hold well. Twist drills are available with countersinks, similar to a tapered drill. Solid countersinks (without a longer drill) are also commonly available. A spoon bit looks somewhat like a gouge that ends in a slight point. This is helpful for starting the hole, as the point helps to prevent the bit from wandering. Spoon bits are used to cut tapered and slanted holes, mainly in chairmaking. They are turned with a brace, never with a power drill. Their key advantage over regular drill bits is that the angle of the hole can be adjusted. Also, they do not have a lead screw. Hand drilling is done with an eggbeater drill or a bit brace. It is difficult to drill a straight hole with an eggbeater drill (a lost art, no doubt) and all too easy to break a HSS twist bit. A bit brace looks like a crank with a wooden pillow at one end and a chuck at the other. It uses an auger bit with a tapered 24-3 Figure 1. Four Kinds of Drill Bits (top twist, middle-brad point, Forstner, bottom tapered) diamond-shaped end. These have become somewhat quaint but are very useful for heavier drilling, e.g., for larger diameter, deeper holes in fence posts. Finally, smaller screws can be started with an awl or a gimlet. A birdcage awl is square in cross section and is very effective in making a pilot hole for smaller and shorter wood screws. A gimlet looks like a short tapered bit stuck in a “T” handle. Bit Materials Some types of bits are available in a variety of metals and coatings. Most drill bits for woodworking are made of high carbon steel or high speed steel (HSS). Bits made from high carbon steel are more durable than low-carbon steel bits. If they are overheated, however, e.g. 139 Table of Contents Drilling by friction while drilling, they lose their temper, resulting in a soft cutting edge. These bits can be used on wood or metal. High speed steel (HSS) is a form of tool steel. HSS bits are hard, and much more resistant to heat than high carbon steel. They can be used to drill metal, hardwood, and most other materials at greater cutting speeds than carbon steel bits, and have largely replaced carbon steels. Cobalt steel alloys are variations on high speed steel. They hold their hardness at much higher temperatures, and are used to drill stainless steel and other hard materials. Cobalt steels may be more brittle than standard HSS. Black oxide is an inexpensive black coating often found on twist drill bits. A black oxide coating provides heat resistance and lubrication, as well as corrosion resistance. It increases the life of high-speed steel bits. Titanium nitride (TiN) is a very hard ceramic material that can be used to coat a HSS bit, extending the cutting life by three or more times. However, when the bit is sharpened the new edge will not have the coating. 24-4 recommended speeds. I follow these only loosely. Recommended RPM for Drill Bits Diameter Hardwood TWIST 1/16 - 3/16 3000 3000 1/4 - 3/8 3000 1500 7/16 - 5/8 1500 750 11/16 - 1" 750 500 Brad Point 1/8 1800 1200 1/4 1800 1000 3/8 - 1/2 1800 750 5/8 1800 500 3/4 1400 250 7/8 1200 250 1 1000 250 Forstner 1/4 - 3/8 2400 700 1/2 - 5/8 2400 500 3/4 - 1" 1500 500 1c - 1¼ 1500 250 1d-2 500 250 Circle Cutter Spindle Speed The speed at the cutting edge of a bit or circle cutter is proportional to the diameter of the bit. Slower spindle speeds are preferable for metal and large diameter bits in wood. This is especially important for hole saws and circle cutters. High speeds are preferable for routing and shaping wood. Most cordless hand drills now have multi-speed motors. Some drill presses have multi-speed motors, others use a system of pulleys to change speed. With a 1750-rpm motor, depending on the ratios of pulley sizes available, you may have speeds of 700, 1250, 2400 and 4700 rpm. The following are Softwood 1-1/2 - 3 500 250 3-1/4 - 8 250 250 Countersinks 2-flute 1400 1400 5-flute 1000 750 References Drill Press Rules, by Brian Murphy, American Furniture Design Co. Johnson, R, Get More From Your Drill Press. Fine Woodworking January/February 2006 pp 60-65 140 Table of Contents Notes and Reflections While Shaving Wood Chapter 25 – Reflections on Using Jigs O interrupt and delay a project. Sometimes this is in fact true, and sometimes making a jig just feels as if it were an interruption in the process of getting something made. After you experience the increase in speed and accuracy made possible by the jig, you may not feel that way any longer. Without exaggeration, I think that this is one of the main issues that separate craftsmanship from hacking around in a workshop. To be fair, sometimes I have spent hours figuring out what sort of jig I needed and then making it. Ian Kirby’s philosophy is to make simple, quick single-purpose jigs. He says that multi-purpose jigs take too much time to make and too much fiddling to adapt to each use. He may be right. When I make a jig, however, I am usually thinking that I may use it again many times in future projects. A very familiar example will illustrate the value of using jigs. This may seem extreme, but it is historically accurate. Consider a table saw. The tool itself is the motor assembly, the table, the blade and the stand. This basic saw is somewhat useful but unsafe, although some people actually use them just like that. Most of us would not be able to accomplish much with a table saw in that condition. Two jigs increase the usefulness and safety of a table saw so much that nearly every table saw sold comes with them, so much that we tend to think of them as part of the saw itself. Of course, I’m thinking of the rip fence and miter guide. If you appreciate the value that these two jigs add to a table saw, you may see how useful other jigs may be with a table saw or with other tools. My shop is full of jigs (and I do mean full). The most useful and necessary jigs in my nce I thought that a mark of an accomplished craftsman was the ability to carry out freehand work. While there is some truth to that, I have changed my mind. Now I see that an accomplished craftsman knows when and how to use jigs and fixtures. A jig is a device that holds a workpiece in a particular position or relationship and guides the tool to the work or the work to the tool. A jig's primary purpose is to provide repeatability, accuracy, and interchangeability in the manufacturing of a product. A fixture holds the work in a fixed location. You are already very familiar with jigs and fixtures, although you may not recognize them as such. Some jigs have become so familiar that they seem an integral part of a tool. (I’ll give an example below.) Others, however, have to be purchased or made as after-market accessories. Many chapters in this book describe one or another jig. The point of this reflection is to encourage you to appreciate their value and recognize when a jig might be useful. Jigs extend the craftsman’s skill by increasing the precision and predictability of operations, including, of course, safety. To a developing craftsman, learning about jigs is part of learning correct methods of work. Jigs demonstrate a lovely idea – we can use our intelligence to extend the skill of our hands and the capacity of our tools. Learning about how to make and use jigs has been a major part of my development as a woodworker. Before I got serious about woodworking, I don’t recall using shop-made jigs more complicated than push sticks. Since then, almost every project has required one or more jigs, often a new jig. The disadvantage of using jigs is that making them takes time and may seem to 25-1 Table of Contents 141 Reflections on Using Jigs shop, after the miter guide and rip fence on my table saw, are a router table with its fence and starter pin; a variety of push shoes and small parts carriers; a big crosscut sled with a miter attachment (see Chapter 16); a “pattern making jig” (see Ian Kirby); a mortising jig for my router (see Chapter 20); a tenon jig for my table saw (see Chapter 19); a bench hook and shooting board; a thin strip ripping jig (see Chapter 15), and a secondary drill press table (see Chapter 24). As I said, there are many more jigs included in various descriptions throughout this book. A recent example was when I wanted to drill 1/4" holes in some cylinders (wooden mugs) to attach a handle. The holes had to line up parallel to the axis of each cylinder and be exactly radial (so the handle would not be crooked); and be precisely located to receive corresponding dowels in the handle. The cylinders had taken three days to make. I would get no do-over. I considered holding each cylinder in a hand screw but it slipped every time I moved it. I was afraid to tighten the hand screw too much because I might have cracked the mug. I considered measuring the distance between the holes, but the holes had to be exactly the right distance apart.. Finally, in frustration, I made a template to position the holes (a strip of wood with two holes drilled in it) and a vee-shaped cradle to hold the cylinder. The cradle held the cylinder in place with just finger pressure and I could level it so that the holes were 25-2 exactly radial. I used the template to drill holes in both the mugs and the handles. When thinking of ways to perform an unfamiliar operation, always stop a moment and think of how a jig might make it more accurate and safer. References Anon. Building a Sliding Table for Your TablesawKWWSZZZKLJKODQGZRRGZRUNLQJFRP Gochnour, Chris. Expand your workbench with versatile bench hooks. Fine Woodworking Johnson, Roland. Get More from your Drill Press. Fine Woodworking, Jan-Feb 2006. Kirby, Ian The Accurate Table Saw, Bethel CT The Cambium Press 1998. (Fresno CA, Linden Publishing 2004). Kirby describes many jigs, including the pattern making jig and panel raising jig. Schlebecker. Pete. Using push sticks. Shopmade Helpers Are a Cheap Way to Stay Safe. Fine Woodworking #186 (Sept/Oct 2006) Speas, Ed. Shooting Board Aims for Accuracy. Fine Woodworking May/June 1994 142 Table of Contents Notes and Reflections While Shaving Wood Chapter 26 – Sharpening Hand and Turning Tools W on. For flat chisels and plane irons, I have settled on high-speed dry (HSD) grinding for re-shaping an edge and sandpaper sharpening for honong and polishing. I prefer HSD grinding followed by a slow, wet wheel for turning chisels, but I don’t have room in my shop (or budget) for a Tormek. HSD grinding alone will suffice. I use HSD to sharpen turning scrapers. I use a file for card scrapers. Admittedly, some mystery remains, so maybe that explains why we don’t all agree on the best method. orking with sharp tools is a joy. Granted, sharpness and dullness are relative terms, and you may think your tools are sharp. I certainly thought my tools were sharp. You may also feel that your skill is deficient when a tool does not cut as well as you wanted. For example, I wondered if I would ever learn to use hand planes and paring chisels, and if I would ever get shavings instead of chips when turning wood on my lathe. In each case, after I learned how to make my tools sharp – sharper, if you prefer, my skill level with the tool seemed to improve dramatically, and I made fewer mistakes. Trying to define a standard of sharpness is difficult. Let me say that a sharp cutting tool like a chisel or plane iron should be almost as sharp as a new razor blade. It should shave hair off your arm with no pulling. A sharp tool should slice a piece of newspaper suspended from your hand. This is a challenging goal. Many woodworkers – not only novices – lack confidence in their ability to get and maintain a keen edge. This has two important consequences for them. First, if you cannot sharpen hand tools you may tend to compensate by using power tools when they may not be appropriate. Second, you may be tempted to try a plethora of expensive methods, jigs, and gimmicks that are touted to overcome that lack of confidence. Many competent woodworkers believe that you need these. I do not agree. Sharpening, can be mastered without gimmicks. I recommend picking a method and then learning how to get acceptable results with it. I have tried a variety of sharpening jigs, oilstones, and waterstones; hi-speed, lowspeed, wet and dry grinding wheels; and so High Speed Dry Grinding There was a time that I would not have touched a good woodworking chisel or plane iron to a high speed wheel unless it was badly damaged and I was willing to risk ruining it. I have learned how to grind without damaging most tools. Experience in sharpening turning chisels also has modified my attitude about this somewhat. It is all too easy to ruin the temper of some steels (carbon steel, especially) by overheating the thinner area near the edge as you grind. A bluish discoloration indicates that a carbon steel edge has been hot enough to remove its temper. One light pass on a high speed dry grinding wheel, to restore a bevel or edge, may be OK for any tool but changing the bevel angle or otherwise reshaping the edge requires technique and patience to avoid overheating the tool. High-speed steel (HSS) chisels are safer to grind dry on a high speed wheel, but even HSS can be damaged if allowed to overheat. So, here are the main points of proper HSD grinding technique. 26-1 Table of Contents 143 26-2 Sharpening 1. Use an aluminum oxide (ALO, alumina) wheel with a softer bond. Pink, blue or white grindstones have a softer binder, which allows the wheel to wear faster, without glazing, and thus present fresh, sharp surfaces that cut faster and cooler. Avoid grey grinding wheels for sharpening woodworking tools. Cubic Boron Nitride wheels are better because they cut faster and do not produce as much heat. They are very expensive. 2. A coarse stone, e.g., 60 or 80 grit, removes steel quickly and reduces heating. A fine grit stone is more likely to overheat the tool. 3. Use a large, flat, stable tool rest such as the Veritas Grinder Tool Rest. The stock tool keep the wheel from developing grooves or getting out of round. Dress the wheel until it runs true. Dress it so that the surface is slightly crowned (convex). This will make it easier to control grinding rate and heat. 5. If you plan to remove very much steel (more than just restoring a bevel) use a Sharpie pen and a square to mark a line that is square across the back of the cutting edge and check frequently to assure that you are grinding the edge square to the sides. 6. "Let the stone come to the tool", i.e., do not press the tool against the wheel. Use a light touch, and keep the tool moving. Keep your finger close to the edge being ground. If the area ½" or so back from the edge is not too hot to touch, the edge is not too hot. (It can get hot very quickly, however, especially if you press the tool against the wheel.) 7. Quench the tool in water frequently. The little trough on my grinder is inadequate. I use a plastic tub of water. 8. When sparks come over the top of the tool, and you can feel a slight burr along the cutting edge, you have ground enough. Figure 1 Codes Used on Grinding Wheels rests that ship with most grinders are totally inadequate for grinding woodworking tools. The tool rest should be big and stiff enough to support the tool at the desired angle as you grind across the width of the blade. It should be adjustable to the angle you want on the bevel and close to the wheel (within c") so that you can easily slide the tool across the rest while you are grinding. Some steels will still overheat right at a corner of the edge if you don't keep the chisel moving at the right speed across the wheel with a light touch. Dressing the wheel with a slight convexity can alleviate that. A grinder that can turn slower (some will turn at 1700 rpm) may also help. HSD grinding puts a concave bevel on the tool which makes honing easier. It especially makes it easier to hone without a jig. 4. A diamond wheel dressing tool is a must, to 144 Table of Contents Notes and Reflections While Shaving Wood Honing Flat Chisels and Plane Irons -Sandpaper Sharpening ("Scary Sharp") Any flat-bladed woodworking tool, such as a chisel or hand plane iron (blade), can be sharpened with this method. Sandpaper sharpening (SS) is fast, easy to learn, and inexpensive compared to waterstone, oilstone, and, especially sharpening machines. Advice on this method is widely available. See, in particular, the article by Hochstetler and Tindall in the Badger Pond Archive MATERIALS REQUIRED 1. Abrasive. An assortment of wet or dry silicon carbide sandpaper in various grits. Sandpaper with pressure sensitive adhesive (PSA) backing is preferred. This is available from auto body suppliers and some woodworking suppliers, e.g., Lee Valley/Veritas or Tools for Working Wood. Coarser grits (e.g., 120-400) may be needed only for the first flattening and sharpening of more coarsely milled tools or for removing nicks (although the latter is very tedious by any hand method). If the tool is already somewhat sharp and just needs touching up, the finer grits (CAMI 600, 15 micron; CAMI 2000 or 5 micron) may be sufficient. It is important to “run the grits” and not skip from coarse to fine paper too quickly. Especially if you have re-shaped an edge with coarser grits, each successive grit should be able to remove the scratches left by the previous one. The following table gives an example of a sequence of grits, but the exact sequence is not critical. When possible, buy abrasive paper designated by grit size in microns rather than CAMI number. It is not really necessary to go much beyond 15 micron paper. However, it does not 26-3 cost much money or time to go finer, and some people believe that an edge will last longer the more finely it is polished. A Recommended Grit Sequence Designations and Micron Size Microns (approx) CAMI Grit Difference (Microns) 60 220 36 320 24 23 400 13 16 600 7 1000-1500 8 8-9 1 7 .5 .5 2. Flat Surface. A perfectly smooth, very clean, flat and hard surface at least 12" x 12", e.g., float glass at least 1/4" inch thick, a cutoff from a polished stone or Corian counter top, or a piece of Lexan backed up by MDF or flat plywood. 3. Adhesive. If PSA-backed sandpaper is not available, sometimes just wetting the back of the paper will be enough hold it in place. If adhesive is used, it must allow the paper to be peeled off in one piece and should be soluble in acetone, lacquer thinner, etc., for cleaning. I sometimes use sanding disk adhesive, applied in a very thin smooth coat. Lee Valley sells a transfer tape that will lay down a thin uniform layer of contact adhesive. It leaves the adhesive on the surface when the tape is pulled off. Spray adhesive, such as 3M #77, can be used. It is messy and tends to form lumps under the sandpaper that will catch the blade. You could use double sided tape along the long edges of the paper but you would have to 145 Table of Contents 26-4 Sharpening sharpen only between the ridges over the tape. 4. Sharpening jig. A sharpening jig is a great help, especially at first. I recommend the Veritas MK II (05M02.10) "sharpening system" jig and angle gauge. If you are sharpening a previously honed tool you can learn to copy the angle and sharpen by hand. PROCEDURE Start by adhering a third sheet of sandpaper to the glass surface. The finest grit papers may stick well enough just by wetting their backs, but while you are learning it is better to glue them down. Glue them near the edge of the plate, because you will need to be able to reach the paper from the edge to remove the wire edge. I sometimes use a bit of water to lubricate and suspend the swarf. It’s a bit messy, but I feel that it may speed the cutting somewhat and extend the usefulness of the paper. Lap (flatten) the back of the tool near the cutting edge (1-2" back from the edge is fine). It is seldom necessary to lap the entire back of a plane blade or chisel. The right starting grit is not too difficult to figure out. Try a few strokes with, say, 320 and judge if the scratches from the sandpaper are deep enough to remove the imperfections without too many strokes (e.g., fewer than 50). Hold the back of the blade flat on the surface of the sandpaper and stroke firmly, back and forth or in a circular motion. Hone it flat, i.e., do not cut a back bevel, especially not on a chisel. Make sure the pressure applied is even. After 20-30 strokes, examine the back. It should display an even pattern of scratches or shininess. Areas unscratched from the sandpaper indicate the need for further lapping. The cutting edge is the critical part. Examine it carefully to see if the lapping Table of Contents extends all the way to the edge. Brush or flush the metal debris from the sandpaper. If the back of the blade is not evenly scratched, repeat for 20-30 more strokes. Once an even pattern of scratches is seen on the coarsest grit, extending all the way to the cutting edge, the back is flat but not polished. Switch to the next finer grit, and repeat, working your way up to progressively finer grits. Each grit will leave a finer and finer set of scratches on the back of the blade. It's actually inefficient to skip a grit. You want to effectively "erase" the coarser scratches with the next finer grit, before moving on. By the time you have finished with the finest 2000 grit or 0.5 micron sandpaper, you should see a mirror reflection on the back of the blade. The back is now flat and polished and the bevel is ready for honing. CHOOSING A BEVEL ANGLE Most chisels and plane blades are ground at a bevel angle of 25E as they come from the manufacturer. When starting out, you may just want to keep the original angle. If you want to reduce a bevel angle, be prepared for a lot of work with coarse paper. A grinding wheel is much quicker, of course, but you must avoid overheating the edge. (Dry grinding is discussed above.) The optimum bevel angle for a hand tool depends on the type of steel, the type of tool and the use of the tool. The lower the angle, the easier the tool will cut, and the weaker the edge will be. On the one hand, a well tempered chisel that would only be used for paring, carving or turning, and would never be struck with a mallet could have a bevel angle of 20E or less. At the other extreme, a mortise chisel, made of somewhat brittle steel, that would be used with a mallet, would require a bevel 146 Notes and Reflections While Shaving Wood angle of 30E in softwood and 35E or more in hardwood. Steve Elliott has shown that an A2 plane blade is prone to chip when planing hardwood if the bevel is less than about 34E. Also, his work confirms the common sense that higher angles will stay sharp longer. HONING I usually want visual feedback about where the sandpaper is cutting, i.e., whether I have the correct bevel angle. To get this, I mark the bevel with a Sharpie pen before I start honing. Turn the blade bevel side down, and clamp it in the sharpening jig so that the bevel is exactly flat on the surface of the paper. Square it to the axis of the guide roller. (If the tool is not exactly perpendicular to the axis of the guide roller, you will not sharpen evenly across the edge, in effect, you will sharpen a skew into the blade.) Bevel Angle Intended Use 15 to 20 Paring chisel, skew chisels, low-angle planes for softwood, skew-blade planes./20 to 25/All of the above (except skews) intended for hardwood or end-grain. 25 to 30 Chisels used both for paring and light mortising, firmer chisels for softwood, most plane blades and spokeshave blades. 30 to 35 Mortise chisels, firmer chisels for hardwood, plane blades for hardwood and smoothing planes. 35 to 40 Mortise chisels for heavy use, particularly with brittle steel, bevel up plane blades for difficult wood. Based on a Veritas/Lee Valley Product pamphlet Don't rely completely on the jig. You Table of Contents 26-5 are sharpening the tool – the jig is only helping. Push the tool forward with both hands. Apply light downward pressure on the tool with your fingers a little forward of the jig, and horizontal pressure on the jig with your thumbs, a bit downward. On the return stroke, don't press down. It takes a little practice to keep the bevel firmly against the paper on the forward stroke without tipping it over. One tip-over may undo the progress of many sharpening strokes. Work slowly until you get the feel of it. Examine the bevel edge often to make sure that the newly cut bevel is equally wide across the edge. You should be able to feel a fine wire edge across the entire width of the edge. These indicate that the cutting edge is 90E to the side of the blade. Again, work your way up through the grits, coarser to finer. Once the 5 micron sandpaper is reached, there will be an even mirror finish all across the bevel. After the 0.5 micron grit the blade should be as sharp as a new razor blade. Some people advocate that you hone a micro-bevel. If you want a micro-bevel, raise the angle of the honing guide by about 2 degrees and repeat the honing with a few strokes on the 5 micron/2000 grit sandpaper. Finally, lap the back of the blade lightly with the finest grit, removing the wire edge from the back of the blade. You may have to repeat this final honing, alternating between the (micro)bevel and the back several times, to completely remove the wire edge. STROPPING Stropping is a process of polishing the bevel of a cutting edge. Usually, just like the old-time barber, a honed edge is stropped on a piece of leather. The difference is that the leather should be glued to a flat surface, and that the leather may be "charged" (covered) with rouge. 147 26-6 Sharpening Stropping is done away from the edge, the opposite direction from honing. Press the bevel lightly onto the surface of the leather and pull the edge backwards. Then flip the cutter over and strop the back. The precautions are: (1) Don't push the edge forward against the leather. You will just cut the leather and possibly dull the edge a bit. (2) Don't push the edge down into the leather too hard – just wipe it firmly against the leather. (3) When stropping the back, be sure to keep the back flat against the leather. If you tilt it you will cut a back bevel. Before you strop, consider the grit size of the rouge. There would not be much to gain if you stropped with 5 micron rouge right after honing with 0.5 micron paper. Stropping is, however, very useful for maintaining (touching up) a sharp edge. I sometimes strop blades on a felt wheel, just to touch them up once or twice very lightly between honings on micro-grit sandpaper. I first learned about stropping with a felt wheel as a woodcarver's trick, and its best application is with knives and freehand carving tools. It is a bit tricky, however, for chisels and plane irons, so I hesitate to recommend it. Figure 2. Hard Felt Buffing Wheels on Arbor Table of Contents Figure 2 shows one setup. The wheel has to be turning away from the edge, the opposite of grinding. I think 800 -1000 rpm might be about the maximum speed. I use only hard felt wheels (available from Lee Valley, among others) with whatever rouge or compound is available. Very light pressure is important. If the chisel indents the wheel at all it will dull the edge. A brief, (1-3 seconds) light touch is all that I need to restore an edge so that I can keep working a while longer before getting out the sandpaper. You can touch up the back of a chisel if you are very careful not to cut a back bevel, which can virtually ruin a mortising chisel, etc., whose back must register against the work. "Ruin" might be too strong a word, but it can take a long time to lap the back of a chisel flat again. MAINTAINING A SHARP EDGE The rule is, sharpen frequently but very lightly. I keep the strop or sandpaper plate handy, so I can touch up edges with little or no fuss. After a while touching-up and stropping will not restore the edge and you need to return to 15 micron sandpaper. This is much quicker if you can hone by hand. This is easiest with thicker blades, especially if they are hollow ground. Press the bevel down firmly with your thumb placed directly over it. (On thicker chisels or chisels with a fresh hollow ground edge you can feel it settle or "click" into position when the bevel is flat on the paper.) Your thumb should be right at the edge. Hold the chisel with both hands, as close to the edge as possible, pressing down with your thumb. Keeping your elbows close to your sides, rock from side to side, thus moving the edge across the sandpaper while maintaining pressure behind the bevel. Only a few short strokes should be 148 Notes and Reflections While Shaving Wood necessary at each grit to restore the edge. The back of the blade should be touched up lightly to remove any wire edge. When I put my cutting tools away, I wipe them with a rag impregnated with paste wax and Boeshield. SHARPENING STONES The classic sharpening method used by our forebears was sharpening with an oilstone. Oilstones are hard and do not wear very fast. They tend to become clogged (glazed) with metal swarf. Oil is traditionally used to lubricate the stone and to suspend the swarf. Water can also be used on a new (or freshly cleaned) oilstone. Non-oily sharpening lubricants are also available for use on oilstones. Based on the amount of discussion on the web, sharpening with waterstones is today the most common method among people who are serious about sharpening. Or, maybe, the amount of discussion reflects the difficulties in choosing and using waterstones. Waterstones are soft, and wear away easily in use. This is considered an advantage because it continually exposes new, sharp, grit, thereby increasing cutting efficiency. It also means, however, that waterstones develop depressions that must continually be removed. Diamond hones and ceramic stones are also used, especially for finer grits. These are the most expensive and, therefore, get the most commercial promotion. The procedure for sharpening with stones is the same as with sandpaper, except of course stones are thicker and shorter than sandpaper sheets, so more strokes will be needed. Also, the selection of grits will be limited. Waterstones, especially, and oil stones to a lesser extent, need to be flattened frequently, because they seldom wear evenly 26-7 from use. A stone with a dished surface will, naturally, produce a curve in the edge it produces. Various flattening tools are used, including coarse diamond stones. Waterstones are flattened most easily and cheaply by using drywall screen, a mesh that has been impregnated with abrasive. Drywall screen is available in hardware stores. It is placed on a flat surface (similar to sandpaper sharpening) and the waterstone is rubbed on it until it is flat. The easiest way to tell when the stone is flat is to scribble across the face of the stone with a pencil. When all of the pencil lines have been worn away, the stone is flat. Oilstones are much harder than waterstones and will quickly wear out the abrasive grit on a drywall screen. Silicon dioxide (SiO2) paper on a flat surface or a diamond plate is necessary to flatten an oilstone. The grit size used to flatten waterstones is probably not very important. Since oilstones are harder, final flattening of an oilstone should probably be accomplished with paper with a grit size similar to the stone itself. If the grit size of an oilstone is not known (as is often the case) you should just use paper of approximate fineness, e.g., 600. Sharpening Turning Chisels Sharpening turning chisels is quite different from sharpening flat chisels and plane blades for "stationary" woodworking. Most turners sharpen on high speed grinding wheels. An experienced turner once told me to forget everything I had learned about honing and polishing plane irons, flat chisels and carving tools. There was a bit of wisdom in that. A beginning turner really needs lessons or a good book. I can't do justice to the topic here, just offer some comments about my 149 Table of Contents 26-8 Sharpening experiences. Richard Raffan's DVD (See References) shows you how he does it and offers good advice. SLOW, WET GRINDING Wet grinding is a good alternative to high-speed dry grinding for carbon steel especially. I once had a big (12") Dunlap (Sears) fine wet grinding wheel. It is a 70 year old antique (like me). Although it is messy and inconvenient, it was my preferred method for honing turning tools. The binder in this grindstone is fairly soft, which means that the wheel cuts well, but also that it needs to be kept flat and that it produces slurry as a byproduct that should be cleaned out of the reservoir periodically. I put it in a shallow plastic tub to catch the runoff. When I was using it, I added just enough water to keep the wheel wet. The direction of rotation was "up" as you face the front of the tool – the opposite of a normal dry grinder. This greatly reduced the amount of water that ran down the tool onto my hands (clothes, the floor, . .). This direction of rotation prevented the tool from digging in, which it occasionally did when the rotation was downward, but it also tended to lift the tool occasionally and pull it. I believe that this indicated that I was pressing the tool against the wheel too hard. Dry high-speed grinding on a wheel yields a slight hollow-ground bevel, which also registers well against the work when turning. I suppose that hollow grind would become less convenient as the radius of the hollow-grind approached the radius of the work (or vice-versa). I don't find that flat versus hollow ground bevels make much difference in use, but it is important that the bevel not be at all convex from multiple facets being ground on it, e.g., by a hesitant, inexperienced operator. Table of Contents Raffin and Rowley (see References) both recommend a light touchup with a white or pink 80 grit stone, no fine honing or stropping, straight back to the lathe. The burr edge cuts well, Raffin says. I have used a white 120 grit stone with good results. I occasionally hone by quickly touching the edge to a buffing wheel. Raffin recommends the following technique for touching up a chisel. Using a tool rest, start with the handle low, so the back edge of the bevel (toward the handle) touches the wheel first. Then raise the handle until the whole bevel contacts the wheel. When sparks come over the top of the chisel, the chisel is sharp. Stop. BEVEL ANGLE There are three basic shapes of turning tools, and there is a lot to learn about sharpening them, especially about forming and sharpening turning gouges, which usually have a rounded (fingernail) shape. Gouges have to be rotated in two axes while the exact bevel is maintained against the stone. The tool can burn if you are not quick and fluid. Skew chisels are a bit easier, but they have a bevel on both sides and, of course, that 60E skew. Choosing a bevel angle is more complicated for a turning chisel than a woodworking chisel. In addition to sharpness and edge strength (see Part I), choice of bevel angle for turning tools has to consider the relationship between the tool rest and the work. With a few exceptions, a turning tool is supported not only by the tool rest, but also by the bevel rubbing the work below the cut. For example, in spindle turning the tool must be on the tool rest, then the bevel rubs the work, and then the handle is raised until the edge cuts the work. Some bevel angles are awkward to use, and very different bevel angles on different tools may require you to continually change the height as well 150 Notes and Reflections While Shaving Wood as the horizontal position of the tool rest each time you change chisels. When hollowing a bowl, the gouge is turned on its side. It is easier to rub a smaller bevel angle near the opening of the bowl, but a steeper bevel is easier to use near the bottom of the bowl. At my level of experience and skill, this is too complicated. When the bottom of the bowl is getting flatter, I use a scraper rather than a steeply ground gouge. Each of my wheels has a Veritas tool rest. I use the existing bevel angle or a bevel gauge to set the angle of the tool rest. I lock the tool rest on one wheel to give me a 30E grind for gouges and leave it there. I set the tool rest on the other wheel at about 45-50E for scrapers. SKEW CHISEL The skew chisel for turning is beveled on both sides and should be sharpened equally on both sides. On this tool the sharp edge should be skewed at about 60-70º. The length of the bevel should equal about twice the thickness of the chisel at the point where it is sharpened. (By trigonometry this is a 14º angle to the side. Other resources say that the bevels should form an angle of about 20º with each other, which also means 20º with the flat sides. I use a large tool rest (Veritas) to maintain the bevel, and to hold the chisel horizontal (parallel to the axis of the wheel). Then I slide the chisel up to the wheel and along the grinding surface of the wheel. If the angle is correct, I see sparks coming over the top of the chisel. I tried using the Veritas sliding tool rest accessory but it is quite clumsy. I'd rather learn how to slide it across the wheel with only the main tool rest as a guide. I put water on the top edge of the chisel and take care if it starts to evaporate Table of Contents 26-9 quickly. Or – I keep my finger near the edge as Raffin recommends. Cool the chisel in water occasionally when using a dry wheel. The wire edge or burr edge left by the emery wheel may be removed by honing with a slip stone or a piece of 120 grit sandpaper. Just be careful not to let the hone tip over the edge or you will have to sharpen the tool again. The chisel is sharp when the sparks come over the top of the chisel, the edge shaves hair off your arm, and makes shavings rather than chips. GOUGE Wood turning gouges are beveled on the outside and ground so that the nose is rounded. According to Rowley (pp 28-29) the bevel ideally should be about 35E for a spindle gouge, 45E for a roughing gouge, and 55º for a bowl gouge. The exact choice of bevel angle and fingernail shape depends on the purpose of the tool and personal preference. Sharpening a turning gouge requires you to swing the handle from one side to the other, while rotating the chisel at the same time. A long bowl gouge makes this a bit more difficult because the handle is in the way while you are trying to swing it smoothly. It may be easier to stand to the side and to hold the gouge by the blade or ferrule rather than the handle. The roughing gouge, which has a less rounded nose, may be a bit easier – mainly rotate it half way around and back again, while swinging the handle slightly to maintain the slight round. A roughing gouge is heavier and flatter than other gouges, and not as rounded. A bowl gouge or spindle gouge is more rounded, to a "fingernail" shape. The cutting edge should blend into each side. This is necessary to avoid the sharp corners that would result if the nose were left 151 26-10 Sharpening straight across as with an ordinary wood-working gouge. The round nose permits the tool to be rolled to the side while cutting beads to avoid scraping the work. The length of the bevel should be about twice the thickness of the blade at the edge. Work a rounded slip stone inside the gouge to remove the wire edge. Be careful not to put a back bevel on the inside of the gouge. PARTING TOOL The parting tool cuts by scraping. It should be ground on both sides. The widest point of a normal "diamond" scraper should be in the center of its cutting edge, and the bevels must meet exactly at the center, the widest point, and should make an angle of about 50º with each other. If the bevels do not meet at the widest point the sides of the scraper will rub against the revolving work and overheat. The work may burn and the tool may overheat. The bevel should be hollow ground slightly. A thin parting tool should be ground at about 45E. SCRAPING TOOLS Scraping tools are best used as finishing tools, although they can be indispensible for finishing a hollow vessel in face-plate work. They are sharpened on one side only, and the bevel is about 60E (the bevel is twice the thickness of the scraper). Scraping tools become dull quite easily as their edges are in contact with the wood almost at right angles. I sharpen scrapers on a high-speed grinder. One quick pass, I stop when I see sparks fly over the top of the tool. Scrapers are somewhat under-appreciated tools in woodturning, just as they are in flat woodwork. (Many of the "wonder" turning tools on the market are carbide scrapers under another, more glorified, designation.) A dull scraper produces dust and may cause tear-out. A well-prepared scraper should remove very thin, intact shavings, not fine dust. Of course, a scraper can be used as it comes from the grinder, or the burr edge can be tuned or adjusted. You can rub the back of the freshly-sharpened scraper on a piece of 400 grit sandpaper to remove the burr and then raise another, smoother, burr with a burnishing tool. Veritas sells a purpose-made burnisher for turning scrapers, or you can make a close copy. Alternatively, you can burnish a scraper with a straight burnisher or the broad side of a skew chisel. A scraper with a burr or hook raised in this way may cut quite differently than one straight from a grinder. In particular, the cutting angle may be different and can be quite agressive if you used heavy pressure while burnishing. Appendix: Steel Type, Edge Life and Bevel Angle Steve Elliott has scientifically tested various kinds of steel in plane blades. (http://bladetest.infillplane.com/index.html) His testing was intended to answer four main questions. The last question is of principal interest here: •With equally fine honing, do some blades reach a higher level of sharpness than others? • Which honing abrasives work best on each type of steel? • How well do blades of different steels hold an edge when used to plane different species of wood? • What bevel angles give the best performance for various steels in different species of wood? He tested five alloys: carbon steel, A2 steel, and three containing significant amounts of 152 Table of Contents Notes and Reflections While Shaving Wood carbide particles. All blades could be sharpened to a very keen edge. The carbide steel blades needed to be honed with diamond or chromium oxide to reach their greatest level of sharpness. No amount of careful honing on a fine waterstone brought those blades to the same sharpness that the waterstones produced on high carbon steel or A2. He found that the carbide alloys held their sharpness longer than plain steel and that A2 blades retained their sharpness longer than the carbon steel blades. With a 34º bevel on a Hock A2 blade, the force required to plane cherry about doubled after 140' had been planed. Considering a 6" wide by 4' board and a 2½" plane iron, that would be equivalent to about ten full-width, full-length strokes. (!!) The blade would still be sharp enough to use. The A2 blades needed to be resharpened after 500' (about 40-50 strokes in my example). This means that a sharpening system has to be easy and quick so that we can sharpen often (and lightly). The Hock A2 blade showed significant chipping at 32º. Elliott recommends a bevel with a total angle of 34º, 26-11 formed by a 31½º primary bevel and a 2½º secondary bevel. The A2 blades showed minor chips after planing cherry, so for harder woods a larger bevel angle may be needed. Elliott tests sharpness by measuring the force required to cut a 40 wt. rayon thread. According to him, sufficient sharpness for all but the most demanding woodwork is where 100 G is required to cut the thread. This is the point where the edge “shaves hair like a dull razor, catches on fingernail.” References Based in part on WoodTurningWiz.com material from Raffan, Richard. Turning Wood With Richard Raffan. Newtown, CT. The Taunton Press, 2001 (Book and DVD) Rowley, Keith. Woodturning: A foundation course. New Edition, Lewes, East Sussex, England. Guild of Master Craftsmen Publications, Ltd. 1999 153 Table of Contents Notes and Reflections While Shaving Wood Chapter 27 – Hand Planes A had tried to remove too much wood with too dull a blade. s a life-long home handyman, I assumed I knew something about hand planes. I was wrong. I had an older (1960-70) #4 Stanley that was quite frustrating to use. I was surprised (and a bit skeptical, I must admit) to learn that some people actually preferred to use hand planes instead of power tools for some operations. I was curious and wanted to learn more. Difficult Lessons A diagram of my learning rate would look like an elongated “S”: slow (and frustrating) at first, very rapid for 2-3 years as I started to get the hang of things, then slower again, but satisfying instead of frustrating. Many years along, I still am continuing to learn and refine my methods (and spend money on handplanes). Learning about planes and how to use them was difficult at first. I was skeptical about much that I read, and I had to find a way to break the following vicious cycle. Hand planes require a surprising initial investment of time and money. You need to appreciate their value to your work before you are willing to make those investments, but you are unlikely to appreciate their worth until you make those investments. You may need experience working with a plane before you can know what to buy, but you need a good plane before you can get that experience. By a good plane, I mean one that will not be part of the problem, one that will simplify learning. My advice is, take the leap. Perhaps the best way, if you can manage it, is to find someone with decent tools to teach you. Perhaps a woodworking store will let you try a well-fettled plane. Next, would be to buy a plane that is ready to use, so that you know how a good plane works, and so you can learn to use it correctly. Such a plane will not come from Home Depot or Harbor Freight. You could buy a Veritas or LN, or a reconditioned one. I bought a Veritas to find out what a good plane was like. I’m glad I did. Four Things I Did Not Know I found that I did not know quite a bit. First, I did not appreciate how useful hand planes are. They have now become indispensable in my work. Second, I did not appreciate the variety of hand plane types that would become essential to my work. Third, and probably most important, I incorrectly assumed that a plane can be used as received from the manufacturer. (I also assumed, again incorrectly, that I knew how to sharpen a plane iron). All but premium hand planes need some tuning, as I describe further below. Inexpensive planes may require extensive tuning. The best hand planes, e.g., Veritas/Lee Valley, may need only final honing. Apparently this is also true of other high-end planes like Lie-Nielsens (LN) and Cliftons. Also, there are people who sell refurbished (tuned) planes that may also be ready to use as received. I have no personal experience with them, yet. Fourth, if planing is to become a routine method of work, hand planes require more technique (and, eventually) more subtlety than I had imagined. Among other technique issues, there is the amount of wood to remove in each pass (usually a few thousandths of an inch or less). For years, I 27-1 Table of Contents 155 Hand Planes 27-2 Price How much to spend on a plane, and the best value for money, bothered me a lot when I first realized what was available in the market. A Stanley #4 cost about $60, while a Clifton #4 sold for $300. I could not believe that people would actually spend hundreds of dollars for four pounds of iron with handles.d How could the price range be so wide? Some vendors shy away from discussing quality and price, probably as a timewaster. They say that plane quality and price are a “personal preference”, which is insolent and patronizing, unless there are actually people out there who enjoy slicing half way through a cut before the plane stops dead or starts to chatter. They could be correct if they refer to one’s willingness and ability to tune a cheaper plane. If so, it would be kindness to say so. I discovered that people collect old hand planes. So, for a while I thought that the high prices might have something to do with collectors values. I have learned better. Yes, collectibility does raise prices beyond some tools’ practical value. Some older tools are valued, however, because they are better than their newer replacements. Still, low priced and high priced planes are often worlds apart in quality as well as price. That does not mean that expensive planes are necessarily worth the money to me or to any other woodworker. That depends on a lot of factors, including your budget (what d These numbers refer to original Stanley/Bailey model numbers. Basically, #1 - #8 are bench planes, from 5½" long to 24" long; called smoothing, jack, fore and jointer planes as the soles get longer. Block planes have numbers all over the place. See Garrett Hack, Patrick Leach in references. you can afford to spend and what else you would rather spend the money for) and your alternatives. If you know what a good plane is supposed to feel like, and have developed passable technique, you can buy cheaper planes and learn how to make them work correctly. David Savage pointed out in one of his essays that a cabinet maker in the hand tool era would expect to pay a week’s wages for a plane. That does not justify a hobbyist’s paying the prices charged for high-end planes but it certainly changed my perspective. Two Plane Stories This short digression will illustrate two possible outcomes of buying a low-cost hand plane. Partially out of curiosity, I bought an inexpensive #5 Anant plane. I am happy with it, after replacing the iron with a thicker iron and flattening the sole (it was almost flat out of the box). The sides are square to the sole, so I can use it for shooting. I bought another one as a gift, and it just required light flattening of the sole and sharpening of the stock blade to work very well. I also wanted a scraper plane but did not want to spend the price of a LN scraper. I brought home a much less expensive Kunz #112 scraper. It chattered terribly even after the sole was dead flat and the blade was as sharp as a razor (and burnished). I bought a thick, new Krenov style blade for it. It was actually a bit too thick to fit, so I had to file down the blade holder until it fit. Then I could cut thin shavings without chattering, but only if I held it just right. Next, I posted a question about this problem on Woodnet. Among the many wrong answers were two right answers that lead me to Peter Korn’s book on hand tools. I modified some washers to take out some side156 Table of Contents Notes and Reflections While Shaving Wood to-side sloppiness along the shaft, and the chattering went away. The plane still did not cut correctly, however, despite my most exacting efforts to sharpen it and to turn a hook with a burnisher. Next, I bought a jointer jig for a file, and jointed the scraper blade so that it was flat along its edge and perfectly 45E. The scraper plane became barely useful. It still did not cut as smoothly as a Veritas cabinet scraper, however. Finally, I found an article by Chris Gochnour that explained how to set up a scraper plane. Now I can trust it. The final step was setting the angle of the frog correctly for the hook on the iron. I had been using much too steep an angle, which made the cut much to aggressive. That was a long road, and I still may not have reached the end. Since I have never used a high-quality scraper plane, I still don’t know whether one would have been worth the price to me. If a high-priced scraper plane works correctly out of the box, I can see why people would pay the premium price, but even an expensive scraper plane won’t work well if you don’t know how to use it. I find some consensus that Veritas planes are as good or better than LN. Veritas are usually a bit less expensive than LN or Clifton, but still way above most other widely-available planes. (Many hand tool woodworkers insist that most modern American Stanleys should be avoided by a cabinet maker.) Tuning a Hand Plane Tuning (fettling) a plane can turn a plane that barely works, and is quite unpleasant to use, into a quite decent plane. (See details below) The many articles describing how to fettle a plane are, so to speak, structural definitions of 27-3 a useful plane. In terms of operations, if a plane can cut very thin (.001 - .002"), continuous, long shavings it will be useful and pleasant to use. There seems to be one more issue. According to David Savage, The casting on modern planes tend to be so green that you can spend half a day getting the plane dead flat only to find it continues moving and is no longer flat in a few weeks time - so you are back to square one. So the alloy, casting and annealing are important in a new plane. In an older plane, this presumably would have settled down, but you can’t tell what will happen if you remove very much steel from the sole. A Practical Set of Hand Planes I didn’t have ready access to a thickness planer until recently, and I still prepare some rough lumber partially by hand. Of course, I flatten wide panels with hand tools. (I bought the planer when I needed to surface a large quantity of rough-sawn oak in the middle of a hot Florida summer.) If the board is not too wide, you can surface two adjoining edges on a jointer, set for as thin a cut as possible. Then scribe the desired thickness all around with a mortise gauge and handplane the remaining two surfaces. You can cut the stock to approximate dimension on a table saw and then finish surfacing with a few light cuts. If you need to remove more than c" from the flat surface of a board, you can resaw it on a band saw (about 1/16" oversize) and then smooth it with the planer or jointer and follow up with hand planes. I prefer to use planes for finish smoothing. Here is my list of necessary hand 157 Table of Contents 27-4 Hand Planes planes, followed by a wish list. This was influenced by a similar list by Robert Feeser (see references). Also, Robert Savage wrote articles for Good Woodworking magazine giving lists of recommended tools, including handplanes, for his fine woodworking students. The lists seem appropriately frugal and wise. These articles are not listed on the index page of his website but are available as I write this in early 2016. (See References) THE SHORT LIST I think that every woodworker should have three hand planes and a shooting board. The planes are: a good #4 or #5 bench plane, the best low angle block plane you can afford, such as the Veritas, and a shoulder plane. Beyond that, my advice on buying hand plane types is to choose based on what operations you need to do. A DETAILED LIST For Flattening and Smoothing Faces 1. A scrub plane is the first step in preparing rough lumber or for leveling glued-up panels. A scrub plane is one of the simplest planes and should not be very expensive. It cuts deeply with a highly cambered iron. Adjust it carefully because it is easy to cut much too deeply with a scrub plane. A second-hand smoother or jack plane can be converted to a scrubs by heavily cambering (rounding) the plane iron and opening up the mouth of the plane, but this is not ideal. Cut diagonally across the grain to get rid of coarse saw marks, take off the highest spots, find defects, etc. If the board is still too thick, repeat the process on the opposite diagonal, i.e., 90E from the first. Normally you would set the iron to cut about the thickness of a piece of typing paper or a little less. Make sure to cut with the direction the grain is rising, or the scrub plane will tear out chunks of wood that will be difficult to repair. In flatsawn lumber, the rising grain causes “cathedrals” that point in the direction that the plane should go. A scrub plane will leave you with a flat surface with small ripples or ridges. Even if you buy S4S lumber, a scrub plane is useful for flattening the surfaces of glued-up panels if the boards aren’t quite even. 2. Next, use a #5 jack plane. Depending on progress in step 1 and the type of wood you are working, you may use it diagonally or lengthwise. Finish this step with the grain with about a .002 to .003 thick shaving. At this point, the board is looking pretty smooth. With figured wood, including many pieces of oak, tear-out becomes a problem as soon as the ridges are gone and the surface is flat. See the section on tear-out below. 3. Some people recommend using a #6 fore plane next, for flattening faces and getting rid of the coarse plane marks left by the jack plane. I don’t have a #6 plane. 4. If the board is very long, use a jointer plane like the Veritas #7 low angle jointer to make it flat. 5. Use a smoothing plane like a #3 or #4 bench plane for final smoothing of faces. I prefer a Veritas low-angle smoother This is my primary smoother. It is an excellent tool, and has reduced the amount of rough sanding I do to nearly zero. The edge must be kept extremely sharp. One great advantage of a bevel-up plane is that the angle of attack can be changed by changing the plane iron. The Veritas low-angle bevel-up smoother has a bed angle of 12E. Irons 158 Table of Contents Notes and Reflections While Shaving Wood are available with 25E, 38E, and 50E bevels, giving an angle of attack from 37E to 62E. If I see any tear-out or expect it, e.g., from oak, I replace the 38E bevel iron with a 50E bevel iron. Planing oak with the 50E iron is a real workout. I take very thin (0.001") shavings, use a tight (0.004" - 0.008") mouth, and lightly touch up the edge frequently to keep it as sharp as possible. I have not cambered the iron on any of my planes but the scrub, but I may someday round the edges of the iron on this one. A well-fettled #4 would also serve this purpose. 6. Use a cabinet Scraper, scraper plane, or a card scraper to finish smoothing the board, usually after the parts have been cut out or the piece has been assembled. The cabinet scraper is the easiest to use. Even after all my fiddling with the scraper plane (see above) I find the cabinet scraper preferable to the scraper plane. Toothed Irons and High Bevel Angles Luckily, the first woods I planed were straightgrained mahogany, walnut and poplar. I got a bit of a surprise when I began to plane oak. Unless the grain is straight and parallel to the surface you are planing, you run the risk of tearing out small chunks of wood instead of slicing it smoothly. That is, unless the plane just stops dead. (A power planer or jointer won’t stop dead but it will tear out chunks of wood.) Softwood is likely to split out large pieces. You have to keep the plane iron as sharp as possible, learn how to read the grain, and plane in the direction that the grain is rising. (See Working With Red Oak) Wood with figured or inter-locked grain usually cannot be planed “with the grain” because the grain rises and falls unpredictably. The solution to planing such wood is to use a toothed iron for wood removal and a high angle of attack for the smoothing step. Some authorities say that you may make as deep a cut as you want with a toothed iron. I prefer shallow cuts for two reasons: first, I have 27-5 had to work too hard to remove the grooves when I made deep cuts. Second, I tend to use hand planes for final work where shallow cuts are necessary. I make them diagonal to the grain and at right angles to each other. A bevel down bench plane normally has a 45E bed angle, which is also the angle of attack. Planing oak with a 62E angle of attack is quite a workout, which is why I prefer shallow cuts with the toothed iron. The highest angle of attack is obtained using a scraper plane, cabinet scraper, or card scraper. I have never had tear-out occur from using a scraper. (But I have no experience smoothing really difficult exotic woods.) In fact, I have removed quite a few tear outs with a scraper. Card Scraper. The best scraping tool for small surface areas, final leveling of panels and edge-banding, or removing dried squeeze-out is a card scraper. A card scraper is a very inexpensive and useful tool. Card scrapers were among the pleasant surprises of studying this subject. They are surprisingly under-rated or under-publicized tools. A Bahco carbide paint scraper, if sharp, also makes a useful wood scraper. Use it with the handle close to or on the surface. Scrapers make a much smoother surface on hardwoods than on softwoods, which may become pitted with tiny tear-outs. A flush plane is also useful for cleaning up squeeze-out and other small projections from a smooth surface without actually re-planing the surface. Careful use of a card scraper will also remove imperfections, drips, etc from a varnished surface. I also use my shoulder and small rabbet planes for cleaning squeeze-out but is somewhat of a mis-use since it messes up the plane. For Straightening and Squaring Edges 1. Inspect the edge and plane off any bumps, knots, and waney places. Measure the width of the board at both ends – it may not be rectanguIar. 159 Table of Contents 27-6 Hand Planes 2. Use a #5 or longer plane with a jointing fence. The fence makes it easier to obtain a square edge to a face. The jointing fence can be of two types. It can be a straight flat board clamped to the rear face of the workpiece which protrudes an inch or two to act as a guide for the side of the plane. Or, it can be a metal plate attached to the side of the plane, forming a right angle. This fence is kept pressed against the front face of the board. I could not resist buying a Veritas iron edge plane. Sometimes I use it instead of the #5 with fence to joint shorter boards. It will produce shavings as thin as I could want, in long beautiful curls. Using either of these planes takes some care. I understand that old-timers could make square edges without sissy things like fences. I can, if I have to, but it is so much easier to use a fence. In contrast, I also have to be careful to keep the fences registered against the face of the board. It is too, too easy for me to let the sole of the plane guide the angle instead of the face of the board. Trimming End Grain – Block Plane and Shooting Plane Block planes are primarily for trimming end grain. or occasional light edge planing, e.g., trimming edge banding. I use an old no-name high-angle block plane, a Stanley #63 low-angle and a Veritas low-angle block plane. Like the other Veritas planes, this one fits my hand and is a pleasure to use. The advantage of a low-angle block plane over a normal angle one is that it cuts end grain more smoothly, everything else being equal. An adjustable mouth plane is an advantage because a tight mouth reduces tear-out when planing with the grain. Although experts say that the size of the mouth does not matter when planing end grain, I keep the mouth about the thickness of two pieces of paper. If I get very thin shavings (rather than dust and chips) from end grain I know that the plane iron is very sharp. High-angle or low, and no matter how sharp the iron is, it is essential to have a backer board tight against the “exit” edge of the work before trimming end grain. Otherwise, the board may tear along the edge. At first, I just clamped a board in place. But then, I discovered shooting boards and I made myself one. It is now indispensable. Not only does it hold the work square horizontally and vertically (or at 45E) but it prevents the plane from tearing the board at the back. Shooting Plane Almost every crosscut made with a hand saw or a band saw needs at least one pass to true and smooth end grain. A wider hardwood board takes significant effort, and you soon become aware of how awkward it is to use a bench or block plane on its side against a shooting board. The body casting of a shooting plane has a running surface set at 90E to the sole. It is intended to be used with a shooting board. In normal use, the sole is vertical, against the shooting board and the running surface is horizontal. (It can also be used for jointing, when the sole is horizontal and the running surface runs along a vertical board clamped to the workpiece.) The Veritas shooting plane is a large plane, 16" long and 7¾ lb. The bed is lowangle and skewed. It slices end grain very well. It’s an expensive plane, especially for such specialized use, but I use it a lot. If I still used my table saw I might not have made this purchase. I had tried various make-do approaches. A low-angle block plane will serve as a shooting plane but lacks the mass needed for larger boards. I put an auxiliary tote (sometimes called a hot dog) on the side of my #4 Stanley bench plane. This made it easier to push while on its side. It has more mass than the block plane but the angle of attack is too high for easy planing of end 160 Table of Contents Notes and Reflections While Shaving Wood grain. I am happy that I bought the shooting plane. Shoulder Plane & Rabbet Plane A shoulder plane has an iron as wide as the sole, so it can cut right into the inside edge of a rabbet or the corner where a tenon shoulder meets the cheek. I cut tenons slightly (1/64") thick, and lightly sand them down with sandpaper. I also use a Stanley (UK) #90 rabbet plane that converts to a small chisel plane for trimming into corners and up against edges. Spokeshaves – These were a late addition. I bought them when I started making chairs. A concave spokeshave is very useful for rounding arrises and re-sizing round tenons. A flat or convex spokeshave is very useful for fairing concave surfaces like chair slats. Router Plane – A router plane is nearly indispensable for cleaning out the recess for an inlay. I cut most grooves and dadoes with a router, but sometimes the router plane is easier and more accurate. It is useful for cleaning & slightly deepening dados and grooves cut with a router. I bought one from LeeValley. To cut dadoes (across the grain) with a router plane, it is necessary to deeply score the layout lines (edges of the intended dado) with a knife and then chisel out the edges somewhat, from the center toward the edges. Derek Cohen recommends chiseling to the depth of the intended dado and then removing the center with the router plane. This is probably necessary when cutting plywood because the grain of each ply is at right angles to the adjoining plies. However, some plywood is made with abrasive glue and is best cut with a carbide blade (electric router). Molding Planes I use four simple, wooden 27-7 block planes. Two of them are a matched hollow and round pair for trimming ½" rule joints. They are very handy. Before I got them I made sanding blocks for this purpose. They worked OK but were slow. The third one has two radii for easing (rounding over) arrises (edges). The fourth one cuts a 45E bevel. Plow Plane. A plow plane is used to cut grooves and rabbets, for such things as inlet drawer and box bottoms and cabinet backs. A plow plane has a fence but no scoring knife. Therefore, to avoid chipout when cutting across the grain you should score the wood with a knife. Because of this, cutting grooves in plywood using a plow plane is difficult and time consuming and I prefer to use a router. See the paragraph above about router planes. Dovetail Plane – I inherited a funny-looking wooden plane from my father. It turned out to be a 6Edovetail plane, which I have used once to trim sliding dovetails. It was indispensable for that use, however. Dowel Plate – A dowel plate is a piece of tool steel with holes in it that will trim a rectangular stick into a precise dowel size. (I don’t know whether I should have included it here, but think about it – if it isn’t a plane then what is it?) Pullshave – I bought this to scoop out a chair seat. It is also useful for trimming and cleaning up any concave radius. It’s important to use it “downhill.” Cutting against the grain is quite difficult and is likely to tear the wood. A pullshave is often quite difficult to control, and I get my best results with rougher work. Beading Tool – What I mean here is a type of small concave or convex card scraper in a handle with a fence. I bought one because it 161 Table of Contents 27-8 Hand Planes seemed like a great idea for beading table legs. I have not needed it for that, yet. It tears the wood when used across the grain and so was not useful for decorating drawer fronts. Tenon Cutter – I bought a round tenon cutter for making chair rungs but I have not used it yet. (I’m not sure that this is a hand plane, but once I included the dowel plate I could not stop myself from including it here.) Getting Started -- Comments on Using Handplanes Most of what I have learned is summarized above in my list of planes. However, planing is personal. It really illustrates the esthetics of craftsmanship and tools. Some details of preparing and using handplanes may seem fussy and unimportant, unless you have tried and failed to become proficient. a. Sharp iron. Ideally, a plane iron should be sharp enough to shave hair off your arm with no pulling or slice a piece of newspaper hanging from your other hand. You can use a plane with an iron that is almost sharp enough but it will be a bit more difficult. See Chapter 24, Sharpening. b. Learn how to use a plane, or try out a new plane, on poplar or other fine-grained soft hardwood. Do not judge yourself or the plane on oak, hard maple, etc until you are experienced. c. If the grain of the wood is not parallel to the surface, plane in the direction of rising grain. If the grain runs up and down, the wood is challenging and d. e. f. g. h. i. perhaps should be scraped or sanded instead. At least, don’t try to learn on such wood. A bit of oil or wax on the sole of the plane makes a big difference.(Apply sparingly and rub off vigorously.) Start by retracting the iron so that it does not cut at all and then advance it by tiny degrees until it barely cuts. If a plane skips and digs, and is otherwise well-fettled, probably the iron needs to be sharpened or you are taking too thick a shaving. Learn to keep your fingers on the tote (handle) and knob of a bench plane. Keep your fingers over the body of a block plane. Otherwise, eventually you will jam a finger between the plane and a hold-down or other obstruction. At least, that’s how I learned this valuable tip. The main exception to the preceding rule is when using a jointing fence or edge plane – keep the fingers of your left hand against the fence and the side of the board. It is surprisingly easy to plane a bevel on the edge while using a fence that is supposed to keep the edge square. Remember that the plane is a guided knife blade. On difficult wood, try to cant the plane slightly, so that the iron can slice sideways as well as forward. (You still hold the plane flat on the board and 162 Table of Contents Notes and Reflections While Shaving Wood j. push it parallel to the grain direction.) Especially, when trimming edge banding or the ends of box joints, don’t push the plane forward as much as use it like a knife in a holder. Rest the sole on the surface that you are trying to match, and cut off the proud material with a sideways slicing action. How to Tune Up (Fettle) a Plane There are many articles describing how to tune up a plane. One of the most respected is the one by David Charlesworth, available from Fine Woodworking’s web site. Garrett Hack’s The Handplane Book is considered a classic and also describes the procedure. Much advice about planes is received wisdom. For example, how flat must the sole actually be? Some question the emphasis a dead-flat sole. See the article by Schortzcopft. The performance of the plane may depend on where the hollows are located on the sole, and what work you want to do with the plane. Certainly, the plane must not rock on its sole. The scratches (or polished areas as the case may be) from your lapping must show on the front, rear and both sides of the sole, as well as around the mouth. Hollows in the middle of the sole may not matter. Reportedly, some Japanese planes are made with hollows in the soles. It would reduce friction. Also, excessive lapping widens the mouth of the plane and can conceivably allow the iron to warp For example, the sole of my Stanley #4 is flat everywhere but around the mouth, where it hollows a bit. It still works well for stock removal and initial smoothing. It will take very fine shavings from smooth-grained 27-9 wood. The slight depression around the mouth, however, may increase the risk of tearout with unevenly-grained wood. How important is a perfectly seated frog? My no-name #3 plane has a sheet metal frog (!) and the original iron. All I did was flatten the sole somewhat. It cuts beautifully, shavings as thin as you like, as long as it’s sharp. My opinion at present is that a complete tune-up on a new plane often may not be necessary, especially endless work lapping the sole flat. I would rearrange the conventional order as follows. The following leaves out some details – it assumes that you have Charlesworth’s or another article available. a. First, check that the sides are square to the sole. If they are not, you can still use the plane to flatten or to smooth, but not to joint edges with a fence. You can’t use a shooting board or fence without some fiddling to adjust the effective angle between the sole and the fence (shooting board) to 90º. You can grind and lap one side square or send it back for a refund. b. It’s probably wise to hone and polish the end of the cap iron (chipbreaker) as described by Charlesworth, to keep shavings from getting under it. Re-install the cap iron a few thousandths back from the sharp end, if possible. c. Sharpen the iron and try out the plane. (That means about as sharp as a razor blade.) Start with the iron even with the sole, so that the plane does not cut. Advance the iron about a quarter-turn at a time until the plane just cuts. It the plane can cut paper-thin shavings without tearing the wood, the iron may be sharp enough. d. If the plane rocks at all on a flat surface, skips, or digs in and is impossible to adjust to cut thin shavings, flatten the sole or 163 Table of Contents Hand Planes 27-10 reject the plane. Flattening the sole is straightforward, but tedious. Most people recommend flattening the sole while the plane is assembled and the blade is fully retracted. 1. Find a dead flat surface. You can use a piece of heavy float glass, polycarbonate (Lexan) glued to a piece of good plywood (like Baltic birch), or a cutoff from a stone counter-top. The tables of your jointer or table saw are also probably dead flat but you might not want to wet them or gunk them up with swarf and glue. This is messy, so you may want to lay down a towel under the flat surface, if the surface is stiff enough. 2. Lay a half-sheet of 400 grit silicon carbide wet or dry sandpaper on the flat surface. It may stay in place with just a few drops of water underneath it. Wet the paper and push the sole of the plane across the sandpaper, or move it in a figure “8" about 10-20 times, being careful to keep the plane perfectly flat on the sandpaper. The scratches from the sandpaper will be brighter than the rest of the sole. If the sole is flat enough, the scratches will show all around the mouth of the plane, along the sides, at the front, and near the back. Starting with 400 grit is conservative. This will not remove much iron, but you can use it to inspect the flatness. If the test shows that the sole needs to be lapped, replace the 400 grit with a coarser grit, e.g., 220 grit. Repeat the lapping process until the coarser shiny scratches show all around the mouth of the plane, along the sides, at the front, and near the back. 3. The sole is now flat. You can go on to successively finer grits to polish the sole: 320, 400, etc. I usually do. After all that work, I want the sole to shine. 4. Wipe the plane carefully and apply a thin coating of non-silicone paste wax to the sole. e. If the plane still chatters at all, and the iron is not at least 3/16" thick, consider replacing it with a new one from Hock or an A-2 iron from Veritas. The plane iron (blade) should be as heavy as possible, e.g.,3/16-1/8" thick and as sharp as a razor. In most cases, the plane iron must be perfectly flat across all or most of its width. An exception is a scrub plane, which is cambered. (Some people very lightly camber or relieve the corners of the irons on smoothing planes, also.) It will also be nice to have a spare iron handy. (Check first to make sure that a thicker iron will fit in the plane.) f. If it still chatters with a thicker blade, after flattening the sole, then see to the fit of the plane iron to the frog, etc., as recommended in the references. This is a bit more tedious and more difficult to do well. That’s why I recommend it only after you have tried the other steps. Frequent, light sharpening (15 μ and finer paper or stone) and light application of silicone-free wax (e.g., Butcher’s) should keep a tuned plane working nicely. Wipe the wax on (plane iron retracted) and scrub it off with a rag. Some people use oil from beside their noses for the same purpose. If a welltuned plane starts to skip and dig, or if it makes dust instead of shavings, it probably needs to be sharpened. 164 Table of Contents Notes and Reflections While Shaving Wood 27-11 References a.html David Charlesworth: HandplaneTune-up Step-by-step instructions to get maximum performance from any plane. Fine Woodworking September-October 2004. http://www.taunton.com/finewoodworking/ ToolGuide/ToolGuidePDF.aspx?id=2933 Deneb Puchalski. Tiny Teeth Tame TearOut. Popular Woodworking #178 October, 2009. Robert Feeser was a contributor to the hand tool forums on http://www.woodnet.net/. He wrote extensive postings on woodnet in response to questions Some of my choices and remarks above were influenced by his writing. Garrett Hack: The Handplane Book is considered to be the classic compendium. Hack is, however, a bit of an afficianado and collector. He does not give enough attention to modern medium-priced planes. Chris Gouchnour. Beyond the Basic Scraper. Fine Woodworking p.70-75, Jan/Feb 2007 David Savage. A Cabinetmaker’s Bench Tools. Part 1, http://www.finefurnituremaker.com/publish ed_articles/gw_tools_part1.htm ______________. An Apprentice Cabinet Maker’s Tool Kit Part II. http://www.finefurnituremaker.com/publish ed_articles/gw_tools_part2.htm _______________. Saws and Sundries. http://www.finefurnituremaker.com/publish ed_articles/gw_tools_part3.htm Seymour Shortzcopft. Perfecting the Steel Plane. http://www.highlandwoodworking.com/inde x.asp?PageAction=Custom&ID=124 (A pdf version is available from the “Library” index.) Shooting Board Kirby, Ian. He's in Fine Fettle http://www.woodworkersjournal.com/index. cfm/ian Patrick A. Leach: The Superior Works: Patrick’s Blood and Gore. This is an encyclopedic listing of handplanes, mostly antique http://www.supertool.com/StanleyBG/stan0 http://www.fineboxes.com/ShootingBoard.h tm Speas, Ed. Shooting Board Aims for Accuracy. Fine Woodworking May/June, 1994. 165 Table of Contents Notes and Reflections While Shaving Wood Chapter 28 – Glues and Adhesives M highly stressful , unpleasant glue-ups. For assembling turning blanks, I prefer PVA II for its slightly shorter working time. That’s why I prefer Type I for complicated gluing operations. PVA glues (especially white glue and Type II) remain thermoplastic for a while after they set and can be used as hotmelt glue in veneering. All three types of PVA, hide glue, and polyurethane are stronger than the wood they join if applied properly. (See the article by Schofield) Liquid hide glue is an under-rated product. It has a long working time, you can clean it off with water, it does not clog sandpaper, and it is reversible. That is, you can soften it with hot water and re-glue the joint with fresh hide glue. Liquid hide glue is excellent for veneering, although it tends to soak through veneer and the long cure time can slow down a project if you are allowing each surface to set before veneering the next. I do not own a glue pot and have never used hot hide glue. Polyurethane is a preferable adhesive for materials other than wood. I also value its longer working time and water resistance, but I rarely use it on wood. As an adhesive for wood, polyurethane is not as strong as PVA. It is messy and does not clean up with water. (You need lots of acetone.) Cured polyurethane glue cannot be removed from fabric by any known means. It will not fill gaps between the mating surfaces, despite claims that it will.The foam that develops while polyurethane adhesive cures is weak. Only epoxy can fill gaps and still provide a strong bond. If a mortise and tenon joint is a bit loose, I mix up a bit of slow-set epoxy to use on that joint. (If it’s very loose, I remake it or shim it out using any kinds of adhesives may be used in woodworking: PVA Type II (e.g., Titebond II), PVA Type I (e.g., Titebond III), polyurethane, hide glue (hot or liquid), casein glue, fish glue, epoxy, cyanoacrylate “SuperGlue”, resorcinol-formaldehyde (or urea-formaldehyde) resin glue, hot-melt adhesive and contact cement such as a sanding disk adhesive. I use mainly PVA Type I or II and twopart epoxy for joinery. (And liquid hide glue occasionally.) The others have specialized uses, especially the cyanoacrylate and sanding disk adhesives. Technically, only hide glue is a glue, but this distinction does not seem to matter except to adhesive chemists.) For general joinery I prefer PVA I because of its slightly longer working time, especially if a glue-up is complicated and may take some minutes. It is water-proof and, like PVA II is often stronger than the wood itself (depending on how tight the joint is). The dried glue line is slightly darker than with PVA II. PVA II is great for some jobs but its working time is too short for others. The main advantage of short working time (“quickness”) and short clamp time is that I can recycle clamps, say, if I am making a set of doors or boxes. I can often resume work on a glue-up piece within a few hours, although this is always somewhat risky. A thin coat of PVA II is slippery for about 30 seconds, and then it starts to grab. This can be a major problem for joints with larger surface area. If the box joint, dado joint, etc. is not completely together in 30-60 seconds, I may have to use clamp pressure or a rubber mallet to move the joint. The quickness of PVA II is probably second only to lack of planning as a cause of 28-1 Table of Contents 167 28-2 Glues and Adhesives PVA and re-fit the joint.) Cyanoacrylate (CA) is the quickest to set (less than 5 minutes) and is great for tacking things in place. You can use an accelerant to speed curing to seconds. CAs are available in various viscosities (thin, medium, thick, and gel). Thin will penetrate cracks and is useful for hardening a divot or crack filled with sawdust. You can use it to harden spalted or punky wood. Medium and thick are better for gluing wood. CA glue cures by polymerization, but the cured glue is soluble in acetone, “debonder,” and other organic solvents. CA glue joints tend to be brittle and to be weaker in resisting shearing than tension. CA fumes when it is curing and the fumes can be very irritating, so it should be used for smaller jobs in a well-ventilated space. Both polyurethane and CA glues polymerize in the presence of moisture. Their shelf life after the container has been opened is not long in a humid environment. For that reason, I buy small amounts and keep them in a plastic bag with a silica gel dessicant. Also, the purity of CA glue may affect its shelf life. For a comprehensive comparison of glue types, and to expand your knowledge beyond these few, see the article by Young. Spreading Glue It is essential to apply just the right amount of glue and to remove the squeeze-out correctly. I tend to apply glue liberally and sometimes get significant squeeze-out unless I have been especially careful. If a glue-up is going slowly (my usual case) it is difficult to patiently apply just the right amount of glue to the next surface as I go along. That’s why I prefer to use glue with a longer working time. Squirting glue out of the container on to the work tends to apply too much glue, unevenly. My best procedure is to put some glue in a bowl or on a plate and apply the glue like paint with a stiff glue brush. "Acid" or "flux" brushes work well and are inexpensive. They just about fit into a ¼" or larger groove or mortise. A flat stick like a popsicle stick works well. For larger surfaces, I use a small "trim" sized painting pad or a small metal putty knife/ spatula. Glue brushes can be cleaned with hot water and re-used. Plastic glue brushes remain useful indefinitely in a jar filled with highly diluted white vinegar in water. Flux brushes eventually corrode, but I can get many uses out of them. It is essential to spread the glue evenly and to work the glue under any dust, etc. that might remain on the surface even after wiping. For face or edge gluing, you can rub the two surfaces back and forth to spread the glue evenly, otherwise, brush it out well. I do not use my finger to spread glue because I will end up smearing it on the work or wiping it on my clothing. Polyurethane glue is really terrible in this respect. The mess it makes reminds me of the tar baby in the Uncle Remus story, and it is difficult to remove from your skin before it dries, even with acetone. After it has cured, nothing I have found will even soften it. I have permanent glue stain on too many work shirts from polyurethane glue. I now try always to wear an apron when using glue, especially polyurethane. Experience is useful, because you need a sense of how heavy a coat of glue to use. You are aiming for a tiny bit of squeeze-out. You cannot be sure that you have applied enough glue unless some squeezes out under moderate clamping pressure. A few beads squeezing out here and there along the joint after you clamp it is a very favorable sign, if you had applied the glue evenly. 168 Table of Contents Notes and Reflections While Shaving Wood Clamping Glue should entirely cover both mating surfaces and should be .0004" thick or less. A thicker PVA or polyurethane glue film is weaker. With two exceptions, glues do not fill gaps. The exceptions are slow-setting two-part epoxy adhesive and a thin putty made of wood flour (very fine sanding dust) mixed with PVA. Clamp pressure is needed to force glue into the wood, to overcome swelling caused by the water in the glue, to hold the surfaces tight against each other and to immobilize them while the glue dries. According to the Titebond.com website: Appropriate clamp pressure is essential for a successful bond. We recommend the following guidelines: For softwoods (pine, poplar): 100-150 psi; for medium density woods (cherry, soft maple) 150-200 psi; for hardwoods (oak, birch): 200-300 psi. Clamp time is dependent on wood species, moisture content and environmental conditions. For most of our wood glues, we recommend clamping an unstressed joint for a minimum of 30 minutes to one hour. Stressed joints, such as bent laminations, need to be clamped for 24 hours. Do not stress the new joint until totally cured - at least 24 hours. For polyurethane glues, clamping should be at least 4 hours depending on moisture levels. 28-3 According to Roman Rabiej, in addition to the wood species, appropriate clamping pressure depends on whether the mating surfaces are radial (quartersawn) or tangential (flatsawn). According to his work, a flatsawn oak butt joint requires a minimum of 900 psi for a strong glue joint. A quartersawn glue face would require about half as much pressure. Softwood would require somewhat lower pressures, about 300 for quartersawn glue face and 150 for a flatsawn glue face. The higher pressures recommended by Rabiej would be difficult or impossible to achieve in a home workshop. For example, consider a panel glueup of two 48" x 7/8" oak boards with quarter-sawn show faces (flatsawn glue faces). Each glue joint has 48" x f” = 42 sq. in. of glue surface, needing a total force of 37,800 pounds. A pipe clamp, operated by a strong man, can exert about 1400 pounds of force. Therefore, to receive the recommended clamp pressure, this panel requires about 27 pipe clamps along its length, that is, one clamp every 1¾". I do not doubt the accuracy of Dr. Rabiej’s engineering studies. Perhaps his recommended clamping pressures would be necessary to make a cantilevered panel or one that would be supported only along its edges. On the other hand consider joints that are not clamped, such as mortise and tenon joints. I doubt that the pressure exerted on tenon cheeks by the walls of the mortise are anything near the pressures recommended by Rabiej. Also set against this are years of experience by many respected cabinet makers. Evidently, for a panel such as a table top that is supported across its width by an apron, joints made with much less pressure seem strong enough. 169 Table of Contents 28-4 Glues and Adhesives Removing Squeeze-Out No cured glue accepts stain well. It reduces penetration or adhesion of dyes and stains. Gel stain will color the surface over glue, or in a pinch you can paint on oil stain if you are willing to let it dry and then seal it with shellac. Neither is likely to be a successful approach. Wiping PVA (or polyurethane) glue while it is still wet is not effective for me. When I wipe wet glue it just smears the glue around a larger surface area. If I give in to the urge to wipe it while wet, then I should scrub the area clean with water and white vinegar (for PVA) or acetone (for polyurethane) on a toothbrush or synthetic steel wool pad. It is better to wait about 30-60 minutes, until the glue has dried somewhat (PVA will have darkened and polyurethane will have turned into a spongy yellow foam). Come back with a sharp putty knife and card scraper and scrape off the squeeze-out. Keep the card scraper clean with a paper towel moistened with water, white vinegar, or acetone. I find that a card scraper is best but you can carefully use a chisel, the corner of a sharp paint scraper, etc. It is easy to scratch the wood if you are not careful. If you do it right, that will get most of the squeeze-out. If you open up any pockets of wet glue, don't fuss with it, or you will smear any remaining liquid glue around. Let the rest of it dry a few hours (overnight is too long) and then come back with a sharp card scraper for a final cleaning. By the way, if you don't have a card scraper, get some. They are cheap and one of the most useful and under-rated finishing tools. I have tried pre-finishing with Seal Coat shellac (blue tape to keep the shellac away from the glue surfaces). This is reasonable if you are not planning to stain or if the shellac is being used as a wash coat. If you want to use oil stain, it is essential -- but difficult -- to keep the stain off the glue surfaces. Michael Fortune recommends using wax to protect surfaces from squeezeout. I fear that I could not control this and the wax would interfere with final finishing. This approach deserves more study. Re-Gluing It is useful to know whether a joint was originally glued with hide glue. Hide glue is reversible, the only reversible glue used in furniture. (Some thermoplastic glues are reversible in the sense that the joint can be pulled apart if heated.) You can soften hide glue with a bit of warm water, acidified with a few drops of white vinegar, and re-glue it with fresh hide glue. Others have written that the softened glue will re-bond without fresh glue, but I have not tried that. If a piece was originally glued long ago, say before the 1940's, chances are that it was glued with hide glue or casein glue. Dried hide glue is brownish tan in color, and brittle. Other, newer glues are likely to be lighter orange-tan and resilient. If the dried glue on a surface will soften all the way through after an hour or so of dampening, chances are it is hide glue. If the piece is more recent, and if dampening will not soften the glue, you have to remove all of the old glue if you want to repair the joint permanently. Usually, water and vinegar with a steel brush and very judicious scraping with a sharp chisel will remove old PVA. It takes patience. Just regluing old PVA may seem to work for a while, but the joint will soon fail as the PVA cures. I have, in desperation, re-glued old PVA with 170 Table of Contents epoxy or polyurethane glue. That has lasted for years if the joint is otherwise completely sound. A joint that failed had usually loosened somewhat or failed structurally in some other way. Inspect it for cracks and other signs of mechanical failure. Removing all the glue will inevitably loosen the joint even more. A piece of hard cotton or linen fabric (like a piece of bedsheet) soaked in glue and wrapped around the male part of the joint will often re-tighten the joint. References R. Bruce Hoadley. Understanding Wood: a craftsman's guide to wood technology. Newtown, CT. Taunton Press 2000. Roman Rabiej. Get Serious About Clamping: Most Woodworkers Are Underclamping Their Joints. Fine Woodworking Nov/Dec 2007. M Schofield . How Strong is Your Glue? Fine Woodworking, July-.Aug 2007. W. T. Young. A Working Guide to Glues. Fine Woodworking, Jan-Feb 1999. Glue Titebond III Ultimate Titebond II Premium Titebond II Dark Titebond Original Bond Strength 4000 psi 3750 psi 3600 psi Chalk Temp 45°F 55°F 50°F Shelf Life 2 years 2 years 2 years Adhesive Proprietary Polymer Cross-linking PVA Aliphatic Resin Cured Color Light Brown Yellow (brown for Dark) Yellow Water Resistance ANSI/HPVA Type I (waterproof) ANSI/HPVA Type II (water resistant) indoor/dry only 28-5 Table of Contents 172 Table of Contents Notes and Reflections While Shaving Wood Chapter 29 – Abrading Wood – Rasps and Sandpaper A sandpaper. brasion is an alternative to cutting. Hand planes, scrapers and chisels cut the surface. They make shavings and leave a smooth surface. Abrasives grind the surface. They make chips or dust and leave scratches, although the scratches may be too fine to see with the naked eye and may easily be filled in by the finish. Wood rasps and files occupy a kind of middle ground. They cut, but do not necessarily leave a smooth surface. Sandpaper Sandpaper (sometimes called glasspaper) is a heavy paper or cloth with abrasive material attached to its surface. Sandpaper is used to remove small amounts of material from surfaces, for example, to shape, sharpen and smooth; to remove a layer of material (such as old paint); and to prepare the surface for finishing or gluing. Sandpaper is used by hand or by machine. You should always support sandpaper with a sanding block (or stick). Use a sanding block if the surface is flat, like a table top (or the edge of a board); otherwise you may round off the edges inadvertently and cause hills and valleys. Any machine that can move sandpaper in a regular motion – back and forth or round and round – can become a sanding machine. The simplest are rubber sanding disks chucked into an electric drill, a drum sander on a drill press, a rigid sanding disk on a lathe. Rotary sanders leave circular marks on the work that have to be sanded out by hand. Dedicated sanders, especially randomorbit sanders (ROS) and vibrating detail sanders are much more useful than disk sanders for finish work. Hand-held belt sanders are too aggressive and hard to control for furniture. (I learned to use hand planes so that I would not be tempted to use a belt sander.) Before you start sanding, give serious attention to dust control. Wear a dust mask and, if you have one, operate your shop vac or air cleaner. Wood Rasps and Files A wood rasp is like a file, but the shape of its teeth is different than a file’s. Rasps are meant to be used on wood, plastic, etc., Rasps remove material more quickly than files, but leave a rougher finish. Power rasps are available, e.g., rotary rasps for use in a hand drill or drill press. I rarely use them, and when I do, I use very fine rasps. The idea of putting my fingers that close to a spinning, unprotected cutter is not appealing. Hand rasps have small, sharp teeth cut into the surface. They can smooth and shape wood with a great degree of control. A Stanley Surform® is a hand rasp with cutters made of sheet metal. Many shapes and sizes are available. Some are called rasps and some are called planes. They are mainly useful, limited to rough work on soft woods. I think of them more as a carpentry tool than a cabinet making tool. Hand rasps are graded for coarseness of cut. Bastard Cut (cut 1) is the coarsest, with 60 teeth per inch (tpi). Second Cut has 100 tpi. Smooth Cut (cut 3) has 140 tpi. Patternmaker's Cut (cut 5) has 210 tpi. For a smoother finish than that, you can use a cabinet file, smoothing plane, cabinet scraper, or 29-1 Table of Contents 173 Abrading Wood – Rasps and Sandpaper RUNNING THE NUMBERS ABRASIVE MATERIALS Wood that has been planed with a power planer or jointer, either by you or at the mill may have ripples that need to be smoothed before finishing. Planers and jointers are not finishing tools. A smoothing plane, cabinet scraper or card scraper, on the other hand, is an excellent finishing tool and may reduce your use of sandpaper to almost nil. Sanding should be done in stages, so that each grit removes the scratches left by the previous grit and leaves successively finer scratches in the wood. It is often tempting to skip a grit or two, but that is usually self-defeating because you just have to sand longer with a fine grit to remove the deeper scratches. To shape wood, I might resort to hand sanding with 60 grit sandpaper, but normally 80 grit is the coarsest paper I would use. If I were sanding with a machine, I would be very cautious with 80 grit. I would follow it with 100, 120, 150, 180 and (possibly) 220 or 240. For fine work that will be stained or dyed, I hand sand the final grit with the grain to remove the little loops and swirls left by the ROS. Often I will not sand a table top, but finish smoothing it with a card scraper. As a general rule, I do not sand hardwood plywood with a machine. The sheets are flat and smooth as received, and light hand sanding with 180 grit should suffice. It is very easy to sand through the surface ply with an electric sander, which creates an appearance defect that is hell to correct. If the surface has a drop of glue or some other surface defect, I reach for a sharp card scraper. Sandpaper is available in many varieties, of which the following is a brief survey. Aluminum Oxide (ALO) Table of Contents 29-2 Aluminum oxide has largely replaced garnet as the abrasive of choice. ALO is hard and tough (resists fracturing), so it is long lasting. It can be used on wood, ferrous and non-ferrous metals and solid surface materials. Finer grits can be used in sharpening, e.g., sandpaper sharpening. Alumina-zirconia Aluminum oxide - zirconium oxide alloy is a very fine, dense, man-made crystalline grain which can be used for aggressive stock removal. It is often used for machine grinding applications. Very hard and sharp grained, it works well on stainless steel, titanium and other hard steels. It can also be used on wood. Zirconium is a very dense material with a unique self sharpening characteristic, which gives it long life on heavy stock removal operations. Iron Oxide Crocus Cloth is a very fine grit iron-oxide coated abrasive on a cloth backing. It is used to polish after coarser emery or aluminum oxide. Normally used only on metals. Chromium oxide Used in extremely fine micron grit (micron level) papers Emery Emery cloth is a piece of fabric with fine metallic or synthetic particles bonded to it. These particles are normally a variety of corundum that contains aluminum oxide set in iron oxide. Wet and dry emery paper is similar to emery cloth, but the backing is a form of paper that remains intact when wet. Good for rust removal, polishing, deburring and corrosion and paint removal. Garnet A naturally occurring mineral, which was once the choice of woodworkers. Not as hard or long lasting as man made abrasives. It is relatively sharp, but has a very weak bonding structure so it wears out quickly. It is inconsistent when compared to synthetics. Garnet dulls too quickly to be used in metalworking. Garnet paper, especially the cheaper brands, may leave fine particles of grit in the wood, which are not always easy to remove afterwards. Notes and Reflections While Shaving Wood SiC waterproof Silicon carbide. Can be used wet or dry. A very hard, very sharp, man-made abrasive suited for glass, plastic, rubber, ceramics, solid surface materials, and nonferrous metals and for final wet sanding on varnished wood, creating a high gloss surface. A very friable grain, silicon carbide cuts faster under light pressure than other abrasives. SANDPAPER TYPES In addition to type of abrasive, various backings, densities and additives are available. Backing Cloth Weights JF - Light weight, very flexible X - Heavy, stiff J - Light weight, flexible Y - Very heavy, very stiff XF - Heavy, flexible Backing Paper Weights A - Very light weight, flexible D - Heavy weight, stiff B - Light weight, flexible E - Heavy weight, flexible C - Medium weight F - Very heavy weight, stiff Open vs Closed Coat Open coat means the grains are set apart from each other, with a surface coverage of about 60%. In situations where loading is likely (soft, non-ferrous materials, painted surfaces, wood, etc.) open coat will resist loading and clogging and extend the useful life of the abrasive. Closed coat means the abrasive grains are adjacent to each other with no space between. Most applications will benefit from closed coat material because it allows for more material removal. Stearated Metallic stearates are insoluble soaps that retard loading Table of Contents 29-3 (clogging) of the abrasive by soft, resinous woods. It may leave a residue that can interfere with bonding of water-based finishes and polyurethane varnish. Other Abrasives PUMICE AND ROTTENSTONE (RAW GRIT) Raw grit is often used as a rubbing compound in conjunction with steel wool or cloth. Like the finest steel wools, it is for rubbing down before and after the final finishing coat. The most common abrasives are pumice stone and rottenstone. Pumice comes in medium and fine (dull or satin finish) and rottenstone is very fine (semi-gloss to gloss finish). FIBER PADS Fiber pads are man made nylon or polyester wool, which is impregnated with grit. They are available in coarse, medium, fine and very fine. A grit free polishing pad is also available. ABRASIVE BLOCKS Abrasive blocks have an aluminum oxide paper bonded to either a soft foam pad or a stiffer Styrofoam block. They come in many grits and are particularly good for sanding moldings and curved surfaces. STEEL WOOL Steel wool comes in eight “grits” and is used for everything from removing paint and rust up to fine furniture polishing. The three coarsest grits (No's. 2, 3 & 4) are for removing varnish, paint, rust, etc. The No.1 can be used for applying wax and polish for hardwood and tile. 0 (fine) is for cleaning woodwork, painted surfaces and floors. 00 (Very Fine) is for metal polishing. 000 (Extra Fine) is for rubbing paints, shellacs and varnishes before the final coat. 0000 (Finest) is for the final rubbing of 175 Abrading Wood – Rasps and Sandpaper shellacs & varnishes. Particle Sizes The particle size of abrasives (the coarseness or fineness) is often denoted by an arbitrary mesh size or numbering system. It is increasingly common for fine sandpaper to be denoted directly by particle size. Mesh Size is sometimes used for grinding and sharpening stones. This system appears to parallel the FEPA P- system. Note that an 8000 grit waterstone is considered very fine but finer sandpapers are available Mesh Size Microns 50,000 1/2 14,000 1 8,000 3 3,000 6 1,200 15 600 30 325 45 270 60 180 80 100 150 29-4 Abrasive Grit vs. Particle Size Particle FEPA CAMI Particle Size Size P Grade Standard Inches Microns (Europe) (USA) Grits coarser than CAMI 180 are approximately the same size in both systems ---P1200 -P 1000 -P800 -P600 -P500 P400 -P360 -P320 P280 -P240 P220 P180 P150 P120 P100 P80 P60 P50 P40 P36 P30 P24 P20 -- 1200 1000 800 -600 -500 -400 -360 --320 -280 --240 -220 180 150 120 100 0.00026 0.00036 0.00048 0.00060 0.00062 0.00071 0.00077 0.00085 0.00092 0.00100 0.00112 0.00118 0.00137 0.00140 0.00158 0.00172 0.00180 0.00204 0.00209 0.00228 0.00257 0.00304 0.00363 0.00452 0.00550 6.5 9.2 12.2 15.3 16.0 18.3 19.7 21.8 23.6 25.75 28.8 30.2 35.0 36.0 40.5 44.0 46.2 52.5 53.5 58.5 66.0 78.0 93.0 116.0 141.0 80 60 50 40 36 30 24 20 16 0.0075 0.01045 0.01369 0.01669 0.02087 0.02488 0.02789 0.03530 0.05148 192.0 268.0 351.0 428.0 535.0 638.0 715.0 905.0 1320.0 176 Table of Contents Notes and Reflections While Shaving Wood Chapter 30 – Repairing Surface Defects prevent or reduce it by applying adhesive sparingly. I usually want to see some glue squeezeout, however, to judge the fit of the joint and the tightness of the clamps. Second best is to remove it when the glue has begun to set but is not yet hard. Third choice is to scrape it and plane it off after it had hardened. This is laborious and risks adding gouges if the plane or scraper slips. A single-edged razor blade is useful as a miniature scraper – not to cut the glue but to abrade it away. Very thin bits of glue left after scraping can be sanded out with light sanding. After that, all I can suggest is to try to tint the glue to match the stained and finished surface surrounding it. As I imagine my work, there are no mistakes. Each piece of wood is smooth and regular, or at least I can make it so with a bit of planing, scraping and sanding. Reality, however, frequently intrudes on my ideal. Perhaps a tool knocked loose a chip or splinter, a plane iron caught and left a shallow cut in a show surface, or there is a bit of tearout. These problems are an annoyance if I notice them before I have assembled the piece. They are worse than just annoying if I notice them after the piece is assembled and ready to finish. Of course, I want to repair or hide them. After all the work of making the piece, and my hopes for its perfection, I want it to look as good as possible. I cannot offer an encyclopedic list of solutions, but here are a few things that have worked for me over the years. Sanding out the mistake – scrubbing it out with sandpaper as it the sandpaper were an eraser on a pencil – is tempting, almost instinctual. It is not likely to be satisfactory, however. Unless you can sand the whole board evenly, the irregular surface will show up in the sheen after finishing unless the surface is as dull as mud. Suggestions The preliminary step is to see what the defect will look like. You can simulate the appearance of the defect as it might appear after finishing. Moisten it with mineral spirits or alcohol. Question whether you can actually improve its appearance. Whatever you do can make it worse. Maybe the best solution is to leave it alone. If you have a dent, try using a steam iron to raise it. Just use a medium hot iron and a damp pad to steam the wood in the dent. Steam the wood around the dent as little as possible. Sometimes the wood will swell up and the dent will disappear. If the dent or depression does not respond, it may be possible to fill it with a burn-in stick (shellac stick). Dents in a surface that is yet to be finished often Squeezeout Glue squeezeout is a special case. It is not a defect, exactly, but it is unsightly. As a general rule, tinting and finishing will make the squeezeout more visible. (Sometimes varnish on red oak will closely match the color of dried PVA.) The best solution for squeezeout is to 30-1 Table of Contents 177 30-2 Repairing Surface Defects should not be corrected until the finish has been applied. It is very difficult to match a color that you cannot yet see. The difficulty in waiting to repair the defect until the surface has been finished is that any filler you use must adhere tightly to the surface. Some fillers may not stick to some finishes like oil or polyurethane. Finally, you will have to level the repair with the rest of the surface. This is possible by the careful use of a sharp card scraper but very difficult to do well with sandpaper. If possible, try your repair on a piece of finished scrap to get the color. I have had some success with the clear (colorless) burn-in stick. You could use a colored stick if you knew exactly what color the finished product would have, or if a small dot of a slightly different color would not be unsightly. Remember that the fill and the immediate margin won’t take stain or dye like the rest of the surface. This is definitely a skill to practice on scrap before you try it on your already-damaged work piece. You may be able to partially fill the depression with wood filler and then bandage the surface with a thick shaving off of a board of the same species and grain, even the back of the board you are repairing. Glue it over the defect. Shellac and wood flour works well. (See below.) Finally, you can correct a surface defect with an inlay. Inlays like this are called Dutchmen. They are a classic way to repair wood structurally, but can also be used for cosmetic repairs. The Dutchman should match both the color and the grain of the board with the defect. The best source is a cutoff or scrap from the same board, or even the back of the board in some desperate cases. Select a flatsawn scrap that has a 2-in. to 3-in. section clear of any growth rings. The best thickness is c” to 3/16 ", but you can cut the recess to fit. A long diamond shape will be less obtrusive than a square or circular piece. Whether to take the patch from an area with or without grain is an artistic decision. Grain pattern may be very difficult to match, but grain direction must match. Draw in grain lines with a oil pencil of the appropriate color. Inlays can be done by hand, but a router is very convenient for cleaning out the recess. Router kits and templates are available. See the chapter on inlay work. Some people suggest making multiple Dutchmen of a contrasting wood species, e.g., bow-ties. Instead of trying to make an invisible patch, you make it a part of a design. I have not tried this. In unfinished wood, a chip or splinter that popped up from the surface often can be glued back in place. You have to clamp it or hold it in place until the glue sets. Cyanolacrylate glue (Crazy Glue) is often a good choice, especially with an accelerant. Sometimes you can use painters tape to clamp it in place, especially if it is on an edge. If you clamp or weight it be sure to put waxed paper between the caul (clamp block) and the workpiece. Miscellaneous small cracks and defects can be corrected with on-the-spot wood filler. Put a minimum of cyanoacrylate glue into the crack or other defect. (A drop may be too much) Thin or medium viscosity is probably best. You do not want any glue on the surrounding wood or standing proud of the defect. Sand the surrounding area with fine (e.g., 180) sandpaper using a flat sanding block. 178 Table of Contents Notes and Reflections While Shaving Wood You are not trying to remove wood – you are trying to create fine wood flour to push into the defect. If necessary, spray a bit of accelerant on the defect, wait a moment until it flashes off and the glue hardens, and repeat. The following repairs are mainly for finished wood, or at least for wood that has been dyed or stained. A pit like a cigarette burn can be filled with a burn-in stick. Scrape the charred wood (if any) to provide a good base. It is not difficult to work the melted filler into the defect, but it is very difficult to get the surface smooth and even. You often wind up scraping or sanding the area around the original damage. Then you have to give the whole surface another coat of finish. Gunk to Glow recommends French polishing to fill pits like cigarette burns. I have done this successfully. You can eventually fill the pit, scratch, etc. with shellac this way, but after all the tedious padding the surface may not look right and you may have to rub the whole surface out with steel wool or pumice, etc. Use a fairly small pad and a lot of patience. I understand that clear (colorless) fingernail polish is useful for filling a pit, but I have not tried it. You can buy wax sticks to fill small holes, cracks, etc in finished wood. Plywood If you sanded through the top layer of plywood, you can carry out the following procedure after the surrounding wood is stained or dyed. However, do not apply stain or dye to the damaged part because the core will take too much color and appear too dark. If you get stain on the core, carefully sand it off. 30-3 1. Smooth the surface (it probably already is smooth if you sanded through the top veneer. 2. The wood grain of the top (show) ply may have 2-4 shades or colors in it. Use an artists pencil that matches the lightest color in the wood grain, and lay down a base coat. Scribble it on in the same direction as the grain and rub it with your finger or a blending pencil to smooth it out. 3. Draw some grain effects to match the surrounding grain using the next darker pencil 4. Use the darkest colored pencil to add some final detail. Minwax sells felt-tip pens with various colors of wood stain/dye that might be useful for final toning of the repair. 5. Mask the area of the repair and apply a few coats of spray shellac to seal in the repair. The surface is now ready for the final coat(s) of finish. Wood Filler Wood filler should be used after the wood is stained or dyed, and before the finish coats. Wood filler rarely, if ever, can be made to match the color of the surrounding wood unless you can paint or use colored pencils to create faux wood grain. Otherwise, it can make the defect more noticeable if the color of the filler bleeds into the surrounding wood. This will have the effect of making the defect like the bullseye of a target. Probably not what you want. (Also, the wood filler will seal the surrounding wood if it is unfinished, thus preventing it from taking 179 Table of Contents 30-4 Repairing Surface Defects stain or dye as expected. This is why I recommend that you mask the area around the defect and why I recommend it for wood that has already received its final color.) Before filling, mask off the area around the defect with masking tape. 1. Pour a tablespoonful or so of shellac into a small bowl. e (This will make about 2 Tbs of filler, but you can add more shellac later if you need to mix a larger batch.) Dewaxed shellac would be preferable but regular shellac is OK. 2. Mix in some fine (floury) sanding dust e.g., from a sander dust bag. (Sawdust is much too coarse) You can generate this dust by sanding with 120-150 grit paper on a power sander and collecting the dust in the bag. Some people stress that the wood flour come from the project wood or at least the same species. This may be important if you are patching bare wood. To patch stained or even varnished wood, I have found that a better color match is possible using different species, e.g., oak, mahogany. Add the wood flour a pinch at a time until you get a “meatball” texture, i.e., until it will hold a ball shape fairly well. Then, wearing rubber gloves, knead the mixture a bit with your fingers to make sure that it is uniformly the right consistency. If it's too thin, add e Some people substitute liquid hide glue or epoxy for shellac. I think shellac will tend to color the wood flour about as much as the finish will, and it sticks to wood and almost any finish. more wood flour. If it’s too dry initially, add shellac. 3. Working quickly, press the paste into the defect with a plastic spatula, and then use the edge of the spatula like a card scraper to level the patch. If you are patching bare wood, leave the patch a bit proud of the surface so that you can sand it down after it dries. Collect the excess back into the bowl. Scrape off the spatula. Keep it in a compact ball to retard drying. If the mixture starts to crumble, add a few drops of alcohol and knead it again until you get the pasty consistency. 4. Drying time depends on how deep the patch is. The patch will shrink somewhat as the alcohol evaporates, so don’t make too deep a patch at one time. 5. Sand and repeat if necessary. The patch may take stain or dye differently than the wood. It may be necessary to seal it with a bit of shellac and touch up the color to match the stained or dyed surroundings. You may need to draw faux woodgrain with wax pencils as described above. Do not use wood dye to color the mass of putty, because it will stain the surrounding wood and call attention to the patch (actually, it will make the problem much more noticeable). Latex based fillers offer fast drying times without the disadvantages associated with 180 Table of Contents Notes and Reflections While Shaving Wood lacquer and oil based fillers. Many commercial latex fillers can be stained or tinted. Some are available in pre-tinted colors. If the defect is deep, clean away any loose material. Apply the filler in several thin coats and allow each to cure completely before adding another layer. If you apply it in one thick coat it may skin over and dry very slowly. Then, as it dries it may shrink and crack in the center. After each coat has dried, sand it smooth and repeat until the original profile has been regained. Larger defects can be filled with a two-part epoxy glue. The epoxy will cure hard without the risk of cracking or shrinking and it can be scraped and sanded like wood. It will cure clear or translucent white, but it can be tinted with TransTint® or universal tinting colorant 30-5 (UTC) pigments. Please see the sections on toning and glazing in Chapter 32 for a discussion of tints. Conclusion As I reviewed this chapter, it seemed a bit “underwhelming,” even though it includes many tricks and ideas that I had to learn gradually, as I needed them, usually in a bit of a crisis. Having these ideas listed out is surely helpful. The inescapable fact, however, is that repairing surface defects, if done well, is very “crafty” and requires patience as well as skill. Although there are basic tools to learn, success will come from the creativity and taste of the craftsman. While you are learning, please practice on scrap. References Groz, George. From Gunk to Glow or The Gentle Art of Refinishing Antiques and Other Furniture. New York, NY. Random House, 1973. Marc Schofield How to Fix Flaws and Mistakes Fine Woodworking Sept/Oct 2012 181 Table of Contents 182 Table of Contents Notes and Reflections While Shaving Wood Chapter 31 – Signing Your Work L necessary to hold pigment, e.g.,a pencil line. Graphite paper (sold in craft stores) is similar to so-called “carbon” paper (probably that’s where carbon paper got its name). Carbon paper, however, leaves a waxy blue line, while graphite paper leaves a line much like a pencil does. The advantage of graphite paper over a pencil is that an image can be traced from a template, while pencil lines must usually be freehand or stenciled. Graphite lines from pencils or graphite paper will not fade. A graphite line can be coated with varnish or shellac to keep it from smudging or being rubbed off. Most “drugstore” marker pens are water or alcohol based. The latter will smear badly if shellacked. In any case, everyday marker pen images will fade. Graphic supply and hobby stores sell “archival” pens that contain India ink. India ink is made with finely ground carbon, usually lampblack or graphite. It will never fade. These are the proper kind of pens to use for writing on wood. Archival white inks are also available for marking darker woods. In my experience, archival inks, after they have dried, can be coated with varnish or shellac to protect them with no smearing. ike many other woodworkers, I like to identify my productions as my own. A signed work shows that it was made by a craftsman. A signature symbolizes the handwork that went into it, and expresses pride in the result. Maybe our work will someday become “collectible,” and appear on some future version of Antiques Roadshow. “Oh, a maple writing desk from the turn of the century, and it’s signed on the bottom – See?” That’s a nice fantasy that may never come true for most of us, but even if our work never becomes interesting to collectors, a signature will increase its value to family members and their descendants. It’s worthwhile to become familiar with different methods, so that you can adapt to various circumstances. By all means, try any new method on scrap until you are confident using it. A wellexecuted signature can add interest and appeal to a project. On the other hand, you would hate to mess up a successful project with an amateurish signature. However you sign your work, I recommend that you locate your signature in an unobtrusive place – the bottom or the back, or underneath a drawer bottom. It’s fine that you want to take credit for your work, but everybody else may not be as excited about it as you are. Engraving and Burning Writing on the Wood Engraving the wood represents a step up in difficulty, but the markings will be more permanent than marks on the surface of the wood. The two basic methods for engraving wood are mechanical engraving and wood burning. Mechanical engraving is most easily accomplished with a The easiest way to sign work is simply to sign it with a soft graphite pencil or marker pen. The surface should be sanded very smooth or, perhaps, coated with varnish or shellac and then sanded smooth. Very fine scratches from sandpaper are 31-1 Table of Contents 183 31-2 Signing Your Work have been created. This fills the engraved reciprocating engraver, a tool that lines with sawdust, which I then remove hammers a sharp point into the wood about with a stylus, ball point pen, etc. Then I 120 times a second, for example, a Dremel darken the engraved lines with a soft Engraver. Small routers can also be used pencil, or black India “archival” ink. The but are quite difficult to control in pencil can be soft graphite or an artist’s freehand work. “oil pencil”. On dark wood, e.g., a walnut Woodburning may produce a bowl bottom, I color the inside of the somewhat smoother line than engraving, engraved lines using or a fine-tipped white because the stylus can burn though the paint pen (Sharpie makes one). If harder latewood, which an engraver might necessary, I wipe the image again with tend to ride over. A good woodburning fine sandpaper and then apply a coat of tool is expensive, however. I find that lowshellac or varnish to protect it. Some cost soldering iron type irons are people prefer deeper engravings, which cumbersome and often do not get hot they then fill with a colored wax pencil. enough. I do not think they are worth the effort. Labels When engraving wood, I prefer to trace out the image with a pencil or For a more complicated image, e.g., a logo transfer it from a template with graphite or permanent inscription, a small label can paper. Then, I gradually engrave the lines, be printed with a computer. Probably, a starting with lighter pressure and ending laser printer would be best, since the with slightly heavier pressure (or higher image would be composed of iron oxide, temperature) once I’m confident that the but black ink-jet images may do as well. stylus won’t wander out of the groove I am The image can be printed on plastic trying to follow. transfers or on light card stock like an Woodworking supply companies index card. Paper labels can be attached sell “branding irons,” heated either with white glue or shellac. A paper label electrically or with a blowtorch. You can may not be very durable, but it can be order lettering or even custom logos. I protected with a few coats of shellac or hear that it is difficult to get the correct varnish. The transfers are usually attached temperature with a branding iron. with heat. Also, you can order engraved brass It’s possible that paper-backed coins that you can inlay or glue into a veneer could be printed in some computer piece. I admit that a custom engraved printers. Then, the veneer could be glued brass coin appeals to me, but I have not to the work. I have not tried this, but it used either of these methods, mainly seems like an interesting project for a because I prefer to place a hand-made “rainy day.” signature on my hand-made furniture. In other words, a factory can brand a logo What to Sign onto mass-produced furniture or make up a bunch of machine-made coins, but is I usually sign my normal signature and the unlikely to hand-sign a piece. date on the underside of a drawer bottom After the wood is engraved, I sand or table top, etc., never in an obvious the image lightly with 150 - 320 grit place. For smaller objects, e.g., turnings, sandpaper to remove any ridging that may Table of Contents 184 Notes and Reflections While Shaving Wood I put a stylized version of my initials and the year on the bottom. I draw the logo out on a piece of paper, re-enforce it with transparent tape, and use that as a template to transfer the image through graphite paper. Occasionally, I use a brief Table of Contents 31-3 dedication, e.g., “Custom designed and made for _____ by C.D. Hepler, 2011.” I would normally print a longer inscription like that on a paper label and then adhere it to the work as described above. 185 Table of Contents 186 Notes and Reflections While Shaving Wood Chapter 32 – Wood Finishing F need to strip off the old finish and follow the procedures for finishing bare wood. If your interest is in finishing a new (previously unfinished) piece, you may prefer to skip to the second section, Bare Wood. A note about safety before we begin. The solvents, removers, dyes, etc can be dangerous. Some are caustic, some are toxic, and some are inflammable. Some dyes may be carcinogenic. These risks can be controlled. Read and follow all package directions pertaining to safety, especially the directions about protecting your eyes, skin and respiratory system. The directions on the label will provide details about safe use that I may have omitted from this brief summary. inishing is the step when a project receives the last surface that we will give it before we deliver it to the client (customer, friend, or family member). The finish greatly affects the impression made by a piece. It is often a major criterion by which a client judges quality. For these reasons, I give as much attention to finishing as to any other step in production. I also find it fun (i.e., challenging and satisfying) to put a beautiful finish on a nice piece of furniture. Many woodworkers, however, dread this step and try to get it over with as quickly as possible. Some may feel intimidated, as if finishing were filled with mysteries of art and chemistry. Well, maybe so. Finishing can be confusing at first because there are many opinions on every aspect of it. Part of the confusion is because finishing is to some extent subjective, although there is an objective science beneath it. Also, finish manufacturers try to accommodate (or maybe exploit) peoples’ insecurities about finishing, and often conceal or misrepresent the nature of their products. Finishing does involve some skill, and therefore involves some risk of failure, especially as you are learning. But so does almost every other aspect of woodworking. This chapter describes a workable finishing or refinishing procedure for a home workshop that requires only brushing, wiping, or spray cans. It leads, if you like, to an old-fashioned brushed and rubbed shellac or varnish finish. It can lead to an attractive oil finish as well. We begin with the subject of restoring old finishes. If the restoration is successful, you would not need to read the rest of the chapter. If it is not, you may Table of Contents Previously Finished Wood -- Consider Restoration Some older pieces of furniture need to be stripped to bare wood and refinished, but some pieces have a damaged or worn finish that can be restored. I recall an upright piano that had a beautiful voice. It had been owned by a community playhouse, and was painted white. When I opened the top, I could see the original manufacture’s label and finish. It was a Ludwig piano. It had been made of quarter-sawn white oak, and the underside of the top showed a beautiful patina. Unfortunately, some esthetically challenged person had painted the piano, then it had been repainted many times – maybe ten times over the years. I found layers of paint in various colors. I could not scrape it all off. If I had tried to strip it with chemicals the pigment from the many hues of paint might have penetrated 32-1 187 Wood Finishing the wood and left a muddy pinkish color. I felt that I had to sand it smooth and repaint it. But if it had its original shellac finish, even if worn and beaten up, I would have taken the trouble to restore the finish. It would have been a very valuable antique. So, my advice regarding an older, previously-finished piece is to be conservative, especially if the piece has any possible antique value. Also, you may be able to avoid the mess and effort of stripping off an old finish. Perhaps you can restore it instead. The first step is cleaning the old finish, but before you do that you should find out the composition of the original finish. C OMPOSITION OF T HE O LD F INISH Widespread use of shellac as a furniture finish began in the early 1800's. From then through the 1920's and 30's shellac was the coating of choice for fine furniture. Then it was gradually replaced by lacquer and then catalyzed lacquer. Shellac and some older kinds of lacquer do not change chemically when they dry. Their original solvent will dissolve them again, even after many years. Varnish and catalyzed lacquer do change chemically when they dry, and cannot be dissolved by their original solvent. You can test the finish on an inconspicuous part, a location that is hidden from view, such as the top inside of a leg, or the inside of a table apron, or the back (if it is finished). Moisten the finish with a cotton swab wet with denatured alcohol. Be careful not to let the alcohol run down the surface to a more conspicuous place. If the finish softens, gets tacky, and begins to come off on the swab, the finish is shellac. If it does not, Table of Contents 32-2 try lacquer thinner. Lacquer thinner will soften both shellac and older types of lacquer. If lacquer thinner does not readily soften the old finish, then it is varnish, paint or some sort of catalyzed finish. C LEANING AND P OLISHING There are at least three alternatives to consider: 1. Cleaning with so-called miracle furniture restorer. Miracle furniture restorer is traditionally made of equal parts of boiled linseed oil (BLO) , mineral spirits (MS) and vinegar. This is way too much linseed oil for me. I recommend about 1 part oil to 6 parts turpentine or mineral spirits and vinegar. You can replace the vinegar with a ¼ part of Murphy’s Oil Soap (adding soap and vinegar makes a useless gunk). Wash the piece gently with a rag or fine plastic abrasive (Scotch-brite 7445 (white) or 7745 (tan)). Try not to get it really wet. Just sponge off the surface, wipe and rinse. Restoring with BLO is very tempting and looks good at first, but BLO can take a long time to dry. If you can apply it to a smooth surface and then vigorously polish it with a rag in a few days, it might be a good finish to use, and may hide alligator cracks on an old finish. With an old shellac finish, you can experiment by adding alcohol (½ part or more) to the restorer. This makes a somewhat dangerous but potentially useful mixture that may help old shellac 188 Notes and Reflections While Shaving Wood 2. 3. 4. finishes heal themselves because the alcohol softens the very top layer of shellac and the rubbing fills in the cracks a bit. Shake well and wash with a Scotch-brite pad or 0000 steel wool, then promptly wipe dry and buff. Cleaning with 0000 steel wool, either dry or dampened with mineral spirits. Polishing with rubbing compounds (for furniture or cars). Try Howard Restore-a-Shine. a A fourth method is somewhat heroic. It is for hardened old shellac finishes. It requires a certain finesse to restore the finish without removing much shellac. I would not try this unless I were willing and able to remove all of the shellac and refinish the piece if this failed. Shellac is soluble in alcohol, so an old shellac finish can be softened or removed with denatured alcohol. Hardened shellac dissolves very slowly, however. The idea is to lightly wash the shellac with denatured alcohol, removing or redistributing the top layer of shellac but not disturbing the rest of the finish. (To find out whether it is shellac, see the previous section.) Wash a small amount at a time with a Scotch-brite pad dampened with alcohol. Don’t let the alcohol stand on the shellac, and don’t rub very hard. A dark-colored, old, factory-made piece, e.g., a dark mahogany table, probably was finished with colored shellac. If you remove much finish, the spot will be lighter in color than the surrounding wood. If you wash it quickly and lightly and then let it dry, you may restore the finish. If cloudy, whitish streaks appear, quickly wipe them off with a clean pad dampened with alcohol. This may leave a very dull surface, which you can then wax or coat with dewaxed shellac. (Zinsser Seal Coat is premixed, dewaxed shellac in a 2 # cut). (See Shellac and Acrylic below.) If you are satisfied at this point and feel that you do not need to add new finish, you could wax it with a good silicone-free paste wax, e.g., Johnson’s or Butcher’s paste wax. Wax does not add much protection, despite claims to the contrary, but it does fill tiny scratches, etc. and helps to raise the gloss. If some sort of additional finishing will be necessary, be sure to clean the old finish first with 0000 steel wool rinsed in a bowl of MS. This is necessary because most furniture has been waxed and may have traces of silicone. Gently scrape off dirt, etc. If a chip of old finish has exposed bare wood of a much lighter color, tint some dewaxed shellac with Transtint and use a fine brush or Q-tip to repair it. You can buy felt-tip pens of touch-up stain from a hardware store. If the finish cleans up well but is mildly cracked, etc, there are two next steps. At this point, the finish should be dull. If not, scuff sand gently with very fine (600) sandpaper or 0000 steel wool. 1. a www.howardproducts.com Table of Contents 32-3 This is the choice that I recommend. Brush or wipe on 189 Wood Finishing dewaxed shellac like Zinsser Seal Coat right over the cleaned and lightly scuff-sanded old finish to fill small cracks, etc. The Seal Coat label claims that it will adhere to any clean substrate, and I have no reason to doubt this. Then, if desired, you can add additional coats of shellac or varnish as described below. Whether you pad or brush the dewaxed shellac, practice on scrap until you can apply a smooth coat in one stroke. The alcohol solvent may soften the old finish if you brush it or rub it too much. Also, the alcohol solvent may evaporate quickly. Let the shellac dry a few hours and decide whether to apply more finish (see below). If you are happy with it, give it a coat of paste wax. 2. If you prefer not to try #1, rub on a thin coat of dilute (1:6) boiled linseed oil in MS using a soft rag. The MS will evaporate in a few hours but the oil may take overnight to dry. Then buff well with a soft rag. This is often OK for items that will not be handled, e.g., cabinets, clocks, etc. but may leave tables or chairs a bit sticky in humid weather. If much of the old finish is badly discolored or is chipping off, then you may need to strip it. If you don’t need to strip it, then go to Toning or Top Coats. S TRIPPING If you have to strip the old finish, test it (as described above) to see whether it is shellac or lacquer. It may be much easier Table of Contents 32-4 to dissolve shellac with denatured alcohol (or lacquer with lacquer thinner) than to use paint remover. Many of the more effective paint removers are nasty, especially those containing methylene chloride. Follow the directions on the container carefully. Make sure of good ventilation. Wear substantial rubber gloves. Wet the finish, let it sit a moment and then scrape and scrub the finish off with plastic or fine steel wool, about 1 sq. ft. at a time. If you use a putty knife to scrape off the softened old finish, first round off the corners of the putty knife with a file to make it less likely that you will gouge the work. Make sure that the stripped wood is cleaned of residue. Scrub with 0000 steel wool and clean lacquer thinner until the bowl of lacquer thinner stays more or less clear. I have had problems with a residue after stripping, even though I washed it with lacquer thinner as a final step. You can see the surface kind of repel the finish. This is why I recommend sealing with dewaxed shellac before applying the finish coats. Bare Wood (New or Stripped) Learning how to finish well is really no different than learning other operations of woodworking. I suspect that people do not try to learn to finish well in part because they are told that it is messy and difficult and in part because it comes at the end of the project, where a mistake might seem to spoil everything. You learn to finish the same way you learn any other skill. Practice on scrap until you feel confident enough to try it on a piece of serious work. 190 Notes and Reflections While Shaving Wood 32-5 The basic steps in finishing bare wood are (1) preparing the surface, (2) staining or tinting the wood itself, (3) toning, (4) finishing and (5) final smoothing and polishing. Staining and toning can be omitted if you don’t wish to change the color. eventual frustration than any other step in finishing. Sometimes it is more difficult to correct surface problems after stain, dye or finish has been applied than it is when the wood is bare. P REPARING THE SURFACE If you are uncertain of your finishing skills or if you are in a hurry, a reasonable compromise is to color the wood with gel stain and then apply three coats of wiping varnish. Some people swear loyalty to Danish Oil, which is available in pale tan (natural) and colored versions. Wipe on, wipe off. Allow it to dry overnight, smooth with 0000 steel wool, wipe clean, and repeat for a total of three coats. After the oil has cured, apply paste wax as desired. Please see the discussion of Oil/Varnish blends below. New (previously unfinished) work should have been planed, scraped or sanded flat and smooth. It is often easier to do this before assembly. If you sand it, start with 100 or 120 grit and work up by degrees to 180 or 220 grit. (Usually 180 is fine enough.) Always sand with the grain. If the surface is not perfectly flat, at the end of the finishing process you will not be able to level the final coat. (You will still be able to rub it with steel wool or pumice.) You need a sanding block to accomplish this. I prefer a cork block. I avoid heavy sanding of previously finished wood after stripping, if possible. An age-darkened wood surface should not be disturbed if you can avoid it. Examine surfaces carefully, especially show surfaces like tops, for imperfections such as glue squeezeout, scratches, gouges, dents, plane tracks, etc. Dampen the surface with alcohol to make imperfection easier to see. The finish will not hide defects. Oil stain and dye can greatly magnify them. The wood should look as uniform as you want the finished wood to look. (This is not to argue against a distressed look, which may be more attractive than a new look.) See Chapter 30 for more on repairing defects Moving past this step too soon has probably caused me more extra work and Table of Contents A C OMPROMISE S TAINING /T INTING Staining, and especially dyeing, will certainly change the color of bare wood. You can imitate mahogany by staining birch, but this has some disadvantages. On new work, it is much better to choose a wood which naturally finishes to the color you want, or nearly so. On a stripped piece, it is better to aim for a tint that is not too far from the natural color of the wood (plus finish, of course). The names that manufacturers give to stains (especially) and dyes may have very little connection to the color of the real wood. Colors from the same manufacturer will differ between products. The same color of oil stain will produce different colors on different woods. It is, therefore, essential to try out stains on scrap samples taken from your 191 Wood Finishing project, or at least samples as close to possible to the wood used in the project. Don’t be afraid to blend or dilute stains or dyes with other colors of the same product. (Never try to mix stains and dyes together. They are incompatible.) Gel Stain Gel stain is a newer product than oil stains and wood dyes. Unlike the two older products, gel stain typically does not need to penetrate the surface to color it. It sits on the surface. Gel stain is very easy to use – wipe on, wipe off. It can be used over non-porous surfaces, especially fiberglass faux-wood or even previously sealed wood surfaces. In this respect it can be used as a toner (see below). Because it dries and is opaque in thicker coats, it can be used as a glaze. Most gel stains rely on pigment (opaque colored particles) and so tend to obscure grain rather than accentuate (“pop”) it. A gel stain is much less likely to reverse the grain of softwoods or blotch difficult woods like cherry and maple, but it is more likely to obscure delicate grain. Because it dries, gel stain is sometimes touted as a one-step finish. In my opinion it must be top-coated on surfaces like table tops and chair arms that will receive any wear. Otherwise, the color will disappear as the gel stain layer wears off. Grain Filler Varnish, etc. will not easily fill the pores of open pore woods like ash, oak and mahogany. Frankly, if you plan four coats of varnish on mahogany, you probably do not need grain filler, but even four coats of varnish may not fill the pores of oak. Table of Contents 32-6 If you want a very formal finish with no pores, you should use a paste grain filler, aka wood filler. This is an entirely different product than the wood filler used to fill nail holes, etc. I have had good results with Bartley’s Paste Wood Filler and with CrystaLac. Bartleys is an oil based varnish filled with silica. CrystaLac is water-based. You paint grain filler on with the grain, let it set, and wipe it off across the grain. It is usually better to start with 12" square sections at a time so that you can control drying time before removing excess filler. How long to let it set is a matter of judgement, usually 5-10 minutes for Bartleys. How long to let water-based fillers set depends on relative humidity. (Follow detailed directions on the can.) The objective is to let it dry enough that it will stay in the pores but can still be can be easily troweled off. A plastic squeegee or wide putty knife is very useful for this, or you can use burlap. Burlap will fill up with excess filler pretty fast. Wipe, trowel or squeegee across the grain. Beware of black rubber squeegees. They contain carbon black and can discolor the surface if any rubber wears off while you are removing the grain filler. Leave a very thin coat, as thin as possible without scrubbimg so much that you start to empty the pores. The idea is to leave the filler in the pores but not on the surface. It will dry like a varnish, but then you can lightly sand it off the surface. Avoid breathing the silica dust. Wear a mask. Depending on how much you sand it, grain filler may inhibit penetration of dye or oil stain. You can buy tinted grain 192 Notes and Reflections While Shaving Wood filler or a neutral color that you can tint yourself using UTC colors or oil stain. This is a messy procedure, and if you let the excess dry too much before scraping it off, it is difficult to remove. I usually limit grain filling to a show surface like a top, apron and drawer front. It can be difficult to wipe off of irregular surfaces, but a few pores might not be noticeable there, anyway. Oil Stain Oil stains, e.g., Minwax Wood Finish Penetrating Oil Stain®, are easier to use than dyes, and are more readily available in stores. They consist mainly of opaque pigment and transparent oil-soluble dye in a diluted drying oil like boiled linseed oil. Oil stain is not the best way to significantly darken light wood or to match lighter/darker boards. If you are trying to darken the color a lot, use wood dye (aqueous or alcoholic, non-grain-raising (NGR) dye with retarder as necessary. Oil stains are fine for popping (intensifying) the grain and for moderate color change on evenly grained bare wood like mahogany and walnut. Very dilute oil stain can be used for washcoating as well, see below. Full-strength oil stains have some significant disadvantages. They tend to “reverse” the grain on softwoods, because the lighter grain takes up more stain than the darker grain; to blotch on maple, cherry and birch; and to amplify the porosity of oak. They tend to obscure finer grain details. Before applying oil stain, I recommend that you pre-wash on softwoods, maple, cherry and birch. I would not use oil stain on bare oak, especially red Table of Contents 32-7 oak. (See Finishing in the Note on Working Red Oak.) Oil stains tend to dissolve (run) into oil-based varnish unless the oil stain was wiped off well after setting and allowed to cure well, overnight at least. Choose the color of oil stain (blend if desired with other oil stain, artists’ oil paints or pigments), wipe it on and scrub it off well with a clean rag after 5-15 minutes. Synthetic steel wool (white Scotch-brite) might be a good choice to apply the stain, as it will also scrub the wood clean. Let the oil stain dry overnight (minimum) and consider exposing the stained wood to warmth and light, e.g., consider putting it in sunlight for an hour or so. Check it often – don’t let it get too hot. After exposure to light and heat, polish the wood again with a clean rag to remove any oily residue that came to the surface. Don’t apply a second coat of oil stain. The pigment in the oil stain may obscure the grain and make the work look muddy. A thick coat of stain takes forever to dry, is likely to run under a subsequent coat of varnish, and may cause unevenness, poor adhesion, and long drying time in the varnish coat. For the cleanest look, seal oil stained wood with a coat of dewaxed shellac. Colored Varnish Colored varnish like Polyshades® is advertised as a one-step, easy finish. In my experience, it is too difficult for a beginner to apply it evenly as a finish coat, especially in darker shades. It will show lap marks, brush marks, etc more prominently than clear varnish. 193 Wood Finishing Colored varnish is very useful, however, if diluted and used as a toner. You can also expect satisfactory results if you use diluted, colored varnish as the first coat and then cover it with untinted varnish. (See the section on Toning, below.) Wood Dye Wood dye is transparent and can achieve greater color change than gel stains or oil stains, without obscuring the grain nearly as much. Examples are Transtint, Transfast or NGR dyes such as Behlen Solar-Lux. Be sure to wear rubber gloves (and an apron) when handling wood dyes. It is more difficult to get an even color with wood dye than with oil stain, but repeated applications will increase color and saturation (darken) so they are useful in matching boards. The best way to use NGR stain on bare wood is to mix a dilute mixture so you can soak the wood with it and then wipe it off before it dries. Otherwise, uneven colors, lap marks, etc. may be a problem. You can even out the color somewhat with a rag soaked in dye mixture. Let the dye dry completely before judging color match. The hue will change as it dries. You can darken it with a second application or lighten it by washing some of it off with alcohol. You can mix retarder into the diluted wood dye mixture to give it some body and slow the drying time. Don’t dye wood with tinted shellac because it is too difficult to make it come out even. It can run and/or dry from the edges, leaving dark trails – generally a mess. Any irregularity will show and may be hard to remove. Seal dyed wood with dilute or fullTable of Contents 32-8 strength oil-based sealer, e.g., varnish. It is difficult to apply shellac directly over dyed wood because the dye runs into the shellac. Washcoating Some woods, especially softwoods, poplar, cherry and maple, will blotch (stain unevenly) and require pre-treatment before pigment stain is applied. Washcoating (pre-stain conditioning) works by partially filling wood pores. The idea is that it will fill the more porous parts of the wood surface, the ones that would absorb more stain and appear darker, i.e., blotchy. Because washcoating reduces the effect of pigment stain, if you are trying to darken the wood significantly washcoating to reduce blotches will be a trade-off against being able to darken the wood as much as you desire. There are products on the market for washcoating, e.g., Minwax Wood Conditioner®. They are just very dilute drying oil or varnish. You can make your own. There are two main approaches to washcoating. One approach is to partially fill the pores with solvent and apply the stain before the solvent has dried or soaked in completely. The other approach is to partially fill the pores with a drying finish like dilute varnish or shellac. Clearly, the first approach works only as long as the solvent remains in the pores. The second approach works best after the washcoat has dried completely. The label instructions for Minwax wood conditioner use the first approach. I don't really know its composition, but it seems to be mainly solvent, with a bit of 194 Notes and Reflections While Shaving Wood oil/varnish, perhaps to retard drying. Flexner recommends letting this product dry overnight, despite the label instructions. So, it must contain enough oil/varnish mixture to seal the pores somewhat. I prefer the second approach, as it is more controllable and predictable. It does take a bit longer, but not too much. Use a 1 lb cut of dewaxed shellac, such as SealCoat® diluted 1:1 with denatured alcohol. Don’t use this under NGR dye. An alternative is a fast drying varnish or sanding sealer diluted 1:3 with a mixture of naphtha and MS (quicker) or just MS (slower). When I use sanding sealer as a wash coat, I pour the clear supernate off, leaving the white stuff in the can as much as possible. (The white stuff is zinc stearate or another soap to make it easy to sand. This is not necessary for a washcoat. Also, stearates may reduce adhesion of polyurethane varnish.) Flood it on, then wipe it off. It will usually feel hard dry in 2 hours. In some cases, e.g., for a dark stain, limed effect, etc. you can use diluted paint (oil or latex) as a washcoat. For example, try two tbs. of latex paint (of an appropriate color) in a pint of water. Lightly scuff sand, wipe off the dust, and apply the stain. The pores will show as white while the rest of the wood has its natural (or stained) color. Always try a test piece first! Popping the Grain. A variation on washcoating for figured wood like curly or fiddleback maple is to washcoat with a mildly tinted finish. Some people use boiled linseed oil (BLO), which Table of Contents 32-9 is amber, amber shellac or tinted shellac. (The tint should be mild, e.g., 3-5 drops of Transtint per ounce of dewaxed shellac or oil stain diluted about 1:10.) It should match or harmonize with the color of the final finish, usually an amber tone.) For example, I used the following mixture to washcoat/pop the grain on soft maple: BLO, 6 parts, MS 6 parts, and Minwax Colonial Maple oil stain, 1 part. I brushed it on, let it soak in, and wiped it off. After the washcoat was dry, I sanded the surface with 220 grit sandpaper, somewhat aggressively, re-exposing the wood. The BLO remained in the pores, darkening them. When I applied the finish coat, the wood figure was intensified. T ONING Toning is the application of a transparent colored layer to partially finished wood, i.e., wood with one or more base coats already on it. Toning differs from staining and dyeing, which is done on bare (unfinished) wood. It differs from glazing, which is application of opaque colorants to finished wood. Toning is used to correct color, match boards, etc. You can paint a faux plank effect or even subtle wood grain by toning long triangular shapes or long rectangular strips. Many media and colorants can be used for toning, as summarized in the following table. Universal Tinting Colorants (UTC) are finely ground pigments in a paste that are compatible with all four media. They are used by commercial painters and are widely available. Behlen offers “furniture 195 Wood Finishing Medium Colorant acrylic varnish (water-based) acrylic artist paints, Trans-tint® shellac Trans-tint resin varnish, including Polyshades® oil stain for plain varnish, choose tint for Polyshades lacquer UTC & proprietary tints powder” which are finely ground dry pigments. The tiny amount of pigment needed for toning should not opacify the medium to any perceptible degree. Some toning is possible with gel stain, but it must be applied in a very thin and even coat to provide a consistent and transparent layer. I prefer to tone with the first three approaches listed in the table. I don’t use lacquer. which is tinted dewaxed shellac, and have also had good results with tinted varnish. Toning uneven surfaces with a brush is tricky because runs will show up very distinctly as dark areas. You should consider applying toner with a rag or paint pad. If the wood was sealed with acrylic varnish, probably any of the three media should be satisfactory. If the wood was sealed, stained, washcoated, etc., with varnish, I would use tinted shellac for toning. Mix small amounts of Transtint® dye with the shellac to tone the wood color. Do not make it too dark, or lap and brush marks will show. I find about 1-4 drops of Transtint per ounce of shellac is about right. If the wood was sealed, stained, washcoated, etc., with dewaxed shellac, Table of Contents 32-10 then consider toning with dilute tinted polyurethane varnish, e.g., Polyshades® . Adherence of polyurethane varnish to dewaxed shellac is excellent. Adherence to normal (waxy) shellac is a bit unpredictable, although the labels of some newer polyurethanes do not warn against using them over normal shellac. Undiluted Polyshades is sometimes too dark for use in toning, in my opinion. I often dilute Polyshades with a “wipe-on” clear gloss polyurethane (or, equivalently, regular gloss polyurethane diluted about 1:1 with MS.) I prefer to use a “finishers color wheel” and classic colors such as raw and burnt umber and raw and burnt sienna. Tinted varnish may look cloudy in the jar, but still appear perfectly transparent on wood. Good brush technique is important to avoid uneven toning. Brush a thin portion of the edge first and then apply about 1 brush width or 1 board width with a full brush. Brush out toward the edge. Don’t drag a full brush onto an edge or it will run. Brush tinted finish out to the leading edge (inside) to feather it and then start the next brush load about a foot away from the feather edge. Brush into the feather edge and tip it off once. b You can also apply tinted wipe-on varnish with a lint-free rag. A second coat of tinted shellac may b I applied a second coat of tinted shellac, in streaks to simulate darker wood grain, immediately after the first. I feathered out the ends and tried to make sharp margins on the sides of the streaks. For some, I used masking tape laid exactly parallel to the edges to simulate separate planks. The result was fairly convincing, certainly more interesting than the very consistent color that resulted from the first coat. 196 Wood Finishing not work out well. It may dissolve the first coat and make the color run in very unnatural-looking ways. Although it is tempting to go right back over a coat of tinted shellac with another, resist the impulse. Don’t even consider a second coat until the first has dried at least 45 minutes. If you do get runs, brush it well to smooth them out. If that does not work, wash it off with a rag dampened with alcohol and start again. A second coat of wipe-on tinted varnish is no problem after the first coat of varnish has cured. If there are specks of dust in the dried toning coat, scuff lightly with dry 220 sandpaper to smooth them. (Avoid stearated sandpaper if the next coat will be waterbased or polyurethane.) Be careful not to cut through the toner layer, because the area that you sanded too much will show up as a lighter (untoned) area.. All you are trying to do is smooth the surface and scratch it slightly. Be especially careful not to rub through at the edges. Check the color and appearance after the last coat of toner. This is how the finished piece will look, except for the pale tan or yellow tones added by the finish coats. Top Coat At this point, the work to be finished should be flat and smooth as described above. It has been sanded to 180 or even 240 grit and dusted clean. Whatever tinting or toning you wish to do has been done. Alternatives for the top coat for a small wood shop are oil, varnish, shellac, water-based acrylic, and brushing lacquer. I won’t have much to say about lacquer because, frankly, I have tried it and not found any advantage, and many lacquer Table of Contents 32-11 products stink. (It has advantages in a spray finishing system, however.) In my view, selecting a top coat involves balancing three issues. These are (1) ease of application, (2) appearance, (3) protection and durability, including ease of repair. Ease of application seems to be the most important factor for many woodworkers, but in my opinion wipe-on finishes extract a high price in return for ease of application. The appearance of a wipe-on finish can be acceptable, but the appearance of many coats of brushed and rubbed-out varnish is far superior The thicker film of brushed-on varnish enhances chatoyance significantly. (Chatoyance is a striking three-dimensional appearance, also called wood iridescence, moire, vibrancy, shimmer or glow.) Wipe-on finishes are not as protective and durable as brushed-on varnish because they are thinner. Nonetheless, acceptable results are easy to obtain with, say, gel stain and 3 coats of wipe-on varnish. Many coats of varnish, followed by leveling and rubbing out, requires a lot of time. Wipe-on finishes are much quicker. Many manufacturers cater to woodworkers who want to avoid brushed finishes, so they promote an “oil” finish that can be wiped on and wiped off. Also, oil finishes are claimed to be more “natural” and even food safe. Some wipeon varnishes claim a hand-rubbed look, apparently hoping that the consumer will think that wiping on and wiping off is somehow equivalent to hand polishing. Many finishes that are labeled as oil finishes are nothing more than overpriced, thinned, varnish, i.e., wipe-on varnish. 197 Wood Finishing L INSEED O IL AND T UNG O IL Two oils are used as top coats: boiled linseed oil (BLO) and tung oil. BLO is very easy to apply. Wipe or brush it onto the wood. Keep the wood wet with it for 510 minutes, or until the wood won’t absorb any more oil, and then wipe off the excess. I scrub it vigorously with a dry rag to wipe off all excess oil after each coat. Allow it to cure at least overnight at room temperature. The next day, sand it lightly to smooth out any raised grain and apply one or two more coats, rubbing all excess oil off the surface. Allow each to cure overnight, at least. Exposure to warmth and sunlight seems to speed curing, but if the wood gets too warm, oil may seep out of the wood. Oily rags must be left spread out to dry so they don’t heat up and spontaneously combust. (This is not an old wive’s tale.) BLO cures to a soft and unsatisfactory film. That’s why it needs to be scrubbed off the surface. What remains may have a soft sheen and look nice, but it is too thin to be protective or durable. Tung oil (real tung oil) will build to an attractive sheen. Just because a product label says “Tung Oil” does not mean it is pure or even that it contains any tung oil. You can tell if a product is actually tung oil by reading the label or by smelling it. Pure oil smells nutty. Blends smell like varnish or MS. You brush or wipe it on, like BLO, but you need five to seven coats. Each coat requires two to three days to cure, and you have to sand between every coat. Frankly, it’s as much trouble as wiping varnish. Table of Contents OIL-VARNISH BLENDS V ARNISHES 32-12 AND WIPING Except for BLO and pure tung oil, most of the “oil” finishes on the market are oil/varnish blends or diluted varnish. Some of these include the words tung oil in their names. Oil/varnish blends do overcome some disadvantages of oil finishes. Examples of oil/varnish blends are Watco Danish Oil and Teak Oil; Behlen Danish Oil, Salad Bowl Finish, and Teak Oil; Deft Danish Oil; Maloof Finish; Minwax Tung Oil Finish and Antique Oil Finish. Oil/varnish blends are applied as described for BLO above. They dry a little harder than BLO but do not build well. Nonetheless, experts recommend that you use 3 coats of oil/varnish. Smooth each coat with 0000 steel wool and dust well before recoating. (See Arceneaux in References) Wiping varnish is varnish diluted with MS. You can make your own by diluting varnish with MS or turpentine until it is thin enough to wipe off with a rag. That’s roughly 1 part varnish to 4-5 parts MS. The advantage of wiping varnish is that it can be applied with a brush or rag, like oil finishes and oil/varnish blends, yet build like a varnish. Actually, it is better to apply wiping varnish with a brush, and then wipe it off with a rag. The disadvantage of wiping varnish is that the dissolved solids are lower than brushing varnish, so it takes many more coats to get the same build (film thickness) as a traditional brushed-on varnish – about 5 coats of commercial wipe-on to one coat of brush-on. Examples of wiping varnish are 198 Wood Finishing very numerous, but here are a few that try to conceal their true identity: Formby’s Tung Oil Finish; Zar Wipe-on Tung Oil; Val-Oil; Hoppe’s Tung Oil Varnish; Waterlox; General Finishes’ Arm R Seal. 32-13 bowls and such. I am confident that nothing will leach out after the varnish has cured. I also use two-part epoxy for small objects. T WO -P ART E POXY F OOD -S AFE F INISH One of the examples of oil/varnish blends mentioned above is called salad bowl finish. This suggests that it is more suitable for contact with food than other finishes. But, since it is an oil/varnish blend, it is no safer than the other oils and varnishes sold as wood finishes. There is no evidence that any finish leaches out dangerous or deleterious materials after the finish has fully cured. I do not know of actual food safety testing done by finish manufacturers, but some two-part epoxy adhesives are called food safe by their manufacturer. For maximum peace of mind, (at the cost of minimum protection and durability) consider finishing the wood with heavy mineral oil (liquid petrolatum) bought from a pharmacy. It will never cure and will eventually wash away, but you can reapply it whenever the wood needs it. Walnut oil is edible. It will dry, very, very slowly. Don’t use other vegetable oils. They may turn rancid and add unpleasant odor and flavor. Vegetable oils eventually do create a somewhat permanent finish, however. I have some salad bowls that have lost their original coat of varnish or lacquer and absorbed salad oil over many years. Next might be linseed oil, except that it is commercially produced as an industrial, rather than culinary, product and is slow to cure. I normally use thinned varnish (wiping varnish) to fill the pores of wood Table of Contents Slow setting two-part epoxy adhesive can be used as a finish. Because they are expensive and require extra effort I use them on smaller surfaces. I have experience with two of these products: West System Resin 105-A and Slow Hardener 206-A; and Max Clr Two Part Epoxy. Max Clr is called “food safe” by its manufacturer. These products leave a thicker coat than other finishes – think of a “bar top” finish. They are thick enough to fill the pores of walnut or oak with one coat. They do not penetrate the wood very much, so I sometimes pop the grain with BLO and then apply a coat of shellac as a base coat for the epoxy finish. I apply the mixed epoxy with a disposable brush. Brush it out well to ensure even coverage. Epoxies tend to run during the early hours of their 24-48 hour cure time. Put the object on waxed paper or the equivalent or you may find that it is permanently glued to your workbench. Epoxies often do not level themselves, so the cured finish may look dauntingly uneven. They can be scraped, sanded and polished like varnish, however. They are completely waterproof, although if water does happen to get into the wood and swell it, the finish may crack. I use Max Clr for drinking mugs, and small bowls meant for salad cereal, etc. I very much doubt that they can stand up to soaking, and certainly would not machine wash them. 199 Wood Finishing B RUSH -O N V ARNISH Varnish is a chemical combination of a resin and a drying oil, dissolved in a solvent like MS. I have been calling it brush-on varnish to distinguish it from wipe-on varnish, but they are chemically the same thing, except that wipe-on varnish has been diluted many times with MS. As may be clear by now, I prefer finishes that build – i.e., shellac and varnish. The three classic types of resin used to make varnish are called phenolic, alkyd, and polyurethane. Furthermore, the amount of oil left in the mix falls into two categories. Long-oil varnish, also called spar varnish, is intended for exterior use. The cured film is more flexible. Short-oil varnish (again, simply called varnish) dries harder and is intended for interior use. I prefer clear gloss alkyd varnish. c Polyurethane will wear well but is harder to rub to a nice sheen. It is very hard to repair because many finishes will not stick to it when it is fully cured. Polyurethane may not adhere to normal shellac so be sure to use dewaxed shellac under polyurethane. Also, polyurethane is infamous for yellowing over time. This may add a nice amber glow to some projects but may degrade the appearance of others. Traces of old finish, wax, silicone, etc remaining on stripped (previously finished) wood can cause imperfections in c I have had success with Sherwin Williams Wood Classics Fast Dry Oil Varnish and Benjamin Moore Benwood Fast Dry Clear Varnish. It has slightly more solids but seems a bit softer when dry. Satin varnish contains silica or some other suspended material that will make the finish cloudy if too many successive coats are applied. Table of Contents 32-14 the finish coat of varnish. When you have finished staining or tinting the wood (or if you skip that step) seal it with a coat of dewaxed shellac. Application Despite the terminology, undiluted varnish can be wiped on, it just can’t be wiped off. I have applied varnish with a cloth pad, a short-bristle paint pad, a foam brush and a bristle brush. I do my best work with a bristle brush. I prefer medium-priced natural bristle brushes. Cheaper brushes have fewer bristles and so hold less varnish. Also, they tend to shed bristles, which is an unnecessary complication. I keep brushes just for varnish and do not apply varnish with a brush that has been used with paint or oil stain. Cleanliness is critical. It is a good idea to vacuum the area and the work a few hours before finishing and then to wipe off all dust with a tack rag just before varnishing. Lighting is critical. I arrange the work and the light so that reflected light will show where I have already applied varnish and where I have not. On bare wood, I often thin the first coat of varnish about half with MS. Brush Technique Trying to describe brush technique is a bit like describing a dance step. You will have to develop your own by trial and error. Get a piece of plywood and practice. But here is an outline, with a few tips. (See also the references by Bob Flexner.) When the brush is still dry, dip it in solvent (MS for varnish, denatured alcohol for shellac, etc.). Dip it all the 200 Wood Finishing way to the ferrule, then let the extra finish drain out and blot the brush to remove most of the solvent. This will prevent the brush from soaking up the first few brushfuls of finish and make it easier to clean when you are done. Good brush technique is important but not especially critical with early coats. The main objective is to spread a thin, even coat. You will be sanding it before the next coat, so you do not have to be very fussy. You can brush it out until you feel that you have a thin, even coat. Most varnish sets up pretty fast, however, so don’t keep going back over it. The brush will begin to drag and you will create rough brush marks. The goal is to apply a generous but even coat. This is difficult to define. You can get an idea of how much varnish to apply, i.e., how much varnish should be in the brush and how thick the coat should be, by varnishing a vertical surface with good reflected light. The brush should be full enough to spread a coat without having to brush back and forth more than once or twice, but not so full that the varnish sags or runs. Hold the brush with your fingers on the ferrule and the handle cradled in the angle between thumb and index finger. Dip the brush in the varnish to a depth of about ½". This is not critical, but do not dip it too deeply. Lift it out of the varnish and let the varnish run off for a second or two. When the varnish has almost finished dripping back into the can, touch the last few drops against the side of the can. Don’t rub the brush against the side of the can because that usually puts air bubbles into the brush. Spread varnish in a strip about as wide as the brush from one edge of the surface to the other. Start the brush stroke Table of Contents 32-15 in the middle. Run a fairly full coat out to the right edge, or as far as the varnish will cover the surface well without heavy brushing. That’s usually about 12-18", maybe less depending on the finish and the brush. The beginning of that band will be thicker than you want. Use a touch-and-go motion (like an airplane landing) to spread the beginning of the first stroke out to the left. If the strip of wet varnish does not extend from edge to edge, repeat the process, starting the second stroke in the middle of what is left to cover. Out to the right edge, out to the left, to meet the varnish put down in the first stroke. When the strip is finished, tip off the varnish – just lightly run the tip of the brush along the whole length of the strip to smooth it out and pop any bubbles. It is better not to touch this strip again with the brush unless you see a serious problem. If I see a small skip at this point, I let it go. I’ll get it with the next coat. Then I start on the second strip, and so forth. Don’t brush over a corner, edge, or any projection because it will squeeze out extra varnish and make a run, sag or puddle. Re-coating On my best furniture, I prefer at least 4 top coats of clear varnish because I like to rub the final coat. Jewitt recommends two coats of varnish, leveling with 320 paper, and a final coat of varnish before rubbing. I would not consider wet sanding with 320 grit unless I had at least three coats over the stain or final toning coat. See details below. Regardless of the drying time on the label, let each coat dry at least overnight for varnish that says you can 201 Wood Finishing recoat it in 4 hours. d The drying time gets longer after the first 2-3 coats. Warmth and sunlight speed curing time, but you have to be careful. If the surface gets too warm it may bubble or uncured varnish in the pores may bleed out. The reason for longer drying time is that varnish “dries” in two stages. First, the varnish flattens and the solvent evaporates. Now the varnish looks dry. Second, the varnish cures (polymerizes). This chemical reaction requires time. How much time depends on ambient temperature and the chemical composition of the varnish resin. Lightly scuff sand between coats (320 or finer sandpaper). I f y o u t r y t o sand out brush marks, dust, bristles, etc., you risk sanding through the adjacent finish before removing or flattening the imperfection. e In the days before sprayed finishes, professional finishers would remove imperfections with a very sharp card scraper. If the scraper is sharp and you bend it just enough with your fingers, the edge will just catch on the imperfection and scrape it off. You can level fairly large brush marks this way as well. d After 2 undercoats I find that successive coats of fast dry varnishes may still soften the undercoat a bit. That is, they must be applied quickly, with even less tipping off than shellac or they will start to pull on the brush. This happens to some extent, even after 36-40 hours. e A full coat of S-W Fast Dry Oil Varnish is about 1.3 mils (.0013") when dry, according to the SW tech sheet. For comparison, a bristle from a good paintbrush is around 4-5 mils. You could easily rub through two coats of finish if you were trying to sand out a bristle and not being careful. Table of Contents 32-16 In my opinion, this use alone justifies learning how to use a card scraper (if you plan to varnish very often). This is a very useful skill because it removes some of the pressure to develop perfect, consistent brush technique. I do not, however, recommend that you try to acquire this skill on your nearly finished work piece. S HELLAC AND W ATER - BASED A CRYLIC Shellac is promoted as a sealer more than as a finish. It is actually under-rated as a finish. Shellac gives a very nice, hard, durable finish, silky to the touch, and it dries faster than varnish. It is resistant to atmospheric moisture. It dries quickly. You can usually apply the second coat after the first has dried for only 30 minutes to an hour. It does take longer to dry the more coats you apply. It is not ideal for table tops, etc where it may be exposed to prolonged wetness. Shellac is available as flakes that you dissolve in alcohol, or pre-mixed. Flakes are available in a range of colors from blonde to amber or garnet. Zinsser Bullseye shellac is readily available. It is not dewaxed. It is available as a three pound cut, meaning 3 pounds of shellac per gallon of solvent. It is not recommended as an undercoat for polyurethane varnish and some lacquers (use dewaxed shellac instead). It is usually acceptable as an undercoat for alkyd varnishes and is an excellent final finish. It is a bit less expensive and more widely available than dewaxed and comes in clear (yellowish) and amber colors Dewaxed shellac is clear, where normal shellac is somewhat cloudy in the can because of the suspended wax. The 202 Wood Finishing only pre-mixed commercial example I know of is Zinsser Seal Coat. Unless the label says, “dewaxed,” assume that shellac contains wax. It is said that you can dewax shellac by letting it settle for a few days and then decanting off the clear liquid. I find that this is impractical because the wax does not settle out enough. Bullseye shellac packaged in an aerosol can is dewaxed, I believe. Dewaxed shellac will apparently bond to any clean finish. I have no reason to doubt this. It is certainly an excellent undercoat for polyurethane varnish. Shellac has a limited shelf life. Old shellac does not dry hard. I mark the date on the can when I open it, and test unused shellac after about six months to a year. You can always test it on a piece of scrap. If the first coat dries hard in an hour, it is OK to use. Otherwise, discard it. I always finish the interior surfaces of a piece, to prevent warping with changes in relative humidity. I typically use shellac for this purpose. Varnish will off-gas for weeks or months and make the interior smell unpleasant. Shellac will dry in a few hours, and any residual smell is not unpleasant. Shellac is an excellent barrier to water vapor. Shellac requires quick, confident brush work because the solvent dries so quickly. On the other hand, it is very forgiving. Just sand the rough spots lightly and apply another coat. The first coat of shellac will feel rough after it dries because it raises the wood grain. A light sanding with 180 grit paper, followed by a second coat and another light sanding, makes a very smooth, silky surface that is ideal for drawers, both inside and out. A coat of wax on a shellacked drawer runner makes the drawer slide very nicely. Table of Contents 32-17 Jewitt recommends 3-4 coats of shellac, leveling with 320 paper, and 3-4 mo r e c o a t s b e f o r e r u b b i n g . H e recommends letting shellac dry at least a week before rubbing. Water-based acrylic, like Polycrylic® is about as colorless as any finish you can use at home. In fact, if you are used to a bit of warm tint from varnish or shellac, acrylic finish takes some getting used to. It has a cloudy bluish cast when wet, but dries quickly to a very hard clear finish. It dries fast enough that good brush technique is necessary, and it is so hard that it is difficult to rub out imperfections. Clean-up is with water. You can get very creative with this product, because it can be tinted with acrylic artist tints. If you ever wanted a blue maple table, with the wood grain showing through, tinted acrylic may allow you to have it. Leveling and Rubbing Polyurethane and acrylic finishes are a bit too hard to rub easily, but they can be rubbed. Alkyd varnish and shellac are easy to rub. If you were able to apply the finish coats smoothly, in many cases a final rubdown with 0000 steel wool, followed by hard furniture wax, will be just fine. For my best finish on a showpiece like a dining room table or living room end table, however, there is nothing quite like the look and feel of a hand-rubbed finish. Most people will visually judge a finish by looking at an oblique reflection of light across the surface. We hope that even a matte or eggshell finish will remind us of moonlight on a quiet lake. Leveling removes remaining brush marks, skips, bristles, etc., to come as close to 203 Wood Finishing that look as patience and skill will allow. Varnish should dry at least two days before leveling. Wet sanding is necessary for leveling varnish. Do not use water to wet sand shellac between coats. Dry-sand it or use mineral oil thinned with MS as a lubricant, although this method is quite messy, smelly and combustible. Leveling is precise work. Use a cork block or a wood block padded with felt under the paper. Don’t let the block tip or you may sand through the finish at the edges. Count your strokes. I know it sounds pretty compulsive, but when I don’t count strokes, I risk sanding through the finish. (See “An Experiment” below.) When leveling the next to last coat, 400 grit is safer but may be too slow if there are lap marks, brush marks, etc. Wet 320 grit sandpaper cuts quicker but requires more caution. Wet sanding removes much more finish than dry sanding because the grit stays more open. Therefore, it is more dangerous. Also, the water tends to cover any warning signs that you are cutting through the finish (until the water dries, that is.) (See important details below) I have never accidentally rubbed through multiple coats of finish with 0000 steel wool. Steel wool or a sanding sponge (400 grit) tends to get into shallow depressions so that the whole surface looks more even, but they do not level the surface. With a sanding block, depressions may show up as shiny. On the other hand, steel wool tends to ride over, rather than cut, foreign objects (dust and hairs) in the finish. Pay attention to the appearance of the water as you sand. If the varnish makes a slurry as it is removed, it is dry enough for wet sanding. If the swarf particles tend Table of Contents 32-18 to roll up or show other signs that they are sticking to each other, stop, dry it thoroughly, and wait another day. Stop frequently, wipe with a sponge, dry with a clean rag and look at the work at 90E to check the color and at a very oblique angle to catch the differences in shine. I lightly wet-sand the top coat with very fine (400 - 600) sandpaper just until the surface is smooth and the water wets it evenly. You can stop here for a matte finish, rub out the finish with pumice and rottenstone, or put it aside until the final step (wax). I prefer to rub the piece with pumice or rottenstone in water or light mineral oil on a felt pad. Before being rubbed with mineral oil (or wax) oil based varnishes should cure for 1-2 weeks or more, depending on the chemistry of the varnish. You can ruin a good varnish finish if you rub with mineral oil or apply paste wax before it is fully cured. After the varnish has fully cured, the final step is to apply a good paste wax (Trewax or Butchers) and buff. This is not protective at all, but it fills in any fine scratches or cloudiness left by the final sanding. References Bob Flexner, Understanding Wood Finishing. Pleasantville, NY Readers Digest Association, Inc. 1999. Excellent introduction to finishing. His introduction is must reading. Bob Flexner. Finishing for First-timers. Popular Woodworking. February 2000 Greg Arceneaux Easiest Finish? Danish Oil. Fine Woodworking January/February 204 Wood Finishing 2014, pp. 70-73. Jeff Jewett. Great Wood Finishes. Newtown CT. Taunton Press 2000 An Experiment After having sanded though the finish on a few projects, I wanted to know how much sanding would be safe while flattening (smoothing) a varnished surface. I prepared a flat, well sanded piece of wood and applied four numbered coats of S-W Wood Classics Fast Dry Gloss Oil Varnish. (I just wrote the number on top of each coat with a felt tip pen after it had dried.) I wet sanded it with a new piece of 220 grit silicon dioxide sandpaper, backed up with a cork block. Wet sanding required about 25 strokes of 220 to penetrate the top coat and about 25 more Table of Contents 32-19 Masaschi, T. Gel Stain Users Guide. Fine Woodworking Sept/Oct 2002 pp. 60-63 to remove it. (Both ways = 1 stroke) The work was smooth and flat except that a few shiny spots remained. Continuing the experiment, 320 grit sandpaper required just about 50 strokes to begin to cut through the next coat and about 50 more strokes (total 100) to completely cut through it. The work was now completely smooth and flat. I conclude that the thickness of brush marks and other irregularities in a varnish coat can be about equal to the thickness of the coat itself. Therefore, I should not consider flattening and handrubbing fewer than three coats of varnish, if I want two coats to remain on the work. 205 Wood Finishing Grit Size 32-20 Mean Particle Diameter Microns mils* 220 65-66 2.6 320 (CAMI) 36 1.4 Pumice (4F flour) 25** 400(CAMI) 23 0.9 600 (CAMI) 16 0.6 Rottenstone 0.51.0*** 0.2 0.4 *1 mil = 0.001" = 25.4 microns ** I have not found sieve data on pumice. I can’t see much difference (after buffing with a dry rag) between a surface rubbed with 600 grit silicon dioxide sandpaper and one rubbed with 4F pumice in oil. The 600 grit surface may be a bit glossier. ***Rottenstone particle size may be very approximate. Rottenstone does break down physically (gets finer) in use. See Chapter 29 “Abrading Wood . . . ” Table of Contents 206 Notes and Reflections While Shaving Wood Chapter 33 – Why I Work Wood There are many people who see craftsmanship as the source of a valuable ingredient of civilization. There are also people who tend to believe that craftsmanship has a deep spiritual value of a somewhat mystical kind. David Pye, The Nature and Art of Workmanship A life, because the alternative is unacceptable. Woodworking, like teaching, is a way of enacting my belief in human potential, certainly including my potential. Woodworking lets me enact values like discipline, patience, precision and competence. In that sense, it is an art. I do not claim that I produce art. I mean that the practice of woodworking (if not the products) expresses important parts of who I am. At least for a while, I can get closer to being the man I’d like to be. I enjoy learning and being useful. I especially like feeling competently useful. I choose to live as if my life were preparation for whatever comes next, as if I must keep my place in “the program” by developing and using whatever abilities I have. A good life is built up brick by brick (or board by board). I choose to live with this philosophy because it gives me a great now. Carpe diem. I admire craftsmanship, as I admire the occupation of subsistence farming. I enjoy using furniture that I made, just as I once enjoyed eating food that I had helped to grow. Like everyone else, I like recognition. I rarely ask for it, however. “Do you like it?” is an invitation for flattery. Flattery is the Chinese food of self-esteem (half an hour later you are hungry again). So I try to make my work stand out, speak for itself. Woodworking reveals a person’s real self because it involves so many kinds of decisions and tasks. How willing was I to plan my work, how well could I see that day, how s I work in my wood shop, there is ample time to reflect. Many tasks require my full attention, but sometimes the work is a bit like repeating a mantra. Planing wood, for example, is rhythmic, physical work. The plane sings across the board. The shavings roll off. The thinnest oak shavings are like lace. The board changes before my eyes from something rough to something shiny and smooth beneath my fingertips. Meanwhile, like those shavings, a thought may roll into consciousness. We once had an acreage in Iowa. We raised livestock, hay, and a large vegetable garden. Many farm chores had a similar effect. Cutting hay with a scythe, even mowing with a tractor and sickle-bar, up and down, across the field, was like that. I often reflect about my experiences: assess their value and explore what they mean to me. I have learned to value this process. Usually, I enjoy it, depending on the particular experience. Socrates’ “examined life” is richer and more worth living. Why Work Wood? Can one’s life really be improved by study, practice and discipline? Craftsmen and craftswomen believe that they can improve on nature. They can make clay into a pot, or a tree stump into a chair. (How could we believe otherwise?) It’s an appealing idea, and should be universally true, despite many apparent exceptions. I choose to act as if it were true in my 33-1 Table of Contents 207 33-2 Why I Work Wood patient was I, how disciplined (or not, for each of those). A piece of furniture, like life, is the cumulative result of myriads of big and small decisions and actions. You may perform thousands of operations correctly, and then in an instant, damage the piece or yourself. As in the rest of life, some woodworking mistakes can be re-done, some can be hidden, and some just have to remain where they are in plain sight, unless you are willing to take a loss and start again. As they say, that’s life. I usually enjoy the process of craftsmanship as much as I do the finished product. I enjoy the process of turning rough ideas and materials into something attractive, useful and durable. I enjoy the problems that I must solve and the skills I must learn to complete each project. It reminds me that I am still able to meet life head-on. Nonetheless, the finished piece is what it is. Results trump rhetoric. This reminds me of farming and gardening. If you let an animal die, or a crop fail, it doesn’t matter why, except perhaps as a learning experience. Woodworking cannot be “spun” into something it is not. Who would be interested in a little plaque on a table that explains why the legs wobble? If you are reading 33 chapters into this book, probably you have worked with your hands to make something that you cared about. Woodworking may have meaning for you, literally or as a metaphor for life. If you have never worked with your hands, this metaphor may seem meaningless and corny. In the latter case, stop reading and go find a piece of wood, or a camera, and try to make something appealing with it. Last, but not least, woodworking somehow creates in me a sense of connectedness. I like the connections to the past represented by my earlier work and older tools. When my uncle Homer died, Dad inherited some of his tools, and when Dad died, I inherited the few hand tools that were left after his down-sizings and garage sales. These included some hand tools, which I happily use today. When I pick up an old wooden-handled screw-driver, or a sliding T bevel with its nickel plating worn off by years of use, sometimes I remember that my uncle and my father handled them. (The fact that so few tools remain sets off another chain of memories about what my father’s life was like.) I also feel a connection to the future. I want to make something nice enough that people will want to keep it. Maybe someday one of my kids, or grandkids, grown up with families of their own, will recall briefly that an ancestor of theirs made an object that they use. Why Write about It? Issues like the meaning of craftsmanship interest me, and seem increasingly important in our throw-away consumer culture. I’d like to interest you in them as well. We need more craftsmen in the world. I write because I enjoy it. It’s another form of craftsmanship. To tell the whole truth, I also write because I have to. Writing is part of how I learn. I write so that I can understand and remember what I have learned. As I age, the understanding becomes easier and the remembering becomes harder. So I write it down. In the hope that it might connect with you, I try to make my writing be like my furniture: durable, attractive and useful. I hope you find it so. 208 Table of Contents Notes and Reflections While Shaving Wood Chapter 34– It’s Only a Hobby Y ou may hear the sentiment that “it’s only a hobby” quite often. The first time I heard that sentiment expressed about an interest of mine was from a neighbor in Iowa. My family had recently moved to a small acreage near Iowa City where we raised a big garden, hay, sheep and other livestock. The outbuildings needed a lot of work. The place needed gates, fences, etc. We all would get up before dawn every day to do chores. We worked like any farm family in the bitter cold wind of winter and the stifling, humid, heat of summer. I put in at least 40 hours a week on our acreage in addition to about 60 hours a week at my regular job. It was hard work, occasionally dangerous, especially because I was learning as I went along and made a few mistakes. We were a bit “house poor” for years after we bought the place.1 So I was taken aback by my neighbor’s unsmiling question, “So, what are you doing out there on the Prizler place? Are you a hobby farmer?” I needed to borrow his tractor for a few hours, and he was a neighbor, anyway, so I did not tell him what I really thought of his question. I just said, “Yeah, if it’s still a hobby after forty hours a week.” So, what about the idea of working seriously at a hobby, especially when it involves craftsmanship? I believe that some hobbies should be taken seriously because of their potential value to society and to individuals. I am proposing that you take your hobby seriously. Hobbies Are Valuable The idea of “its only a hobby” rests on untenable assumptions about the value of amateur work. It suggests that a hobby is mostly aimless relaxation like games or light reading. Even such hobbies as those are valuable, if only because most people need some aimless relaxation every day. We need, almost literally, to recharge our batteries – to restore the balance of neurochemicals in our brains. The importance of aimless recreation is demonstrated by how deeply we resent the spoilsport who tells us that we are “wasting our time” with such activities, or the superior player (or critic) who refuses to give any quarter and spoils our play in that way. Developmental and occupational psychologists see play as exploratory learning. Some play is a natural expression of curiosity. It may develop into serious learning, if the learning gratifies the learner. The origin of the word amateur is “one who loves,” meaning one who pursues an activity for the love of it, i.e., as a hobby rather than for money. It has come to imply a lack of mastery, discipline, or thoroughness, as in amateurish. Hence the idea of “only a hobby.” It’s what an amateur might say to excuse an amateurish performance. This modern connotation, however, is not consistent with the idea of doing something because you love doing it. Although the stereotype of a not-quite proficient amateur may contain some truth, it is a modern artifact. Many amateurs were, and are, very serious about their hobbies. Charles Darwin, who turned biology upside down, was an amateur naturalist. He evidently did 1 The definition of a hobby farmer is a person whose liabilities are in the country while his ass sets in town. 34-1 Table of Contents 209 34-2 It’s Only a Hobby not consider his journey half way around the world to the Galapagos Islands, or his books, as “only a hobby”. Sir William Herschel was a professional musician. With his sister Caroline, he discovered the planet Uranus, two moons of Uranus, two moons of Saturn, a variety of comets, and, for good measure, infrared radiation. Charles Ives, although educated as an undergraduate in music, was an insurance executive and amateur composer who became one of the first American composers of classical music to receive international standing. The final examples are the more than 123,000 active contributors to English Wikipedia. They have contributed millions of articles, and there are an additional 260 languages. These are amateurs who contribute valuable work for free. Furthermore, Wikipedia itself is a kind of “hobby” because all of it is available for free. Wikipedia may seem like a poor example because of its famous inaccuracies and abuses. These occasional inaccuracies are important, certainly, but are beside the point here, which is the existence of 123,000 amateur contributors. The vast majority of Wikipedia entries meet high standards and, by the way, all encyclopedias contain errors. One more irresistible example: Dr. Harvey Einbinder, a physicist and amateur historian, found more than 600 flaws in the Encyclopædia Britannica. In 1964, he published them in a nearly 400-page book, The Myth of the Britannica.2 2 All encyclopedias contain errors. In December, 2005 the journal Nature compared the accuracy of Wikipedia and the Encyclopedia Britannica. They gave 50 matched monographs, Hobbies have economic value to individuals, as well. On our “hobby farm” in Iowa we raised almost all of our own vegetables and meat. We raised much better lamb than was available in local markets. Goose and duck, which we raised for our table, were not available in our rural Iowa meat markets. As an amateur woodworker, in my first five years I made furniture that would have cost somewhere between $5,000 and $10,000 if it had been bought in a shop. Some of it was built to exact specifications. One piece is a reproduction of a hand made antique. These could not have been bought in any shop. The cost of the lumber was about $1000, but of course I made the furniture for the fun of it, so maybe the input cost should be counted as zero. Either way, my work added considerable value to the lumber. A Game Worth Playing The idea of “its only a hobby” also rests on unexamined assumptions about why people work. Most people have to work for a living, it is true, but work has a much broader psychological significance. H. L. Mencken said that a cynic is one who knows the price of everything and the value of nothing. If people go beyond minimum requirements, their work evidently has value to them beyond covering a variety of science topics, to experts who were asked to review the articles rigorously. The reviewers did not know where the monographs came from. The reviewers found 162 factual errors, misleading statements and critical omissions in the sample from Wikipedia, compared to 123 errors from the Britannica monographs. (Peter Macinnis, The Science Show, April, 2006. http://www.abc.net.au/rn/scienceshow/stories/2006/1 612253.htm) 210 Table of Contents Notes and Reflections While Shaving Wood its price. For an obvious example, most people in the West have some choice about how they make their living. Obviously the profession of woodworking offers psychological rewards to the people who do it for a living. Competition from industrial furniture manufacturers and overseas craftsmen makes it difficult to make a living by woodworking. Nonetheless, professional craftspeople remain. Most stay in the field because they prefer it to other occupations. So, the professional and the hobbyist share an attitude about how they choose to spend their time. Both enjoy the craft. Sociologist Richard Sennett observes that many people care about how well they do their jobs. They take even menial jobs seriously, and try to excel at them, with no expectation of additional material reward for their additional effort. In fact, Sennett defines craft as any job a person commits to executing to the best of his abilities. He says craftsmanship is wanting to do a job well for its own sake. To this, Matthew Crawford adds the notion that a craftsperson feels responsible to something beyond himself – to his or her “better self” or perhaps to the thing itself. Crawford sees craftsmanship as a means of learning as well as self-expression. He writes of the “the satisfactions of manifesting oneself concretely in the world through manual competence [which has] been known to make a man quiet and easy.” Craftsmanship must reckon with objective standards that do not depend on the emotions of the moment, but rather depend on the reality of the product, where shortcomings cannot be interpreted away. Craftsmanship is materialistic by its nature, but is anti-consumerist. Crawford points out that the craftsman is proud of what 34-3 he has made, and cherishes it, while the consumer discards things that are perfectly serviceable in the restless pursuit of whatever is trendy and new. Making something yourself, of which you are justifiably proud, is also an act of political expression. Research in the psychology of work has repeatedly demonstrated that, for many people, work is instrumental for a variety of important psychological outcomes such as personal dignity, self-identity, self-esteem, and self-direction. The following list is due to Prof. Estelle Morin: Social purpose - doing things that are useful to others Moral correctness-- actions, process and results that are morally justifiable. Achievement-related pleasure -Enjoyment of activities that stimulate the development of one’s potential and that enable achieving one’s goals. Autonomy – The ability to use skills and judgment to solve problems and make decisions. Recognition -- Activities and results that are visible to others and that are rewarded appropriately Positive relationships -- Activities that promote interesting and supportive associations with others. This framework makes work seem as natural as play, and further blurs the line between making one’s living and living one’s life. People may go beyond the minimum necessary to succeed in a paying job. Moreover, once the need to make money has been satisfied, people may continue to work for the pleasure it gives them. Increasing numbers of retiring Baby Boomers will be making this even more abundantly clear in the coming decades. Erich Fromm has put the same ideas in a much darker perspective. He sees work 211 Table of Contents 34-4 It’s Only a Hobby as an effective means to deal with the angst of death and emptiness. He once wrote: “The principle can be formulated thus: I am because I effect.” Therefore, work is a way to prove one’s existence, and to prove that one’s existence has value. As an aging woodworker, I am aware that I work not only to obtain positive benefits like those listed by Morin, but also to fend off the encroaching sense of diminished vitality. These psychological perspectives appear to have a basis in brain physiology. Kelly Lambert, a neurophysiologist, has reviewed animal and human research on the activation of the brain by various activities. She concludes that using both hands on crafts projects can be as beneficial to the balance of neurotransmitters in the brain as taking psychoactive medication. She believes that mental health requires effort-driven reward circuits to be activated in the brain. Many activities like knitting, planting, hunting, and cooking are primal activities that activate these circuits. They require planning, control, and the use of both hands. They can lead to a sense of accomplishment if done well. Excessive ease and passivity are not good for us because they de-activate these circuits. Perhaps the attitude of “It’s only a hobby” is actually harmful if it encourages us to abandon meaningful pursuits without a struggle. Job Satisfaction? Many people’s work today may not satisfy some of the needs mentioned above. Over the past century, many industries have centralized and routinized many decisions once left to the worker. This may be done for a variety of motives, e.g., to increase efficiency, to improve quality, and as a cheaper substitute for hiring and training competent workers. The American prototype is “Scientific Management.” This term was coined by its creator, F.W. Taylor, in the early 20'th century. For our purposes, Taylor’s fundamental idea was to move decision making about how to do a job from the worker to the manager. Taylor originally applied Scientific Management to manual workers like slag shovelers and bricklayers. He used “human factors” research to decide the most efficient shovel size, hod size, pace of work, etc. for a typical worker. Management would then dictate these efficient methods to all workers. Over the last century, many white collar occupations have also seen centralization of decision making, away from the worker and “upward” in the management hierarchy. Ivan Illich has described how this trend can extend even to “industrialization” of professions such as teaching, engineering, law, and medicine. The result is much more tightly prescribed work procedures. Some putative “managers” make few decisions themselves, but merely carry out decisions programmed by higher management. Also, a company may outsource some parts of the job. Where a butcher once prepared cuts of meat, he now sells pre-cut, pre-packaged meats. Where a chef once chose ingredients and prepared meals from them, a cook may now heat up prepared food products. Where a pharmacist once compounded a prescription, he may now transfer dosage forms made in a factory. The benefit of this manner of organizing and managing business may be economic success and the kind of abundant affluence enjoyed in the West. The cost, however, has been a staggering impoverishment of work for many people. De212 Table of Contents Notes and Reflections While Shaving Wood skilling may block achievement-related pleasure, autonomy and self-esteem. More highly prescribed (and circumscribed) work may lose its meaning as an enactment of the workers’ competence (personal knowledge, skill and values). By reducing the market value of competence, it also may alter recognition and interpersonal relationships. Following rote instructions and pushing a mouse button may deaden the effort-driven reward circuits that Lambert says are so important to mental well-being. Some people surely turn to hobbies to fill the void. Obviously, they would choose hobbies that satisfy the needs that are not satisfied by their “day jobs.” Therefore, a hobby can be even more important psychologically than a paying job. My personal experience was consistent with this. Throughout my career, my workshop time rose when my job was less fulfilling and fell when my job was more fulfilling. Finally, there are the millions of people who do not need to work for a living. I’m referring mainly to people who have retired from their careers, who have acceptable retirement incomes, and who will not tolerate idleness. They are looking for a means to continue receiving psychological rewards from work, if not financial ones. The retiree who chooses to spend his remaining days playing golf or watching TV is a stereotype that is losing whatever validity it may once have had. Possibly, earlier generations were worn out or disillusioned when they retired, but maybe the stereotype was never accurate. In any case, many people who retire today continue to seek the satisfactions of work, although they may work less regularly or less intensively than when they were being paid. 34-5 At least, that’s my situation. When I retired, after 40 years as a teacher and researcher, I was ready to stop being a professor, but certainly not ready to stop working. I spend four to eight hours a day in my wood shop, writing essays like this, teaching English as a second language, making and editing photographs, and building and maintaining websites. I approach all of these (mostly unpaid) crafts as seriously and almost as methodically as I had during my academic career. I study each subject in books, and choose projects primarily on the basis of their value to advance my education. Whether I would ever make money at them is and was irrelevant. Conclusion Not appreciating hobbies presents a tragic paradox: to be unfulfilled at work or in retirement, and yet to exclude serious hobbies as a source of satisfaction. We should at least allow ourselves to “follow our bliss” in our hobbies if not in our livelihoods. Amateurs are potentially valuable to their families and friends, themselves, and society. This, by itself, may motivate serious pursuit of a hobby. A simple economic “work for money” model is an oversimplification for contemporary craftspeople in Europe and North America because some people cannot satisfy all of their important needs in their jobs or in retirement while their juices are still flowing. Craftsmanship allows people to enact values and beliefs that are important to them. Fifteen years of self-directed study, and serious woodworking, photography, writing, etc. have deeply acquainted me with the meaning of craftsmanship. It does nourish my soul. 213 Table of Contents Whether I am paid for my work does matter, financially and in some other ways peculiar to our capitalist society. But whether I am paid does not significantly affect my attitude about my work or my aspirations for my results. (It does influence how much money I spend on tools.) In fact, I would suppose that anyone who has experienced the challenges and satisfactions of craftsmanship would be serious about it. I am surprised to hear any craftsman say, “It’s only a hobby.” References and Further Reading Crawford M. B. Shop Class as Soulcraft: An Inquiry Into the Value of Work. The New Atlantis Summer 2006-7. An expanded version was published as a book under the same title by Penguin Press 2009 Fromm E. The Anatomy of Human Destructiveness. New York: Holt: Rinehart and Winston, p. 235. 1973 Quoted in Morin. Lambert, Kelly. Lifting Depression. New York: Basic Books, 2008 Morin, E. The meaning of work in modern times. 10th World Congress on Human Resources Management, Rio de Janeiro, Brazil, August, 20th, 2004. http://web.hec.ca/criteos/fichiers/upload/MO W_in_MTimes_EMM200804.pdf Reilly, M. (Ed.) Play as exploratory learning: studies of curiosity behavior. Beverly Hills, CA: Sage Publications, 1974. Senett, R. The Craftsman,. New Haven, Yale University Press, 2008 214 Table of Contents Notes and Reflections While Shaving Wood Chapter 35 – Making Glued-Up Panels M aking up flat panels from narrower boards is a basic operation for making table tops and other wide pieces from solid wood. The process seems straightforward. As with most other woodworking operations, however, there can be pitfalls. Careful planning will produce a better result with less strain. Usually, a solid wood panel is made up of expensive lumber. An unsatisfactory result can be expensive as well as disappointing. The main issues to consider are • Selecting the stock • Matching and orienting the grain • Jointing the edges, • Gluing and clamping • • Plane all boards to equal thickness, at least c" thicker than the final thickness of the panel, (This assumes that you will be able to get the surfaces flush within 1/16" of each other when you glue up the panel and that the panel could be a bit thicker than you had intended.) Plane or rip all edges straight. Use a band saw or ripping sled if necessary for crooked boards.(See Preparing Lumber for Use in Chapter 4.) Selecting the Stock Think about matching when you select the boards at the lumberyard. • Select lumber that is at least ¼" thicker than the finished panel. Some final planing, scraping, and sanding will be necessary to flatten the panel after it is glued up. For example, I prefer a large table top to be f" thick, but the process has to go very well if I’m to get a f" thick finished panel from 4/4 lumber. It is nearly impossible to make a ¾" thick panel out of ¾" thick S4S boards. • Boards are often easier to match if you can cut adjacent boards (even the whole panel) from the same piece of stock or from boards that were adjacent when the log was sawn. • Straight grain is easier to match. Matching the Boards Appearance trumps other considerations, except for grain direction. With flat-sawn lumber especially, two adjacent boards with grain running in opposite directions (one with grain rising toward the surface, one with falling grain), are very difficult to plane flat. Sanding a panel smooth in your own shop is also quite unpleasant. Some people send the panel out to a service to have it sanded flat. That has never been an option for me. Bruce Hoadley points out (p.125) that quartersawn boards (flatsawn glue edge) that 35-1 Table of Contents Figured grain is more interesting but much more difficult to match. Allow extra width for edge planing, especially if the boards are crooked, waney, or damaged, or if you will be ripping off sapwood. Reject twisted or bowed boards. Allow extra length for trimming off planer snipe, checked ends, etc. You will also want to trim the ends of the panel after it is glued up. 215 35-2 Glued-Up Panels include wood from near the center of the log should be arranged pith side to pith side and bark side to bark side. He explains that the part of a board from near the center of the tree, with smaller annual rings, may expand and contract more than the bark side. That is, the pith side of the board may behave more like a flatsawn board. It seems to annual rings would be preferable to a bow. Just my opinion at the moment. Lay out the boards. When you get a pleasing pair, mark them with a triangle that will show how they were arranged, and mark them with arrows on both sides to show the direction they should be planed. When you get the panel laid out, mark that with a larger triangle as well and number each board consecutively. Jointing the Edges Figure 1. A QS Board Containing Smaller Annual Rings may exhibit expansion and contraction behavior of both QS and FS boards follow that this would be significant mainly for boards sawn from smaller boles. Figure 1 illustrates this point. It is an exaggeration, however. If possible, I would trim off the part of the board that was close to the pith. Some authors recommend alternating the direction of annual rings for boards with flatsawn show faces. I do not believe that this is necessary for wood that has reached a stable degree of dryness (EMC). Although flatsawn wood may tend to cup toward the “bark side” as it absorbs moisture, this is a minor issue for indoor furniture. In any event, changes in moisture content should be minimized by finishing both sides of the panel with approximately the same number of coats. If the panel will be attached to table aprons, etc. the clips will tend to hold it flat. Finally, I doubt that a washboard surface from alternating the direction of Preparation of the glue surfaces may be the most important aspect of making butt-jointed panels. This is true in part because of the relatively low clamping force achievable in a small shop. (See below.) 1. Fold two adjacent boards together as if the pair of boards were like the covers of a book. The spine of the book will be your joint. Let the top (show) surfaces be to the outside of the book. 2. Plane or sand the edges simultaneously. Use a block if you sandpaper the edges and take long, smooth strokes. Take off as little as possible. The advantage of doing it this way is that if you make a slight bevel on one board you will make the complementary bevel on the other. The disadvantage is that if you make a hollow it will be doubled when you place the edges of the boards together. You should plane off as little as possible because planing can easily create gaps at the ends of the boards. A plane makes a slightly thicker shaving at the beginning of a pass when only part of the plane is contacting the surface. Chris Schwarz found that a hand plane would take about .0055" at the ends of a board while 216 Table of Contents Notes and Reflections While Shaving Wood taking .005" in the center. It may not seem like much, but that end cut is 1.5 times thicker than the center cut. Repeated planing can add up to a gap at the ends of the boards that will make an unacceptable joint. You can also joint the edges with 100 grit and then 120 grit sandpaper. Use a block, not your fingers, to back up the sandpaper, and make long strokes from one end to the other. You can also use a trim bit on a table mounted router. Set the fence even with the bearing on the trim bit. HOW TO JOINT EDGES WITH A PLANE Not surprisingly, there are a few matters of technique that will influence your success when you joint boards with a hand plane. 1. Use a fence. Use your dominant hand to push the plane forward. With your other hand, push the toe down against the edge with your thumb and push the fence against the face of the board with your fingers. It is surprising how easily the plane can tilt, even with a fence. 2. Make sure that the plane iron (blade) is dead square to the fence (i.e., the sole of the plane) and that it spans both boards. 3. Set it for a thin (e.g., .005") shaving 4. Test the edges frequently for square and straightness. 5. Schwarz recommends that we start the cut a few inches in from the end of the board, cutting away from that end, and finish the cut a few inches from the other end. Make one or two cuts like that. Then make a cut all the way along the edge. If you get one unbroken shaving, test the edge with a straightedge or the board's mating 35-3 edge. If the edge is perfect or is a tiny bit hollow in the middle, stop. If it is a bit convex, remove another shaving in the middle. If there is any gap at all it should be less than a few thousandths of an inch and should be toward the middle of the joint, not the end. SPRING JOINT When wood exchanges moisture with the air, either drying or absorbing moisture, most of the exchange occurs near the porous ends of the boards rather than the center. Therefore, more dimensional change may occur near the ends than near the center. This has lead some experienced woodworkers like Gary Rogowski to advocate what is called a spring joint. Spring is a deformity in a board where the edge is cupped. (See Chapter 4) To plane a board for a spring joint, we create a very shallow hourglass shape. How shallow? A crack of light, a few thousandths of an inch, one pass with the plane. The hollow should be about ½ of the length of the board, centered about ¼ of the way in from each end. The hollow should be shallow enough that light clamp pressure will squeeze the edges of two boards together tightly. This is consistent with Schwarz’s planing technique, above, except that we do not have to worry about leaving a very slight hollow in the joint before clamping. Gluing and Clamping The objective of the glue-up is not only to make strong glue joints but also to align the boards so that their tops are flush. Any edges that protrude will have to be planed down, preferably on both sides of the panel. That’s a lot of effort and wastes a lot of wood. 217 Table of Contents 35-4 Glued-Up Panels A butt joint is the best kind of joint to use for gluing boards along the edge. A clean, properly mated and clamped long grain to long grain butt joint with white or yellow glue is stronger than the wood itself. Many people, Figure 2. Clamp Rack for Gluing Panels myself included, intuitively feel that more is needed. I have used rabbets, splines, tongue and groove, dowels and biscuits. They may help to keep the boards aligned during assembly but they do not strengthen the joint. (See the article by Gary Rogowski in References.) Some well-intended modifications (e.g., splines, T&G) may make a joint much harder to assemble and can actually weaken it. The boards have to rest on a dead flat surface so that the panel will not be twisted. You will need to raise the panel off the workbench while gluing so that you can get one side of each clamp underneath it. I use a clamping rack for this purpose. A clamping rack is two pieces of plywood glued together at right angles with notches cut in the upright piece to hold clamp bars. The height overall of each rack should be exactly the same, e.g., 3 ¾", so that the tops of the rack will all be in the same plane if the racks are on a flat surface. The one shown in Figure 2 is 24" long. You would need four of them for a 48" long panel. Prepare 3-5 cauls about 2" shorter than the panel will be wide. The length of the top cauls is not critical, but the bottom cauls at each end need to fit between the clamp racks. You can use almost any scrap, as long as it is strong (e.g., a 2x4 ripped in half lengthwise, 1½" square, and straight. You can use straight or slightly curved cauls. If the cauls have a very gentle arc along one edge, e.g., c" off at each end of 48" cauls, they will straighten under clamp pressure and apply enough force in the center to hold the boards flush. Plan on at least one pipe or bar clamp for each 12" of panel length. For example, if you are gluing up a 48" long panel, you will Figure 3. Panel in Clamps. Clamp rack holds the panel off the bench so that clamps can fit beneath it. (This 20" panel has more bar clamps per foot than recommended in the text.) need at least four pipe or bar clamps. You will need at least four strong “C” clamps for the cauls at the ends and two for the middle. You may need two additional “C” clamps for each joint. You will also need two clamp pads for each clamp, either the cauls or blocks of wood. Waxed paper between the cauls or clamp pads and the work will prevent the sqeezeout from gluing the pad to the work. Sometimes its easier to glue up 2-3 218 Table of Contents Notes and Reflections While Shaving Wood bars may flex under high force. This can put more pressure on the edge of the panel nearer the bar, which tends to distort the panel. The panel can actually buckle from strong clamp pressure on one side. boards at a time. 1. Put the bar clamps in the clamp racks. You will want one clamp about 6" in from each end and about every 24". (But see Step 5.) 2. Lay the boards on the clamp racks and check to make sure that the panel will glue up flat (with no twist or bow). The joints should be invisible when you slide the boards together. 3. Apply a thin coat of glue evenly with a brush or flat stick to the mating edges 4. Clamp the boards firmly but not tightly between the cauls. Make sure that the ends of the boards line up. The idea is to make the show surface of the panel as flat as possible. Place the cauls about 4" in from each end and in the middle. If you have only three cauls, put them on top and use a block of wood under each "C" clamp on the bottom side. Also, put a "C" clamp at each joint at each end if necessary to hold the boards flush. Put waxed paper, plastic wrap, or even typing paper under the cauls and blocks so they won't stick to the boards. 5. Add clamps over the top of the panel about 12" from the clamps in the rack that run under the panel. (From step 1) That should give you a clamp about every 12" along the side of the panel. Tighten the clamps to light-moderate pressure. Alternating clamps on top and bottom is necessary because the 35-5 6. Adjust the surfaces of each board so that they are flush with each other. If necessary, use a deadblow mallet padded with a block of wood to hammer the boards until they are flush. 7. Tighten the "C" clamps and bar clamps gradually and in rotation at least until you see squeezeout along the joint. Tighten the clamps as much as you can without distorting the panel. There is some disagreement among experts about how tight the clamps should be. While it is true that the edges should mate without the need for much clamp pressure, very high clamp pressure on well-mated surfaces is necessary for a strong glue joint. The small panel in Figure 3 is 20" long and ¾" thick. Therefore each glue joint has 15 square inches of glue surface. The five clamps may be able to provide 500 lb each, for an average pressure along the glue joint of 167 psi. This is not especially high. (See Chapter 28) 8. Don’t wipe off squeeze out. Wait until the glue has set up to a semisolid state (about 20-40 min.) and then peel it off very carefully with a steel putty knife or chisel. It should lift right off. If it’s still liquid let it dry longer. 219 Table of Contents 35-6 Glued-Up Panels If you try to wipe off wet PVA glue, it smears and makes a mess. If you wash it with water, the glue soaks into the pores of the wood. On the other hand, if you let the glue harden completely, it is difficult to remove and you risk gouging or tearing the wood. PVA glue is very difficult to sand, because it melts when it gets hot. Flattening The Panels After the glue has cured (overnight is safest) take the panel out of the clamps and lay it on the bench with suitable stop blocks and horizontal clamps (e.g., Veritas Wonder Dogs®). Inspect each joint for a ridge of swollen wood along the glue line. The water in the glue, when it is absorbed into the wood at the glue surface, may cause the surface to swell. A flatsawn show face may tend to expand more than a quartersawn show face. If any swelling from moisture is evident, let the glue cure longer and the joint dry. If the panel is uneven, flatten the back of the panel first. Lay a long straightedge across the panel and mark the high spots with a pencil. Proceed to plane down the marked areas. Plane with the grain, and as much as possible, from the high spots toward the low spots. Be careful not to strike any protruding board edges with the edge of the plane. If you break a piece of wood it may leave a divot. Depending on how level the surfaces of the boards are, I prefer to use a #5 bench plane. If I need to remove more wood, I use a scrub plane. (I typically use the thickness of a piece of paper (about .005") to set the cutting depth on my scrub plane, but it may take a somewhat thicker shaving. Read the grain and be careful to avoid tearout, especially with oak or figured wood. A toothed plane iron can be very useful when tearout is a problem. Whatever plane you use, the iron must be as sharp as possible. Also, take light cuts. Think about the next step when you will need to remove the tooth marks. With a scrub or toothed plane, I prefer to work diagonally across the grain at the high spots in one direction and then diagonally at 90E to the first cuts, until the panel is flat. Then I finish up with a #4 bevel-up smoothing plane, using the highest bevel angle I can manage. (See Chapter 27,Handplanes) Use a marking gauge to mark the final thickness around the edges of the panel, flip it over, and flatten the top. If you remove very much wood when flattening the panel, it may bow or warp in the next few hours. If this happens, it will be because one side of the panel has absorbed more moisture than the other side, and expanded. Usually, you can flatten the panel again by drying out the convex surface. Bowing like this can be prevented by sealing both sides of the panel with a coat or two of shellac. Naturally, you would need to apply dye or stain, if desired, before you seal the surfaces. References Christian Becksvoort. Edge Gluing Boards: Making Flat Panels with Nearly Invisible Joints. Fine Woodworking, December, 1989. Burkin, A. Making Table Tops Without Coming Unglued. Fine Woodworking Sept/Oct 1998 Tage Frid. Tage Frid Teaches Woodworking, Book 1: Joinery, pp54-56. Newtown, CT. 220 Table of Contents Taunton Press 1993. R. Bruce Hoadley. Understanding Wood: a craftsman's guide to wood technology. Newtown, CT. Taunton Press 2000. Roman Rabiej. Get Serious About Clamping: Most Woodworkers Are Underclamping Their Joints. Fine Woodworking Nov/Dec 2007. Rogowski Gary. Gluing Up Tabletops, Fine Woodworking Nov/Dec 2003. Christopher Schwarz. Cheating at Jointing Edges April 6, 2009 http://www.popularwoodworking.com Appendix – Clamp Pressure Glue should entirely cover both mating surfaces and should be .0004" thick or less. A thicker PVA or polyurethane glue film is weaker. (With the possible exception of slowsetting epoxy, glues do not fill gaps.) Clamp pressure is needed to force glue into the wood, to overcome swelling caused by the water in the glue, to hold the surfaces tight against each other and to immobilize them while the glue dries. According to Roman Rabiej, starved joints from excessive clamp pressure are not a worry with hand operated clamps. On the contrary, it is difficult to get enough force with hand clamps. Please refer to Chapter 25 for more details. Titebond recommends 300 lb per square inch for hardwoods. A 48" long panel of f" thick hardwood has 42 sq. in. of glue surface, needing a total force of at least 12,600 pounds. A pipe clamp, operated by a strong man, can exert about 1400 pounds of force. Therefore, to receive the recommended clamp pressure, this panel requires at least 9 pipe clamps along its length, that is, one clamp every 5". Experience teaches that we can get by with lower clamp pressures than this for a panel such as a table top that is supported across its width by an apron. 221 Table of Contents 35-8 222 Table of Contents Notes and Reflections While Shaving Wood Chapter 36 – Making Raised Panel Doors On A Table Saw Definitions wide. Stiles are the vertical members of a door or frame. They usually make up the sides of the door panel or face frame of a cabinet. Rails are the horizontal members. They are usually the tops and bottoms of a door or frame. The rails are almost alFigure 1. A Raised ways inserted into Panel Door the stiles by way of a stub tenon (short tenon or tongue) into a groove or mortise. Raised Panel The panel is located inside the frame of stiles and rails. A panel may be made of plywood or solid wood. A solid wood panel is usually cut with a beveled or cove profile to produce its raised effect. Solid wood panels must float within the frame to keep the panel or door from splitting. Panel Raising Jig. To make a beveled raised panel on a table saw, you will need a panel raising jig. (Figure 2.) This is a tall, angled secondary fence that rides along the rip fence of the table saw. You can also make a flat (non-tapered) panel without a panel raising jig or use ¼" plywood. For the sake of this discussion, assume all stock is ¾" thick. The door will measure 16" wide by 26" high. Doors usually need to be fitted somewhat into the opening. Suppose in this example that the opening is a little ( c" or so) less than 16 x 26. Let the stiles be 2" Making Up the Panel Larger panels, especially, may need to be composed of narrower boards glued together edge to edge. For details on ths subject, see Chapter 35, Making Glued Up Panels. Choose the panel stock and frame stock from the same board, if at all possible. By doing so, you will have better luck matching grain and color. If the same board will not yield enough stock, then try and match grains as closely as possible with other boards. You should expect to lose a little bit of thickness due to misalignment of the boards you glue together, so if you want a ¾" thick finished panel, start with f" thick boards. Set the frame stock aside for the moment. The rough panel dimensions in this example should be 13"x 23". Take the width of the door (16") and subtract the width of the stiles, 2 stiles @ 2" each. That gives 16"4"=12". If the grooves will be d" deep, then add ¾" to the width to allow for the part of the panel that will fit in the grooves in the stiles. That gives a panel width of 12¾" total. Add an extra ¼" to trim off later when sizing the panel to correct width. The panel height is figured the same way. Take the door height 26"- 4"=22" add the ¾" to make it 22 ¾". Add the extra ¼" for a rough panel dimension of at least 13"x23". After the panel is ready, remove the clamps and plane or sand the panel smooth. Resize the panel to a finished width later. Making the Frame: Stiles and Rails 1. Make the stiles and rails from stock similar 36-1 Table of Contents 223 36-2 Making Raised Panel Doors With a Table Saw in grain and color as the panels. Rip the stock to a finished width of 2". The stiles measure the full height of the door (26"). them together with slight clamp pressure. 2. Rip the rails to a finished width of 2". The length of the rails will be the same as the panel width, without the extra ¼" (12 ¾"). 1. Measure the inside dimension of the dryassembled frame. The door was 16x26, so our inside dimension should be now 12x22. If so, we need to add e" to both our panel width and height for the part that will fit into the grooves. The inside dimension from groove to groove should be 12¾ x 22¾; however, we need to allow room for our panel to expand and contract, so we'll cut the panel to a finished size of 12e x 22e. 3. You now have two stiles that measure 2"x 26" each, and two rails that measure 2"x 12 ¾" each. 4. Cut ¼" wide by d" deep grooves in the stiles and rails to accept the panel. You can use a ¼" straight router bit on a router table, a dado cutter in your tablesaw, or a plow plane. You could also use a single blade on the table saw, making the grooves in two passes. Assuming stock exactly ¾" thick, and if the single blade has a kerf of c", set the fence at ¼". Cut one pass, turn the board end for end, and cut the second pass. This should give you a groove ¼" wide, exactly centered in the stile or rail. Raising the Panel 2. The panel must fit into the grooves in the rails and stiles, so there eventually should be a ¼" thick tongue about 5/16" - d" wide all around the edge of the panel. Normally, the highest surface of the panel should be approximately flush with the surface of the rails and stiles. Therefore the tongue should be centered on the panel, ¼" from the front 5. Both the stiles and rails should have one groove each measuring ¼" wide x 3/8" deep running the total length of one side. 6. Now the rails need a ¼" thick by d" deep tongue or stub tenon cut on each end to be inserted into the grooves on the stile. See Chapters 19 and 2. 7. Dry fit the frame pieces together. Make sure that they fit together well and that the faces of the pieces fit together flush. Hold Figure 2. A Panel Raising Jig in Place on a Table Saw (See Chapter 15) surface. To make a beveled raised panel on a table saw, it is best to use a smooth cut “planer” or “glue joint” rip blade with a panel 224 Table of Contents Notes and Reflections While Shaving Wood raising jig (Figure 2). For instructions on how to make a panel raising Jig, see Chapter 15, Ripping on a Table Saw or The Accurate Table Saw by Ian Kirby. Set the jig against the rip fence and set the jig and fence to leave ¼" between the edge of the panel and the edge of the blade. Set the saw height at the required height so the cut will just barely protrude from the panel. Carefully run the panels through on edge, doing the end grain cut first. Bump the rip fence over toward the blade about 1/32" and run the panels through again to clean up any burn marks, or, you can clean up the panels later with a plane or sanding block. 3. After all four sides have been cut, the inside edges of the bevels may not look straight, square and even. Reset the rip fence the distance from the saw blade that equals the width of the bevel on the panel. For a 20E panel bevel that’s approximately 1d ". Set the height of the blade at 3/64" (between 1/32" and 1/16") above the table. Run the panels face down through all four sides. This will cut a reveal that will be square with the face and gives a starting and stopping point to the profile and face. If you lack a panel raising jig, you can cut the bevels by tilting your table saw blade, or you can make square raised panels. You can also make then by hand using a rabbet plane. Start with step 3 above. Use a shallow cut on the table saw to mark out the edges of the center of the panel. Then use a rabbet plane to cut the bevels. 36-3 You still need to create a tongue all around the panel, leaving a small space (c" ¼" or so) in place of the bevel. (Think of this as a 90E “bevel” instead of a 20E bevel). Use a dado blade to make a square reveal close to the rails and stiles, say ¼" or less. Or, you can use a tall fence and cut away a small amount from the front, leaving a ¼" thick by d" tongue all around. Or, finally, for a much more informal look, you can use a ¼" plywood panel. Finishing 1. Plane and sand away any remaining burn marks and saw marks. After the panel is finish sanded to the desired grit, (e.g., 180 grit) you can finish the panel. Coating the panel before assembly will prevent unfinished areas showing if the panel later shrinks a little during dry periods (in a house with dry heat). It also will help prevent the panel from accidentally being glued in place if there is any glue squeeze out during frame assembly. 2. Assemble the frame by gluing and clamping the stiles and rails together. Insert the panel before you glue the last end together. Glue and clamp remaining side after panel insertion. 3. After the glue has dried, finish sand the frame, being careful not to hit the panel. 4. Install hardware. 225 Table of Contents 226 Table of Contents Notes and Reflections While Shaving Wood Chapter 37 – Edge Banding E To get an invisible glue line, joint the mating surfaces together. This can be awkward if the edge banding is thin, but at least plane them flat. I much prefer to miter the corners, but butt corners can be used. For mitered corners, cut the pieces to the length of the plywood edge plus twice the width of the edge band (one width for each end) plus 1/16" or so for trimming. This is precise work. It takes practice and sharp tools. Very accurate, smooth miters are necessary. Do not expect a bunch of nominal 45 E cuts to close the frame with no gaps. If you have a table saw, perhaps your most accurate method would be the procedure described in Chapter 16, Crosscutting on a Table Saw. To use a cutoff (chop) saw, set the horizontal angle as close to 45E as possible. dge banding is a strip of wood or other material applied to the show edge. It is usually necessary in plywood construction to hide the laminations, unless you use face frame construction, frame and panel construction, or mitered joints. You can buy veneer ribbons in many wood species, often with hot-melt glue already applied to the back. This material is easy to apply with a household iron. As far as I know, it is durable unless something sliding across the edge of the veneer might catch on it and pull it loose. You can minimize the risk of this happening by chamfering the edge of the veneer. A very light cut with a small wooden chamfer plane is ideal for this. I prefer to use a strip of solid wood glued to the edge of the plywood, rather than a veneer. I prefer a strip at least d" thick. Thicker strips, say ¾"or more, can be used and will look like a design element. Butt joints are usually strong enough, but if the edge banding is wide or I expect that it may be subject to stress while the piece is in use, I may decide to cut a “V” shaped cross section in the edge banding and the edge of the plywood with a router bit. It takes a bit of fuss to set up the cuts, but it helps to align the edge band to the plywood and increases the glue surface. Avoid the temptation to select any odd cutoffs that are laying around the shop to use as edge banding. At least a part of the banding probably will be in prominent view, so the wood should look decent and match the color of the plywood or the other hardwood in the piece. Also, straight grained wood will be much easier to work with and less likely to curl off the edge as temperature and humidity change over the life of the piece. Figure 1. The grain in this diagram rises vertically from left to right. It should be planed in that direction. (Make test cuts to check it.) If this setup does not produce a precise 90E joint, you can cut adjoining pieces a pairs. Cut one miter as an acute angle (45E) and, without moving the angle setting, cut the mating 45 from the opposite side, as an obtuse angle (135E). Any error in the one will cancel the error in the 37-1 Table of Contents 227 37-2 Edge Banding other and they will form a 90E angle. The principle is that two angles can add up to 90E even if they are not exactly 45E Cut one piece with the work to the left of the blade and one with the work to the right of the blade. Use hold-downs, etc to avoid crossing your arms and to keep your fingers away from the blade. A good miter box may be capable of accurate 45E cuts. If not, use the procedure described in the preceding paragraph. Finally, a shooting board with an auxiliary fence for 45E planing is helpful to get the ends neat and exact. Starting at one corner, cut and plane both halves of the first joint to form a 90E joint. When you are satisfied with the fit, mark the opposite end one of the edge bands and cut it to the outside of the mark (at 45E). Then cut the other side of that joint & plane both sides to fit. Work your way around the top in this manner. “Vertical” grain direction of edge banding matters greatly, especially in coarse grained wood like oak. It is important to plane with the grain. Because there is less wood to provide support to the fibers, planing against the grain (i.e., into the ends of the grain as it rises to the surface) can split the edge banding, or cause very deep chips and tearouts. (See Figure 1 above and READING THE GRAIN in Chapter 27, Handplanes.) Ideally, you would orient the edge banding so that its grain rises or falls vertically in the same direction as the show face of the plywood. That way, if the plane does catch the plywood, it will be more likely to shave it and less likely to dig in and tear the wood. I do not always plan that carefully, but when I don’t, I have to be doubly careful when leveling the edge banding to the plywood. The edge banding must be installed proud of the plywood, preferably 1/32" or less on each side. (It can initially stand as proud as you like, but removing more edge banding is more work and allows more opportunities for a mistake.) Often, ¾" finished hardwood is thick enough, because nominal ¾" plywood is usually less than ¾" thick. It takes some fine adjusting to center the 23/32" thick plywood on the edge banding. Often, I just set the edge banding flush with the back to make sure that I will have something to trim off on the show side, or place a piece of paper under the plywood and then push the edge banding against the benchtop. However you do it, make sure that the edge banding is a bit proud of the show face along its entire length. After the glue is dry, the edge banding should be leveled to be even with, or slightly proud of the surfaces of the plywood. I would always plane it flush on a table top where it might catch something slid across the top. Some people (not me) use a flush trim bit (straight router bit with a bearing on the end) will do this job well. I trim edge banding by removing successively thinner shavings. I might start with a bench plane, e.g., smoothing plane; a low-angle block plane; a cabinet scraper; or a card scraper. (Figure 2) Choice of tool depends on how much wood remains to be removed from the edge banding, how sharp your plane iron is, and how fine a shaving you can take. I would not attempt this operation with sandpaper, even sanding by hand, and certainly not with a power sander. The risk of sanding through the top ply is too great. My preferred tool is a Veritas lowangle block plane. It has an adjustable mouth. As I approach level, I set it to take very fine shavings through a very narrow mouth. Then 228 Table of Contents Notes and Reflections While Shaving Wood I follow with a scraper. If the grain is twisted and irregular, I use a plane with a high cutting angle, close the mouth, and take very thin shavings. (A bevel-up smoothing plane with a high angle cutter, e.g., 50 E is ideal.). Alternatively, for difficult grain start with a cabinet scraper. Begin by finding an area where the edge banding is most proud of the surface, ideally a high corner. (See Figure 3) Start by lowering that area and work out from there in both directions, cutting with the direction of the grain. Hold the plane at about 45Eto the direction of the edge. The heel of the plane should be on the plywood, and the iron should be on the edge banding. Use a slicing action to remove a very thin shaving from the edge banding. (Figure 2) This takes a bit of practice and a lot of discipline and patience. Clamp a guide for the heel of the plane, or use a piece of paper or protective tape on the plywood until you get comfortable at it. In theory, if you hold the plane correctly, with the heel firmly on the plywood, you cannot cut into the plywood. In practice, slice off edge banding until you see the glue line start to disappear (actually it is the shadow that disappears) or see the plane cutting into glue squeezeout near the glue line. When the next stroke might remove too much, switch to a scraper. Be especially careful as you approach corners, especially if they are uneven. (Figure 3) If you bump the toe of the plane into the edge banding on the adjacent side of the corner, you may knock it loose or damage it. That’s a good reason to start by planing down the high spots at the corners. Take great care not to let a cutting edge touch the surface of the plywood, especially when cutting the ends (across the 37-3 Figure 2. Position for Planing Edge Banding Flush direction of the grain on the plywood). If the plane digs into the plywood, it will leave permanent scratches, or worse, a partial divot. Such damage may not show up until you apply the finish, especially if you use pigment stain. Plane marks across the grain of the plywood are very noticeable and will have to be scraped or sanded away. If they are deep, this may result in going through the top ply. Note the thickness of the surface veneer of your plywood. Most hardwood plywood has a paper thin surface veneer, e.g., 1/100" or so. You should not remove any of it, except for very light hand sanding or light scraping in the finishing step. The layer under the surface veneer has its grain running at 90E to the surface grain. It may be a different species, and is often a different color and ugly. In other words, you cannot disguise or patch an area that you have sanded through unless you draw or paint faux woodgrain on 229 Table of Contents 37-4 Edge Banding it. (See Chapter 30) A card scraper is a finishing tool. It is OK to scrape across the edge banding and the plywood, to make a perfectly flat joint. Just make sure of two things: (1) the edge of the scraper must stay exactly on the plane of the surface – do not let it tip at Figure 3. Uneven Corner. all in either If the body of the plane d i r e c t i o n , hits the top edge, it may especially not damage it. dig into the plywood; and (2) you must scrape in the direction of the plywood grain, as you would if you were scraping it as a finishing operation. If you scrape at an angle, the scratches or gouges will show later, especially if you stain or dye the wood. After all of the edge banding has been leveled, scrape off any glue that may have smeared on the plywood. You can scrape or lightly sand to remove any minor plane marks from the plywood. If you use sandpaper, hand sand with 220 grit paper backed up with a block. The final step is to smooth the outside edges (if necessary), repair any damage, relieve the sharp edges, and smooth the miters. After you have finished, moisten the area around the edges with alcohol or mineral spirits and inspect it carefully for scratches before applying stain. Reference: Mario Rodriquez: Trim Plywood Edging Flush Fine Woodworking.com (Short video) 230 Table of Contents Notes and Reflections While Shaving Wood Chapter 38 – Hand Made W however, may have been spun by an industrial machine or by hand, using a spinning wheel (a human-powered machine). A hand-thrown pot can be made on an electric potter’s wheel, and hand sewn clothes can be made on an electric sewing machine. Hand made bourbon is mainly advertising hype, but it means that the whiskey was made e expect a piece of fine furniture to be exceptionally attractive, useful, and durable. For most people, “hand made” adds another dimension of excellence. As a craftsman, however, I have to recognize that some factory made furniture may be as attractive, useful, and durable as some of the furniture I make, and – by the way – cost less. I am sometimes tempted to claim that my furniture is “better” than factory made because it is hand made. But, what does that mean? If hand made furniture is not necessarily more attractive, useful and durable than some factory piece out of a catalogue, is such a claim just hot air? Purely as a thought experiment, for example, how might anyone be the wiser if I bought a factory piece and then just claimed that it was hand made? Moreover, if you press me at all, you will discover that I use machinery to make my hand made furniture. I can see no reason to rip a plank with a hand saw, when my table saw or bandsaw will do the job as well or better, in a small fraction of the time. (And with less sweat.) Even if I didn’t use machinery, the lumber that I used to make it was cut by machinery. So is my so-called hand made furniture really hand made? My dictionary does not support a claim that I can use power tools to make hand made furniture. It says that hand work is performed without the use of machines. (I’m sure it means machinery powered by steam or electricity.) So, I turn to analogies. Hand-painted furniture or glassware, etc., seems selfevidently painted with a brush by a human. Hand made paper is actually made by hand, one sheet at a time. A handmade carpet is literally woven or knotted by hand. The yarn, Purely as a thought experiment: How would anyone know if I bought a factory piece and then just claimed that it was hand made? in small batches. Hand-blown glass is, of course, not touched by hand at all (ouch!) but is blown by human breath, one piece at a time. So, hand-made furniture isn’t quite the same idea as some other hand-made items, but it does share with them the idea of individual character, craftsmanship, one-at-a-time production, and a connection between the craftsman and the user. In David Pye’s terminology, craftsmanship requires the workmanship of risk, where an operation is under the control of the workman and could go wrong if he or she were not sufficiently competent and attentive. This idea applies equally well to hand tools, hand-held power tools, and hand adjusted bench power tools. It seems to exclude automated substitutes for craftsmanship. Pye believed that workmanship of risk would produce small diversities in fit and appearance that would enhance the beauty of the piece, and which factory made furniture could never achieve. This is what many people mean when they say hand made. 38-1 Table of Contents 231 38-2 Hand Made (Workmanship of risk is not a term that I intend to use very often in describing my work.) A hand finish is a good example of the workmanship of risk. Normally, the craftsman will have chosen wood of similar color, grain pattern, etc., so there are few imperfections to hide. Then a clear finish can be used to actually amplify the character of the wood. Factory furniture may be finished with colored laquer to hide mismatched boards and achieve a uniform color. Unfortunately, this also hides the character of the wood. Although one strives to make a hand-applied finish perfectly flat and the gloss uniform, little imperfections give the finish character, compared to a factory laquer or plastic finish, which is “perfect.” People who value hand made furniture may have two more ideas in mind. The first is uniqueness. When a client says, “hand made,” he usually means that the piece is not massproduced. A piece designed to suit a client’s particular tastes and needs for color, style, dimensions, materials, and finish probably deserves to be called hand-made. It is technically custom made or (in the UK) bespoke furniture, but it would inevitably require a considerable amount of hand work, despite the fact that power tools were used in some operations. In my shop, it will certainly be fitted, assembled and finished by hand. A piece of custom hand made furniture is assembled and finished part by part, regardless of whether the parts were made using power tools. Each step is visible to, and under the control of, the craftsman who is making it. The quality then depends on the competence of the craftsman (including his discipline to “do it right or do it over”). The second connotations is a sense of connection to something made by human hands. So much in our world is massproduced, this value may equal the others in importance. Some people are delighted by the idea that somebody actually labored to make a piece of furniture, to their order, specifically to please them. If it were made by a personal acquaintance, friend or family member, so much the better. When you open the drawer of a hand made desk, you feel the smooth motion and hear a quiet whoosh because a craftsman cut and trimmed the drawer until it fit perfectly into the opening. It does not matter whether he used a machine or saw and chisel to cut the dovetails. The silky feel of a hand rubbed top results only from a craftsman’s patient efforts. Such a piece did not come from a factory. You are touching something that a person took pains to make. So, the idea of hand made furniture is on shaky ground if we try to take it literally. The term is valid, however, if it refers to a unique piece made by a particular craftsman to match his skill to the needs and desires of another person. Hand-made, in this sense refers to a metaphorical hand-clasp across time and space. 232 Table of Contents Notes and Reflections While Shaving Wood Chapter 39 – Boxes and Drawers A box is a fundamental form in woodworking. Cabinets and other large boxes are often termed casework. Drawers are smaller boxes. Even smaller boxes usually have lids and are decorative as well as useful, e.g., a jewelry box. This chapter is primarily oriented toward drawers and smaller boxes. Joinery Options There are more available joinery alternatives than I can describe here. Besides, the number of choices can become confusing. All but one of the joinery methods described here can be Figure 1. Corner Tongue-and-Groove Joint used in casework. The drawer lock is the exception. And there is one very useful joinery method used in casework that is not usually used to make drawers – the corner tongue and groove joint (Figure 1). This joint is very strong and hides the edges of plywood. It is an excellent joint for casework. For more information, see Frid and Rogowski in References. For the drawers in my first fine woodworking project, a pair of mahogany bedside tables, I used half-blind dovetail joints at the front and box joints at the back. These are challenging but were great choices for a novice because my objective was to learn, and I could take my time. I was using this project to teach myself some fundamentals. These drawers were the occasion for me to make a box joint jig for my table saw and to learn how to cut dovetails by hand. If you can’t spare that much time, drawer lock joints on a table saw or router table might be a better choice. DOVETAIL JOINT Usually dovetails on drawer fronts are halfblind, but some designers (Krenov school) use through dovetails as a design element. Dovetails are not well suited for plywood, but good dense plywood with no voids, e.g., Baltic Birch, might be OK in certain cases. If I plan to use hand-cut dovetails for a drawer, I normally plan 1-3 relatively wide tails. If I will use a jig, the jig determines the number of dovetails. (See Chapter 23.) Sometimes, I attach a false front to a drawer box made with through dovetails or box joints, although they can show, as a point of style. For more utilitarian furniture, I use box joints, drawer lock, or lock miter joints. BOX JOINT Box joints look neat, they are quick and easy to make once you have a good jig, and they are at least as strong as any other kind of joint. You can cut them by hand, but if you will cut many fingers (e.g., ½" fingers in a 4" drawer means 32 fingers and sockets) the advantages of machine-cut box joints are obvious. You can make them with a table saw, box joint jig, 233 Table of Contents 38-2 Boxes and Drawers and a good dado stack. If I don’t have access to a table saw and a box joint jig, I cut them with a router and box joint jig. (See Chapter 23). DRAWER LOCK AND MITERED JOINT The drawer lock or lock miter joint is quicker than a box joint if you have a lot of drawers to do. I often use drawer lock joints all around for smaller drawers. This joint is strong enough for most boxes made of ½" stock , at least up to about 6" wide, especially if the bottom is glued into a groove. Also, you don’t need to use false fronts with a drawer lock joint, because you can inset the sides ½" to accommodate drawer guides. Layout is easy., say up to 4" deep. The lock miter is very strong and conceals end grain. It is strong enough for case work or the long mitered joints used in legs of Mission style furniture. (Lock miter joints get a separate chapter – see Chapter 22.) The main issues with drawer lock and lock miter joints are set-up time, and tear-out. Setup for these joints may require a lot of trial and error until you get a good fit. The setup time can be shortened somewhat with experience and certain jigs and tricks. These joints, lock miter especially, may cause tear-out and even shatter the inner plies in cheaper grades of plywood. Also, both sides of a lock miter joint have to be the same thickness. Mitered joints are good for plywood because the joint hides the plies. Also, because the grain direction is rotated 90E in each ply, a mitered joint in plywood is not a pure end-grain joint and may be a bit stronger than a mitered joint in solid wood. For small, decorative boxes I like a miter joint, usually pinned or keyed for better appearance and more strength. A splined miter joint is strong enough to be used in a drawer box, and should be considered. A mitered joint can be strengthened by inserting transverse pieces, giving a sort of false box joint look to the joint, or even longitudinal keys cut across the ends. (See the section on mitered joints.) Mitered joints take some tricks to assemble. OTHER JOINTS A sliding dovetail (cut vertically in the rear of the drawer front) might also be a great joint if the sides are solid wood, and if the sides need to be inset to accommodate drawer guides. I rarely make a box with butt joints although I have made strong boxes with rabbeted corners. I would not trust these for drawers, however. Design Considerations Detailed design (planning) is important to get well-fitted and attractive (professionallooking) drawers. Here are some things to think about. 1. Wait until the piece has been dry-assembled before you can get final drawer dimensions. If you cut the parts out in advance, cut them oversized. If the actual drawer opening is out of square, seriously consider a false front or a lipped front. See Chapter 40, Fitting Drawers. 2. The groove cut to hold the bottom will show at the ends of the front and back unless (a) a dovetail or a box joint pin on a drawer side hides it or (b) you cut stopped grooves. For a box joint, that means that the sides 234 Table of Contents Notes and Reflections While Shaving Wood should have a finger at the bottom. For a dovetail to cover the groove a thin (¼" wide) half pin is required at the bottom of the drawer front. This may be strong enough in oak, but not necessarily in other woods. This issue comes as an unpleasant surprise when you are assembling the box and the half-pin breaks. Sketch out this detail to avoid this. 3. Another design issue is the drawer guide style. Classic design uses wooden guides (“L” shaped in cross section) beneath and beside the drawer bottom. These work just fine. Ball bearing or wood guides attached to the side of the drawer require space, usually about ½". Then you have to hide that space when the drawer is closed, e.g., with a false front. 4. The dimensions of a drawer will affect how easily the drawer works. A drawer should be at least 10% deeper than it is wide, or it may tend to bind. You should consider using plastic pads/buttons or ball bearing guides for wide, shallow drawers. 5. If you will use a jig to cut the joint (e.g., box joints), your design should consider the consequences of using the jig. For example, see the first paragraph in the section on making box joints. 38-3 a back that extends from the top of the sides to the top of the drawer bottom. This allows the drawer to be assembled and then the bottom to be slid in. Sometimes I prefer a full-sized back, with a groove, so I assemble the bottom simultaneously with the rest of the box. This may strengthen the box somewhat, but it requires more accurate cutting of the bottom and a bit more fuss when assembling the box. After I cut out the front, back and side pieces, I cut the grooves for the bottom, so I can more readily see which side is “inside” and which edge is my reference edge. I usually cut them so that the top of the groove is ¼" from the bottom edge. I cut the grooves with a dado stack on a table saw (preferred) or with a straight bit on a router table. The table saw method limits the width of the dado to ¼" or wider. Most ¼" plywood is actually less than ¼" thick, so the bottom may rattle in a ¼" groove. However, if the bottom fits tightly into a groove, it is important to make sure that the bottom is seated fully so that the sides are straight. Also, using the table saw requires the groove to go the length of each piece. If I use a router, or plow plane I can cut a narrower groove and stop it before it reaches the ends of the front and back. Box Joint Box Bottom I make box bottoms from ¼" Masonite or plywood set into a ¼" deep groove. For large drawers intended to contain heavy objects, e.g., workshop furniture, I use a ½" plywood bottom, but that is probably over-building. One great advantage of plywood or Masonite is that the bottom can be glued in, which will strengthen the box. Classic drawer construction often uses A design that will use box joints should consider details of layout and construction. For example, the exact height of a drawer box may not matter much in an overall design, but it will matter a lot in the construction of the actual drawer. Most of my boxes are made of ½" Baltic birch plywood, on a jig that is set up to cut ½" sockets and fingers (pins). Therefore I should design drawer boxes that are multiples of ½", preferably odd multiples, 235 Table of Contents 38-4 Boxes and Drawers e.g., 4½", 5½" etc. 1. If the width of the joint (i.e., height of the box) is not an even multiple of the width of the fingers, there will be a partial finger on one side of each joint. If it’s half a pin, e.g., ¼", it may be OK, but a partial finger less than that would be weak and might break off during assembly. 2. An odd number of fingers and sockets will yield a symmetrical joint, e.g., with a finger at both edges or sockets at both edges. Example, for ½" fingers/sockets, a board width of X and one-half inches will work out well, e.g. 4 fingers + 3 sockets = 3½ “, 5 fingers and 4 sockets = 4½" etc. 3. When preparing stock, recognize that you may want the fingers on the box joints to protrude 1/64 - 1/32" initially, so that they can be planed or sanded flush with the sides. If so, add 1/32 - 1/16" to the length of front, back and sides. LAYOUT Layout before cutting is almost as important for cutting box joints with a jig as it would be for cutting them by hand. The jig can help you follow your layout accurately, but it cannot give you the layout. It is easy to get confused while cutting, and to reverse the fingers and sockets on at least one board. Also, cutting box joints is repetitive, and it’s easy to mis-register a board in the jig. If they were laid out with a pencil, you could at least see deviations from the layout and perhaps cut the mating board to match the mistake. A ruined joint is not a disaster, but it does mean that the piece will have to be remade, which means that the box joint jig and dado will have to be removed to cut the new piece(s), then replaced, re-adjusted, etc. Boring. Do it right the first time by laying out at least the starting socket or finger on each reference edge. Mark the sockets with an X 1. I think that a box joint looks best if the sockets are no wider than the stock. I make most of my drawers out of ½" plywood (usually Baltic Birch) so most of my box joints use ½" fingers and sockets. 2. Lay the boards in a rectangle & mark the eight mating ends, A-H as shown in Figure 2. Choose and mark a long reference edge (RE) for each board. For boxes, drawers, etc. the bottom side (next to the groove or rabbet for the bottom) should be the RE. Make sure the ends are square to the RE. 3. Decide how you want the finished joint to look. For example, A. If you cut a ¼" groove in the sides and front, you might want to lay out the joint so that there is a finger at the bottom of each side, to hide the groove. B. On one hand, a partial finger would be hidden best at the bottom of Figure 3.. Shop Made Box Joint Jig 236 Table of Contents Notes and Reflections While Shaving Wood the joint, but you would want to start cutting at that side of the board. Your choice. Avoid the decision by making the joint an odd multiple of the finger width. 4. A marking template is very handy. If you don’t have one, plan to make one out of Masonite after your jig setup is complete. Box Joint Jig I cut box joints on a table saw or router jig (see Chapter 23) and occasionally by hand. You can buy a box joint jig (BJJ) for less than $100 or make one easily. (Some BJJ come as part of a miter gauge extension kit.) A shop-made BJJ can be either a miter gauge extension, which is clamped to the miter gauge, or a sled that runs in the miter gauge slots. The advantage and disadvantage of each is adjustability. The miter gauge extension can be adjusted to get fingers of exactly the right width. This is, however, laborious and has to be done every time you attach it to the miter gauge. The sled type cannot be adjusted, so it requires great precision to make, but once made it is ready to go whenever you need it. The essential point to making a good BJJ is to make B=C=D in Figure 3 within a few thousandths of an inch. The key has to be a hair thinner than B so that the sockets that you have just cut will slide over it easily, but still fit with no side play. If the socket width is B, the outer edge of the indexing pin has to be exactly 2B away from the socket , as shown in Figure 3. Cut a socket in the board that will become the jig. (For ½" pins, cut a ½" x ½" socket.) Then cut a piece of hardwood slightly oversize, e.g., ½" x 17/32" and hand plane or 38-5 sand it until it fits perfectly in the socket with neither binding nor side play. Glue a piece of it (about 1" long) into the socket you just cut. Use another Figure 2. Layout of piece of the ½" x ½" Box Sides pin as a spacer and then cut another socket in the board exactly 1 pin-width away from the pin. The height of the pin can be carefully shaved down later if you ever need to cut sockets that are less than ½" deep, e.g., for thinner stock, or you can place a spacer under the stock . Fasten a piece of plywood or Masoniter to the bottom to act as a sled to support the workpiece. I tweaked my adjustable jig to near perfection (see next section) and then used it to cut the two notches in a board to make a sled type jig that slides on its own guides in both miter slots. Now I don’t need to fuss with adjusting the commercial jig every time I want to cut box joints. SETUP To set up the box joint jig: 1. For ½" sockets & pins, the overall width of (B + C + D) must equal 1.5" exactly. Distance C (the width of the finger) is adjustable by moving the jig to the left or right. My commercial BJJ has horizontal slots that I use both to mount the jig to the miter gauge and to adjust distance C. Once I got it adjusted to cut “perfect” joints, I slid a bolt into the slot, up against the miter guide, and tightened it in place on the jig to serve as a permanent reference point so that I would not need to readjust the jig every time I mounted it on the 237 Table of Contents 38-6 Boxes and Drawers miter guide. This helped a lot, but did not entirely eliminate the need for test cuts and fine adjustments. I mount a homemade BJJ, as in Figure 3, on an aluminum miter guide extension that has horizontal slots in it. If I did not have that slotted miter guide extension, I would have used a router to cut mounting slots into the BJJ, so that I could adjust it from side to side. 2. Set the saw blade height to just 1/32" or so more than the thickness of the stock. (Instead of measuring, lay a piece of stock beside the dado stack, resting on the jig support ledge, if any.) Consider making a cut in scrap as a template of the desired cut depth (e.g., ½"), and then reset the blade height with the template resting on top of the “support ledge,” if any. Adding the thickness of the support ledge to the desired cut depth is not accurate enough. Setting the cut depth just 1/32" high will make the pins proud of the sides . These can be planed or sanded down after the box is assembled and the glue has cured. Set the height of the pins as close as you can, however. Shaving away more than 1/32" can be pretty tedious. CUTTING 1.. You may need a backup board, e.g., c" Masonite, to reduce tear-out. Put the board on the jig after setting cut height. If the backup board is the same length as the jig, position the left end of the backup board about 1" beyond the left end of the jig and short about 1" from the right end. Make a cut. Move the resulting socket over the pin. Make another cut. Drill and countersink holes near the ends of the backup board to receive #10 flat head screws, and fix the backer board to the jig. 2. You must make test cuts, lest you spoil the work. The test cuts should be made on two scrap boards of approximately the same width as the joint you plan to make. The jig is not adjusted properly until the joint will fit together over the whole width of the joint you Figure 4. Simple Box Vise plan to make. The joints should dry-fit tightly enough to stay together but not so tightly that they must be driven together or apart. Assembling too-tight joints in plywood may de-laminate the fingers. 3. Finger width is adjusted by moving the whole jig relative to the saw blade. Move the jig in the direction of the pin to make the pins wider (tighter). The adjustment is critical, especially for wide boards. An error of 1/32" in pin width will accumulate to a 1/8" error at the end of an 8" wide board. However, if you start cutting each side at corresponding ends (i.e., ends that will mate when the joint is assembled) the errors may accumulate in the same direction so the joint will still fit. The adjustment is much finer than you can do by eye or by trial and error cutting test cuts on scrap. Use a dial gauge to measure 238 Table of Contents Notes and Reflections While Shaving Wood changes as you move the jig on the miter gauge. Alternatively, you can measure B+C+D with a vernier caliper & tweak the pin width. Note that B+C+D must equal 3*B. When you get it right, don’t forget to retighten the mounting screws or clamp. (Scratch a reference mark on the back of the jig for ½" pins, but you may still have to adjust the pin width.) Lateral play in the guide of the miter gauge will make a difference in the fit of the joint. I measure and cut with slight pressure to the left, to minimize this source of error. 4 For simplicity (it seems to me, anyway) plan to cut the long boards first, with a pin at the RE to hide the groove for the bottom. Cut the sockets for the first end “A” of the long board. Flip it 180E vertically. That should bring the RE at the other end “H” in position against the jig. Check to make sure that it will be cut symmetrically with the first end “A” (with a pin along the RE). 5. Cut ends “D” and “E” of the second long board in exactly the same way. 6. The short boards now must begin with a socket at the RE. You need a spacer. Either use the edge opposite the RE of a long board, a good test cut, or a template to space the cut over by exactly 1 pin-width from the RE. Cut end “B” of the short board 7. Flip the board vertically, and use the spacer, to cut the other end “C” of the short board so that it is symmetrical with “B”. 38-7 8. Cut ends “F” and “G” of the second short board. FIT UP AND GLUE- 1. PVA glue, e.g., Titebond II, will grab a tight box joint too quickly and may make final adjustments quite difficult. Figure 5. Drawer lock Bit (Lee Valley Drawer Lock Bit. Drawing Consider glue from www.leevalley.com) with a longer open time and less initial tack, e.g.,Titebond III, Titebond Extend, or hide glue. Brush the glue on the mating surfaces with a shortbristle brush dipped in glue. 2. If the pins were cut proud of the sides, set the clamp just inside the joint. Apply just enough clamp pressure to close the joint, as shown by a bit of glue squeezing out, or tap the joint closed with a rubber mallet. If you apply too much pressure you may bow the sides. If the joint still does not close, use a square wooden drift (e.g., just a piece of oak or maple ½" square) to pound each pin down. If the joint is tight, pounding them together is more important than clamping them, and the clamps can be moved if they are in the way. In any event, inspect each joint to be sure that the pins are seated well. The joint may be perfectly strong even if they are not, but the joint will look sloppy. Also, the pins will not protrude and you might have to plane the sides. 239 Table of Contents 38-8 Boxes and Drawers 3. Check diagonals and square the box. A well made box joint should fit together square but that depends on how well you closed each joint. Check the diagonal measurements. The box should need relatively light clamping to stay in position until the glue has set. sides if desired to leave clearance for drawer guides, final trimming, etc. To calculate the length of the stock needed for the drawer sides. Just add ½" to the desired length minus the thickness of the front and the back. As a formula, this is: 4. Make sure that the bottom is seated all along and that the sides or back are not bowed. L = Lo- T + ½ FINISH After the glue has dried completely, plane, scrape or sand any protruding pins flush with the sides. Figure 4 shows a simple box vise. It comprises two strong sticks clamped to the bench top, protruding beyond the edge by the depth of the box. One or both of the sticks has a simple clamp screwed to it to hold the drawer steady against the front of the bench. Drawer Lock Joint Cutting this joint in plywood tends to tear the ends of the board severely, even if you score the cut first with a knife. As shown in Figure 5, one setup on a router table is needed for both sides and front/back. The setup is fussy, however, and may require many test cuts. After that, the joints take little time to cut. The instructions provided with the Veritas Drawer Lock bit are clear.(They may be available on the Lee Valley web site even if you don’t buy their bit.) They include how to set up the router and how to diagnose the cause of various gaps in the fit of the joint. Your bit may require different directions. Here are some additional points. a. I prefer to make the drawer front to go across the full width of the drawer (i.e., not fit between the box sides). You can offset the where L = length of side stock (raw) Lo = desired overall length of drawer, including both ends T = thickness of front plus thickness of back Example: 15/32" front and back, then T = 15/16 b. Feed the stock through the router slowly and firmly because it may tend to tear out and also may tend to move away from the router fence. A high auxiliary fence and a specialized push shoe are very helpful for routing the side pieces. Make an “L” or a “U” by gluing two or three pieces of stock together. Any scrap will do, but the vertical piece must be slightly thinner than the stock.(See Figure 6.) c. If you get tear-out, use a shim between the fence (or table) and the workpiece for the first cut and then remove it to finish the cut. d. If the box bottom will be placed in grooves, you may need to clip off each corner of the bottom, because the corners will hold the Figure 6. Push Shoe for Routing Drawer Lock Joint 240 Table of Contents Notes and Reflections While Shaving Wood 38-9 joints apart if the grooves are ¼" deep or deeper. e. Despite the name, the joint does not lock together automatically on glue-up. Check carefully to make sure that the each joint is properly mated at each end. Again, be careful not to overclamp. You can bend the sides or even split them.. f. Check the diagonals for square Half-blind Dovetails I find it easier to design a half-blind dovetail (DT) joint for a drawer front by looking at the tails, although I cut the pins first. Half-blind pins are more difficult than tails to cut exactly, so I cut the pin board (drawer front) first & then fit the tails to the pins. The front and sides of a drawer usually have a groove cut in the front and sides, into which the bottom will slide. For a dovetail to hide that groove, a thin (¼" wide) half pin is required at the bottom of the drawer front. This may be strong enough in oak. If you want a wider half pin at the bottom, the groove should either stop ¼" from each end of the front or should be cut far enough from the bottom edge to allow space for a thicker half pin, e.g., d" from the bottom edge. 1. Decide on the approximate layout of the tails on the drawer sides first, or make a fullscale sketch. If you will cut the tails on a bandsaw, make the tails symmetrical across the width of the drawer side so you can just flip the tail board over. Be sure not to plan a pin that is narrower than your narrowest chisel. 2. Lay out the pin board (the drawer front). A. Set a mortise gauge to scribe a line Figure 7. Long Miter Jig (Shown on Shooting Board) along the end of the pin board. The line should be at least 3/16" from the front. You have to balance the strength of the finished joint against the strength of the pin board, to withstand the repeated chisel blows as you make the sockets. On a ¾" thick pin board, a 3/16" lap seems to be a good compromise. Set the gauge to measure from the back of the pin board. For example, if the pin board is ¾" thick, the gauge would be set at 9/16". Scribe the line. B. Set another mortise gauge to the exact thickness of the drawer side (tail board), or perhaps 1/32" more. Scribe a line across the back of the pin board. C. Lay out the pins with a dovetail gauge or sliding bevel gauge set to 1:6. D. Chop out the sockets. This can be slow and tedious. You can cut the sockets with a dovetail bit. Or, you 241 Table of Contents 38-10 Boxes and Drawers can hog out some of the sockets with a router or Forstner bit, and remove the final bits of stock with a chisel. (For detailed discussion of making half-blind DTs, see the Small Writing Desk project notes. Also see the article by Collins. 3. Lay out the tails to match the pin board. Precise cutting will be fruitless unless the layout accurately matches the pin board. Also, it is too easy to forget whether the marks correspond to the wide or narrow end of the sockets. (I know this from sad experience.) A. Scribe a line across the inside of the tail board using the mortise gauge set in step 2A above. B. Choose a consistent reference edge. I prefer the edge where the groove is cut for the bottom. Put the pin board in a vise or clamp it to the bench. Line up the two reference edges perfectly against a small block of wood. The wide ends of the sockets should be barely visible under the end of the tail board. Clamp the boards in place. (It is necessary to immobilize the two boards. They can easily move while you are marking the pins, and this will spoil the joint.) Mark the positions of the wide ends of each socket on the end of the tail board, preferably with a marking knife. C. Using the dovetail gauge or sliding bevel gauge, mark the lines for the tails back to the scribed line. D. Cut the tails by hand or with a bandsaw, using a taper jig (a wedge of scrap) run along the fence. Cut to the outside of the layout line so that the tails can be trimmed if necessary to give a good fit. If the drawer side is Baltic birch, you can chop out the pin on the drawer side with a few chisel blows. Just be sure to start on the inside of the drawer and then turn it over to finish chopping. Otherwise, you may tear the wood or delaminate the plywood as the chisel comes through. You can also cut out the waste with a coping saw or benchmounted jig saw. Be careful not to cut too far into the board. Finish up with a chisel. Miter Joint Cutting a long-grain miter joint exactly at 45E is difficult. Theoretically, you can tilt the saw blade on a table saw and achieve acceptable results. In fact, it is difficult to cut miter joints this way because the tilt of the blade has to be exactly 45E or the box will not close properly. For the long (ripping) cuts needed for most box dimensions, if the blade tilts toward the fence, a burn is likely unless a special auxiliary fence is used so that the cut piece can’t be trapped between the blade and the fence. In my opinion, this joint requires a purpose-made sled to cut accurate 45E angles. (See Figure 7.) The sled tilts the workpiece rather than the saw blade and slides along the rip fence or along a miter slot. It is described by Kirby in The Accurate Table Saw (pp 132 and 146). The workpiece must be clamped to the sled because it may tend to creep during cutting. A piece of coarse sandpaper glued to the sled will help to prevent the workpiece 242 Table of Contents Notes and Reflections While Shaving Wood from slipping. If you have only a few joints to make, say one box, it may not be worth the setup time, so you should choose another joint. Assembling a box with miter joints is challenging because the joints slide under clamp pressure. One way to hold them in place is to use corner molding as clamping cauls. Then clamp the box with band clamps, tourniquet, or rubber bands (for smaller boxes), Be sure to wax the inside of the corner molding and use waxed paper to prevent it from becoming glued to the box. Another way is to hold them together with painters tape. Lay out the box sides and ends in order, end to end, on the workbench. The outside surfaces of the box should face up. Make sure that the edges are aligned properly and tight against each other. Tape the pieces together across each joint. Use at least two pieces of tape for each joint. Wipe dust off the surfaces that will be taped, and don’t be stingy with tape, because it will be put under tension when you form the box. Flip the sides over so that the insides of the box now face up. Apply glue and roll the box into its final form. The taped joints will be put under tension, which will apply clamping pressure to each joint while holding it in position. Hold the last joint together with your hands and apply tape to it. Because a miter joint in solid wood is an endgrain glue joint, it may be a good idea to “size” the joint by applying plenty of glue and then waiting a few minutes to let it soak in to the end grain. Then apply more glue as necessary to make the joint slightly wet. Also, it will be very inconvenient to remove squeezeout from inside the box. Apply wax or masking tape to the edges along the miters that may receive squeezeout, being careful not to get any wax or tape onto the mating 38-11 surfaces. You can increase the pressure on the joints by wrapping more tape all the way around the box. Optionally, clamp small blocks to the outside corner of the last joint or use corner molding and then apply clamp pressure to the blocks, diagonally across the joint. Do not apply too much pressure or you will break the tape. After the glue has set the outside of the joint can be lightly chamfered with sandpaper. Splines glued lengthwise along the face of a miter joint will strengthen the joint and also make it much easier to glue up, because the spline prevents the faces of the joints from sliding under clamp pressure.1 After the miters are cut and trimmed so that they fit properly, you place the mitered side (hypotenuse of the miter) down on the table saw and, using the sled, score shallow grooves for the splines on each miter face. The grooves should be a bit off-center toward the inside of the corner, so as not to weaken the outside of the corner. The ends of the splines will show at the ends of the joints (top and bottom of the box corners), so they should fit well. You can use a contrasting color of wood to make them a design element. Long-grain splines are easier to insert and will usually be strong enough. Short-grain splines (cut off the end of a board) are fragile before being inserted in the dado but are very strong across the joint. Cut the splines from a piece of stock that is large enough to handle safely. One safe way to cut off small pieces is to score a larger piece of stock (cut it part way through) for the width of the key, and then cut the pieces away 1 A long-grain miter is technically a bevel but I doubt that this the distinction would help us much here 243 Table of Contents 38-12 Boxes and Drawers from the stock for the thickness of the key. The cutoff pieces should be on the side of the blade away from the fence, so that they will fall away and not be kicked back by the spinning blade. For example, for a small box made with ¼" thick sides, I cut grooves in the miters that were 3/32" deep by c" wide. (The width of a standard table saw blade kerf.) Then I needed keys that were 4" long, c" thick (to fit in the saw kerf) and 3/16" wide. I took a piece of stock that was 12" long and about 1" square in cross section. I scored it so that the ridges were 3/16" wide and then cut off c" wide pieces. I then sanded the thickness until the sticks fit easily in the grooves, and finally cut the resulting c x 3/16 x 12" sticks into 4 splines. A lock miter joint is similar to a splined miter joint, except the spline is an integral part of one side of the joint. These are cut with a specialized router bit on a router table. (See Chapter 22.) You can also cut keys across a miter joint after the glue is set up. These strengthen the joint somewhat and are decorative. They are sometimes called false box joints. Regardless of the finish used on the drawer front and the outside of the case, shellac is by far the best finish for the drawer sides, back, and bottom. Shellac does not smell bad, and it will dry hard faster than varnish, so it has much less tendency to stick to other surfaces. (A well-fit drawer can stick to its guides for some time after being varnished.) Also, after light sanding, shellac has a smooth, silky feel. I usually apply wax along the bearing surfaces on the bottom and sides after the shellac has hardened. References Collins Janet A , Making Traditional Dovetailed Drawers, Fine Woodworking July/August 2002) Frid, Tage. Tage Frid Teaches Woodworking Book 1: Joinery. Newtown, CT. Taunton Press 1993. Rogowski, G. Router Joinery, Newtown, CT. The Taunton Press 2003 (book and companion video). Finishing Drawers 244 Table of Contents Notes and Reflections While Shaving Wood Chapter 40 – Fitting Drawers and Cabinet Doors F are more demanding because the front must be in the same plane as the front of the cabinet, itting drawers is a three-dimensional exercise. It is much easier if the drawer is true. Still, even with a perfectly true drawer, the solution to binding near the front may lie at the back of the drawer, and vise versa. If you can achieve a good sturdy dry-assembly of the piece, it may be easier to identify and access the part of the drawer guide that needs to be adjusted. At least, fit the drawer before the top and/or the back is attached. The procedure for fitting cabinet doors is essentially the same as for fitting drawer fronts. Most of the discussion of fitting drawer fronts will apply as well to fitting cabinet doors. Two basic types of drawer fronts and cabinet doors interest us here: inset and overlay. An inset drawer or door fits inside the opening. It may show some of the frame, usually ½" or less. If the drawer front, face frame, rails, etc., are in the same vertical plane, the style is called a flush front. (Figure 1) All four edges of a an inset front must be parallel to the plane of the front of the case, especially for flush fronts. Otherwise, the front will look cocked when fully closed. This means that the drawer box must be square in 3 dimensions and that the drawer guides must be perfectly aligned. An overlay front covers some of the face frame. An overlay front hides the drawer opening, and usually covers about half of the face frame, rails, etc. (Figure 2) Overlay fronts are easier to fit, although the fronts must look symmetrical and line up well if there are columns or rows of drawers. I prefer to make drawers with flush fronts or inset fronts. (Figure 1) Both must be fitted precisely into the opening. Flush fronts Figure 1 Flush Drawers (Above) and Inset Drawers (Below) which means that the drawer must be foursquare in the opening. Also, any inconsistencies in the gap (reveal) around the drawer front will be apparent. How to Fit the Drawer Successfully fitting inset drawer fronts, false fronts, or doors must begin before the fronts are cut. Check the corners of the opening with Figure 2 Flush Drawers (Above) and Overlay Drawer (Below) 40-1 Table of Contents 245 40-2 Boxes and Drawers a square. When you measure each side of the opening, include any additional dimension needed to allow for a lack of squareness. For example, in Figure 3, the opening (solid line) is not square, but it is a parallelogram. So, suppose that you measure both the top and the bottom length and they are both A inches long. So, if you then cut the front square, and exactly A inches long, the drawer will be too short by B-A inches after you trim it for a consistent gap, and you will have to cut a new one. The nominal length of the drawer front should be B. You will want a gap around a solid wood inset or flush front to allow for wood expansion and contraction. For example, a 6" wide, white oak board at 80% relative humidity (RH) may shrink to 5¾" when the humidity falls to 30%.2 Suppose you want a 1/16" gap all around. Then each dimension of the front will need to be c" less than the corresponding nominal dimension, e.g., (B - c "). It is better to cut B- 1/16 ", to leave a 1/16" margin of safety. You will have to trim off any extra, so don’t make the fudge factor too large. STEPS 1. Cut the drawer front a bit over-size as described above. It is better to cut the sides square. That is, don’t try to copy the angles in the drawer opening at this stage. If you are fitting more than one drawer, mark each drawer with chalk so you won’t get them mixed up. (If you are fitting a false front, mark the back top edge as well so that you will orient it consistently while fitting 2 For example, see http://www.woodbin.com/calcs/shrinkulator. htm Figure 3. Make a drawer front or a door large enough to trim to actual opening. Solid line - drawer opening. Dotted line - rough drawer front dimensions. it.) a. Slide the drawer into the opening. If you followed the suggestions above, the front may just fit or may be 1/16" too big to fit into the opening. If the opening is not square, the drawer front will catch at the corners. (See Figure 4.) If the drawer will not fit into the opening enough to check the fit, use a sliding bevel to transfer the angle of the opening to the drawer front. b. Depending on how many right angles you have in the drawer opening, you can trim the front to fit in 1-4 cuts. (Figure 4 shows the worst case. It will take one cut on each edge.) Until you are proficient, however, it is safer to sneak up on the fit in two stages. Figure 4. Irregular Drawer Opening. (Opening, dotted line. Drawer front, solid line.) 246 Table of Contents Notes and Reflections While Shaving Wood 40-3 2. Choose one corner as your reference. Trim the height first. For a 1/16" gap at the top and bottom, rest the drawer front on a c" thick shim on the bottom of the drawer opening. Mark the amount to trim off at that end. Repeat for the height of the opening at the other end. (For frame and panel doors, you should trim equal amounts from the top and bottom so that both rails remain about the same width, so cut half the distance.) Continue each mark around the edge to the back face of the drawer front. 3. Draw a guideline between the marks. If you have only 1/16" or less to trim off of the top and bottom, use a hand plane and trim it until the line just disappears. Take mediumlight cuts and pay attention to grain direction. Take extra cuts on the “wide” end, but finish it up with a full-length stroke so the edge will be straight. Check it frequently against the opening. If you have a lot to remove, use the table saw. a. Assuming that the rip fence is aligned correctly with the saw blade, slide the rip fence against the right side of the saw blade. Make a sharp pencil or scratch mark on the saw table, near the front, along the rip fence. b. Slide the fence to the right (I mean, away from the blade) and line up one end of the workpiece with the saw blade. c. Using a thin wooden spacer or a taper jig, rotate the workpiece so that the pencil marks on the drawer front line up with the blade and the scratch mark. (See Figure 5). d. Trim the drawer front. Figure 5. Trimming the Long Dimension (Exaggerated) Note that the width of the spacer equals the gap you measured between the drawer and the opening. front and mark the bottom corners for height. Repeat step 3. 5. After the front is trimmed to the correct height, mark the four corners for length as in step 2 above. Use a bevel gauge to copy the angle from the opening to the drawer front. The angles should pass though the marks you made on the drawer front. It is much easier to trim end grain (on a solid wood front) on the table saw. You need a mark near the front of the table that corresponds to the left side of the saw kerf. You can plane it if you have a very sharp lowangle block plane and a means to prevent tearout at the ends. 6. Rotate the workpiece until the trim line is parallel to the saw blade. Use a shim or rotate the miter gauge. (See Figure 6) Trim conservatively and check it often against the drawer opening. 4. For frame and panel construction, you will have trimmed only half of the necessary amount. Lay a 1/16" shim on top of the drawer 247 Table of Contents 40-4 Boxes and Drawers FINAL FITTING to hold the FF securely. Attaching False Drawer Fronts 6. Predrill from the back (using a depth stop (!)) and screw the FF to the drawer front with at least one screw. I usually use two so the FF is held in place independently of the drawer hardware. 1. Pre-drill the false front (FF) for the hardware, e.g., knob/drawer pull. It is easier to get the holes precise when it is off the drawer. 2. Locate the FF where you want it. For inset drawers I use a few shims at the bottom and at the sides of the case opening. You could probably use shims for overlay fronts also. Carpet tape also sounds like a good idea for overlay fronts if the tape isn't too strong This allows you to remove/replace drawer hardware whenever that may become necessary. 3. When the FF is where you want it, mark through one or two holes in the FF to the drawer front, remove the FF and drill the drawer front. (If the drawer is secure, I might just drill through the hole in the FF. Otherwise, you can make the hole in the drawer front a bit oversize to leave a bit of wiggle room.) 4. Install the drawer hardware snug but not tight. Figure 6. Trimming the Ends 5. Recheck the location of the FF, adjust it, re-center it, etc and then tighten the hardware 248 Table of Contents Notes and Reflections While Shaving Wood Chapter 41 – Installing Quadrant Hinges Q uadrant hinges limit the opening of a hinged box lid to about 95-100E. They are an interesting design that will dress up a decorative box. The Brusso #HD-638 (1" x 5/16") is an excellent hinge. You can’t use this hinge, however, if the top is thinner than about ½". (Actually, d" is the minimum clearance for the quadrant link). Brusso also makes a larger version (#HD-680 1½" x ½") and a brass template for cutting the mortises for the #680 hinge. Unless you expect to install dozens of these hinges a template is Figure 1./ Brusso probably not worth Small Quadrant Hinge buying. Necessary tools include a small inlay knife or marking knife and a router with a ¼" straight bit (same diameter as the width of the forward hinge leg) or router plane. Also, before you start, have a few #4 x ½" F.H. steel screws on hand to tap the screw holes for the brass screws. protrude very much beyond the edge of the box, and you must place the hinge far enough forward so that the rear screws can get into solid wood without splitting it. (See below). Make certain that the top, bottom, and hinge knuckle are aligned precisely, or the hinges may bind after assembly. If possible, clamp the hinges in place. Mark the centers of the screw holes with a centering punch and drill the pilot holes (1/16" bit). Slant the pilot holes in the box toward the inside very slightly to avoid splitting out the back when you insert the screws. To be on the safe side, clamp a block of wood to the back to hold the wood in place as you insert the screws. You can screw the hinges to the surfaces of the top and bottom to hold them in place. (Use the #4 steel screws.) Use a marking knife or X-acto knife to scribe around the edges of the hinges. Scribe a shallow line at first and then deepen it with successive cuts. This is to prevent tear-out from the router. Mark the slot for the quadrant link with a fine pencil. Mark Hinge Locations Cut the main hinge mortises with an edge guide and a ¼" straight bit. A lighter router like a trim router or Dremel tool in a base is preferable. A small router plane is also excellent for this. Since the mortises are open to the back, you do not need a plunge router. Because of the shallow depth of cut a powered router should be easy to control, but remember to cut in the correct (counterclockwise) direction. (When the router edge guide is on the right, push the router away from you.) The depth of the hinge mortise Cut Hinge Mortises It will go better if you take the time to set up a small platform under the lid to hold it in the “open” position (face down) and in the same plane as the top of the box. Use angle braces as legs, screwed to a piece of scrap for the top of the platform, or just tack legs on the edges of a piece of scrap.) Line up the sides of the box and the lid with a straightedge. Lay the hinge (without the quadrant link) in position on the top and bottom of the box. It is not necessary that the knuckle 41-1 Table of Contents 249 Installing Quadrant Hinges (pocket) determines the amount of gap between the lid and box. The deeper the pocket, the smaller the gap; however, if the mortise is too deep, the lid will not close. Use a plunge router with a c" straight bit and an edge guide to make the cutout in the lid and box for the quadrant link to drop into. Extend the scribed lines so that they will not be routed away, and cut to the lines (about ½" long). The quadrant link requires about 1e" total depth, but at least d" depth in both top and bottom for the ends of the quadrant link to fit into. If the top of the box is thick enough, cut the mortises about equal depth (13/16 - f") on both parts. Be careful not to cut the mortise for the quadrant link too long, in particular do not cut away the wood needed to hold the front mounting screw. If you do cut too far, about all you can do is glue in a dowel of the same diameter as the groove. This repair won’t show when the hinge is on. Assemble Hinge and Install. The quadrant link is not quite symmetrical. One end has a deeper notch than the other. You assemble the hinge by passing the smaller notch through the slots. It is necessary for the deeper notch to be at the top. It is best to attach the assembled hinge to the box first, and then to attach the top to the hinge. Otherwise, the top will be in the way as you try to insert the screws. You will be “tempted” to slant the screws toward the back, where they may split out, and you may tend to cam out the brass heads. It is best to use the steel screws to 41-2 attach the hinges, then to remove them one by one, replacing them with brass screws as you go along. The heads of the brass screws are easy to cam out and the brass screws can easily break if you try to tap them into hard wood. Also a bit of candle wax or soap on the threads will make this easier. If a brass screw breaks, use a screw extractor or drill a new hole on the drill press, right down into the Figure 2. Framework for Holding Lid shank of the broken brass screw. Repair the hole with a hardwood toothpick and some Elmer’s glue to hold a new screw. To attach the top, you can rest it on the temporary platform, but it is easier to make a small “H” frame to hold it in place while you attach it. Attach a cleat to two upright pieces. Put the top on the cleat and clamp it in place. (Figure 2) See also, http://forum.canadianwoodworking.com/sho wthread.php?t=14711 http://www.whitechapel-ltd.com/tech/quadra nt_hinges.shtml 250 Table of Contents Notes and Reflections While Shaving Wood Chapter 42 – Turning Wood W Choice of lathe should depend in large part on what you want to make.1 I am offering these comments because lathes, like other power tools, are subject to the delusion that bigger is better. I rarely turn objects even as large as 12" diameter and would have wasted money and space on a larger lathe. oodturning is a specialized craft with its own distinct concepts, facts, and skills. Even someone who is already competent in “flat” woodworking might need to study and practice patiently for hours just to learn the fundamentals. (Some useful references are listed below. Videos are especially helpful.) I once thought of turning as making spindles and bowls from solid wood. Apparently, that’s what most turners do. But the range of turning is much wider. With its variety of shapes and methods of preparing turning blanks, turning can be a source of challenge, accomplishment, and satisfaction for years. Even a brief look at the work of people like Malcom Tibbetts will convince you of the potential and challenges. This chapter will summarize the main points and fill in some gaps that I found in my readings. A comprehensive treatment would require at least one additional full-length book. (See References) I learned to turn, and wrote the early drafts of this chapter, with a vintage (late 1930's) Craftsman lathe that might today be classed as a midi. It swings 9" (diameter) over the bed and will hold about 30" between centers. It has a ¾" x 16 tpi spindle and takes a #1 Morse taper. This lathe was completely suitable for learning this craft, and well beyond. I added a bed extension made of oak so that I could turn longer chair legs. Later, I replaced it with a Delta 46-460 with a bed extension because I wanted a larger swing (12"), the choice of a slower speed (down to 200 rpm), and easily variable speed. Both of these lathes are heavy cast iron with durable bearings. Both are versatile enough for general woodworking. Setting Up to Work Lathe setup is important. The lathe must be rigid and heavy. Cast iron is greatly preferable to steel or other materials for the bed, headstock and tailstock. A strong, solid bench or stand is also necessary, and the bench has to be on a solid foundation, like a concrete floor; some turners even load the table with sandbags. The table for my medium-sized lathe has a 2" thick by x 24" wide yellow pine top, a steel frame, and a modified torsion box shelf loaded with blanks, tools and accessories. It is heavy and stiff enough for most of my turning. (Because it is up against a wall, I wish the table were a few inches deeper than 24" so that I could maneuver inside bowls more easily.) For spindle turning, the axis of rotation should be about elbow height. For bowl turning, it should be at least 1-2" above elbow height. Exact height is not critical, but if the centers are too low, you will bend over a lot when hollowing vessels. If you need to raise the lathe, screw a straight 2 x 4 or 4 x 4 to the bottoms of the lathe table legs. 1 See, for example, Raffan’s book, below, or http://en.wikipedia.org/wiki/Lathe . 42-1 Table of Contents 251 42-2 Turning Wood The two lathe centers must be aligned. Otherwise, spindles intended to be cylindrical may tend to be slightly tapered. Ends intended to be trued up square may not be, and if a spindle has to be re-mounted on the lathe it may act as if it were elliptical. High capacity dust collection and good lighting are essential. TOOLS Necessary tools to get started in turning will depend on what you want to do. High speed steel (HSS) tools hold an edge longer than tools made of high-carbon steel, and are less subject to damage if they overheat a bit while being sharpened. HSS chisels are reasonably affordable. Some elite bowl gouges, however, are beyond the needs of a beginner or occasional turner. I’m thinking of models named after a famous turner – not the brand name but the model name , e.g. the “Hepler Grind” model of a bowl gouge. Fancy carbide scrapers are also relatively expensive. Carbide scrapers are actually substitutes for chisels. Learning to use them is easy but they do not teach you how to present a chisel to the work. Expensive tools are even more expensive if you are also learning to sharpen them, because you will grind away lots of that expensive metal. Chisels for turning spindles can be shorter and lighter than chisels for turning bowls. The minimum set for spindle turning would be a ¾" to f" roughing gouge, a ¾" spindle gouge, a ½"-¾" square scraper, a 3/16" diamond-shaped parting tool or extra-thin (c") parting tool and a 1" skew. For turning bowls, add a d" - ½" bowl gouge and a round nose scraper, ¾" or larger. A bowl gouge has a deeper flute, is longer, and is made of heavier stock than a corresponding spindle gouge. The difference between bowl and spindle gouges was not clear to me until I bought a bowl gouge. It became apparent when I actually compared them in hand. I swap chisels often while I am turning, so where and how I keep them becomes important. After many trials, I built a rack on the wall over my lathe table. The handles rest on a narrow shelf and the blades rest against magnetic knife holders from a kitchen store. The rack has two levels. A spindle gouge may be around 11", while a “long and strong” bowl gouge may be 20" or more. Keeping chisels behind the lathe, as I do, has a potential danger. If you have to lean over the lathe to reach a tool your clothing could get caught by the spinning lathe and you could be pulled down into the lathe. This could be a serious, even fatal event. You should never turn while wearing hair or clothing that could get caught by the spinning lathe. Some turners advise against using a magnetic holder because they say that it magnetizes the chisel and makes it cling to the tool rest. I have not noticed this problem. Mounting the Blank A secure mount is one that holds the workpiece securely, is able to survive a catch, and yet is easily separated from the finished turning. Mounts for center turning (spindles) are much more straightforward than mounts for face work (bowls, etc). CENTER TURNING Spindles are held between centers: a drive center (usually, a spur) at the headstock and a 252 Table of Contents Notes and Reflections While Shaving Wood cone or cup center at the tailstock. A new lathe would usually ship with these, but they may be missing if you buy a used lathe. A live tail center has a bearing and rotates with the work. It’s a nice accessory, because a dead center will burn the work slightly. (Burning can be reduced somewhat with a bit of wax on the tail center.) Sometimes a spindle can be clamped in a chuck. FACE TURNING Face turnings can be held by a variety of methods. I feel that the minimum is a screw chuck2 or faceplate and a four-jaw scroll chuck. A small-diameter (3-4") faceplate with a screw chuck and two additional screws will securely hold smaller bowls (up to 8" or so) but will leave screw holes in the turned bowl. It is always a good idea to mark the position of the chuck or glue block and the workpiece, just in case the work has to be remounted. Attach what will be the mouth of the bowl to the screw chuck. The screw holes from the screw chuck will be removed when you hollow out the bowl. After you turn the profile, turn a tenon on the bottom. (See the next section.) Then dismount the blank and secure the tenon in the four-jaw scroll chuck. This mount is strong, 2 Commercial screw chucks are not very expensive. You will need a faceplate anyway. To make your own, screw a 1" thick disk of hard wood to a small (2"-4" diameter) faceplate. put it on the headstock, and mark the center of the disk with the tail center. Then remove it from the faceplate and drill a hole through it at the mark, as close to perpendicular as you can get (use a drill press). Drive a 1¾" #10 or larger Phillips or square drive wood screw through the hole. Screw the disk to the faceplate with 2-3 screws and drill 2-3 holes through the hardwood disk for additional screws into the workpiece.. 42-3 especially if you turn a slight dovetail on the tenon for the scroll chuck jaws to grip. Sizing a Mounting Tenon or Mortise. It is important to fit the diameter of the tenon to the jaws on the scroll chuck that you will use. You want maximum contact. The jaws make a circle only when they are closed, so the tenon should be slightly larger than the minimum size of the jaws. (How much larger depends on how compressible the tenon is plus a margin for error.) A too-large tenon would be held only at the ends of each jaw section, while the jaws won’t tighten enough on a too-small tenon. A similar concern applies if the jaws will fit inside a mortise on the workpiece. For example, using Barracuda #3 jaws a well-fitting tenon would be about 3¼" OD and 3¾" for a mortise gripped from the inside. For the Oneway #2 jaws that’s about 1¾ and 2¼". Use a calipers or make a gauge (template) that reflects both outside and inside dimensions of the chuck when closed or nearly closed. The length of the tenon should be slightly less than the depth of the jaws, so that the jaws will contact the shoulder of the tenon. The end of the tenon should not contact the face of the jaws. Cut a slight dovetail on the tenon, so that pressure from the jaws will push the jaws against the shoulder of the tenon Glue Block If a tenon is impractical, the next choice would be to glue the blank on a mounting block screwed to a faceplate or held in the scroll chuck. This situation may come up, for 253 Table of Contents 42-4 Turning Wood example, if the available stock for the base of a segmented blank is too thin. The mounting block and the bottom of the blank must be trued and flat, or slightly concave. Note that a slight convexity often results when you try to true up a solid surface. It’s much easier to make a very shallow concavity. The important thing is that there be enough well-mated glue surface on the bottom of the turning blank and the mounting block. It is very important that the glue mount be secure and durable. If a catch knocks the blank off the mounting, the blank may be damaged. At best, you probably will not be able to remount it on the same center as before. The blank will have to be turned round again, which may require more wood than you have left. Turn a block of hardwood into an appropriately sized disk (e.g., 3 ¼" x ¾"). Wood that can split longitudinally, like plywood or pine, is not a good choice. If the glue block will be attached to a scroll chuck, follow the advice in the preceding section. If it will be attached to a faceplate, cut a slight concavity on both sides of the block. While the block is on the lathe mark the center with the tailstock. Drill a small hole through the center of the glue block to receive a piece of stiff wire. Make a corresponding hole in the center of the workpiece. Put CA glue on the glue block and accelerator on the workpiece. Line them up with a piece of stiff wire and push the pieces together. Choice of Adhesive. There are five general approaches for mounting a workpiece to a glue block. 1. Cyanoacrylate (CA) glue is the quickest. Use either thick or medium. Thin CA glue tends to soak into the wood too much. This method can be ready to go in 2-5 minutes, especially if you use accelerator. I have had mountings with CA glue fail under some circumstances. 2. White or yellow PVA glue is strong and reliable but takes a while to cure. It’s probably best to let the glue dry overnight. Put brown paper between the mounting block and the blank to make sure that the pieces separate properly when you are finished. 3. Quick setting epoxy is quick, but a bit messy and inconvenient. Again, use brown paper in the joint. 4. A good double face tape (turners tape) if you are not turning too large a piece. 5. Some people recommend hot melt glue. I have not given it a fair trial, but the joint seems quick and strong. You can sometimes tap on the mounting block to break a CA glue bond. Epoxy and PVA glue mounting normally have to be cut off with a parting tool or a saw, after which you reverse the mount and finish the bottom. Cole, Longworth and Jam Cole jaws (large flat jaws, finger jaws) fit on a scroll chuck base. They are extremely useful for holding the “top” of a turned vessel while you finish the bottom. They are normally not strong enough to hold a solid blank for roughing, and the fingers or buttons have to be remounted frequently for different size bowls. A Longworth chuck is a variation on finger jaws. It tightens against the workpiece by means of continuously adjustable fingers. These require much less fiddling with buttons than Cole jaws but can’t be tightedned as securely. Some turners use a home-made jam chuck to hold a piece of face work while they 254 Table of Contents Notes and Reflections While Shaving Wood work on the profile. A jam chuck is a friction drive. For example, a cylinder with one end turned to a cone can be jammed into a bowl tightly enough to act as a drive mounting. The angle of the cone should approximately match the shape of the blank, because you want a fair amount of contact between the cone and the blank. The jam chuck is attached to a faceplate or held in scroll jaws. Use a piece of non-skid rubber foam (or double sided tape – something to increase friction). Squeeze the blank against the jam chuck with the tailstock. Rowley describes how to make a jam chuck on p. 100 of his book. Make sure that it’s secure. Would the bowl remain mounted after a bad catch? In most uses, a jam chuck is on the headstock. Occasionally, I want to re-mount a bowl after the tenon has been cut off. I can hold the wide end of the bowl in Cole jaws or a Longworth chuck, but I hold the bottom of the bowl in a sort of jam chuck made of an open four jaw chuck (3-4" for example) or even a faceplate padded with rubber foam. I have also glued a ¾" oak dowel about ½" into the bottom and held it in smaller chuck jaws. Attachments/ Accessories Turning has its own set of “after-market” extras. The most important are ways of mounting the work, described above. A long tool rest will reduce the number of times you have to move the tool rest along the bed to turn a long spindle. A curved tool rest is very useful for turning bowls. A ½" Jacobs chuck (a normal ½" drill chuck) with a Morse taper, stuck in the tailstock, is very handy for drilling. A steady rest, either commercial or home-made, is useful to keep long, thin spindles from vibrating. 42-5 Turning bowls and other hollow forms requires a good compass (heavier than the kind used for drawing on paper) and calipers for checking the wall thickness. Deep and odd shapes require the “figure-8” shaped kind of caliper. Depending on the lathe model, and how it is offered for sale, a bed extension may be offered as an accessory. I needed a bed extension when I wanted to turn some ladder chair posts. Given the importance of cast iron lathe beds and precise machining, I was skeptical about whether a home-made, wooden bed extension would be satisfactory, but it was completely so. Its main shortcoming is that it is minimally adjustable – the tailstock has to be bolted at fixed positions. Wood – Turning Blanks Perhaps the first decision concerning the choice of wood species depends on the purpose of the turning. A table leg, for example, must be utilitarian, i.e., strong, and probably should be the same species as the rest of the table. Although the turned leg should be attractive, its main requirement is not decorative. For other turnings, e.g., bowls and lamp standards, decorative value may equal or trump utility. Almost any species can be turned successfully. In general, finer-grained wood is easier to turn than coarse-grained wood. Poplar is a good wood to learn with, but does not scrape smooth (must be sanded). Most fruitwoods turn well, e.g., apple, plum, pear, Bradford pear. Cherry is easy to turn, smells good, is attractive, and is easy on tools. Elm, ash, beech and magnolia are recommended as excellent woods for turning. Walnut and mahogany turn well, 255 Table of Contents 42-6 Turning Wood although older mahogany can splinter and tends to powder rather than produce shavings. Some woods turn nicely even if they can be quite challenging to plane without tear-out, for example, maple. Oak is hard and coarsegrained but turns OK. Softwoods, like pine, are easy to cut with a chisel, but prone to tear-out. I have had good results turning clear Radiata pine. For decorative work, another decision is whether to turn seasoned wood or green wood. Green wood turns much more easily, but may dry unevenly and distort or crack the turning. Of course, this can be part of the design. Solid wood blanks are sometimes not available in the desired size, or not appropriate for the job. Blanks can be glued up in a number of ways, to produce a variety of visual effects. The simplest glued-up blank comprises well jointed boards glued together. Many other types are possible. These are discussed further below under Gluing Up a Large Spindle. See also the next chapter, 43, Blanks for Segmented Turnings. Safety Turning presents its own set of safety issues. Most of these involve the wood flying to pieces and striking you or a bystander, clothing or fingers getting stuck between the work and the tool rest, or the turner being struck or sliced by the rotating lathe or workpiece. 1. EYE PROTECTION IS MANDATORY. Even if you wear specs, wear safety goggles. Most books recommend a full face shield. 2. AVOID LOOSE HAIR, CLOTHING, JEWELRY. Tuck in your shirt, beware of loose sleeve cuffs, and put long hair up under a cap. A female graduate student died recently because her hair got caught up in a lathe. Do not locate tools, etc. so that you have to lean over the lathe to reach them. (Reach, maybe. Lean, no.) 3. BE ABLE TO TURN OFF THE LATHE QUICKLY, e.g., install a paddle switch on the front of the table in a place that you can press with your knee. 4. ROTATE THE WORK THROUGH ONE FULL ROTATION EACH TIME BEFORE TURNING ON THE POWER. Make this a habit. 5. REMOVE THE CHUCK KEY FROM THE CHUCK UNLESS YOU ARE ACTUALLY USING IT. In other words, the only time you take your hand off the key is after you have stored it where it belongs. Make this a habit, also. A flying chuck key is a dangerous missile. 6. KEEP THE FLOOR AS CLEAN AS POSSIBLE. Shavings on smooth concrete are slippery. 7. CONTROL DUST and wear a dust mask if necessary. Many bowl blanks are exotic woods that cause allergies. Sanding on a lathe creates fine dust and propels it into the air. Set up a dust collector and shop vac before you start turning. Hold the shop vac nozzle near the work when sanding to collect the sanding dust at its source. 8. CONTROL THE SPEED a. check the speed of the lathe before turning on the power. b. stand off to the side (out of the line of fire) when you start the lathe and whenever possible while turning. c. Whenever the workpiece has a large diameter (more than 3-4") or is glued256 Table of Contents Notes and Reflections While Shaving Wood up, rough the shape into a cylinder at slow speed. D. Use a lathe speed compatible with the size, weight, and length of the wood to be turned. See Table 1 for recommended approximations for balanced blanks. If a blank is unbalanced because it is irregular or drier on one side than the other, slower speeds should be used than those indicated in the table (about half). Regarding the speed of the lathe, my advice to beginners is to turn at slow lathe speeds, e.g., 500 - 700 rpm. Higher speeds can conceal improper technique until you “cross the line,” when really unpleasant things may happen. The surface velocity of a turning triples for every unit increase in diameter (C=πd). Centrifugal force, the force pulling the blank apart, increases as the square of the velocity (F=mv2). A large, slightly unbalanced turning blank rotating at high RPM can substantially stress the workpiece and the lathe. It can come loose and smash into you or bystanders. (See Rowley’s First Law of Woodturning.) It is possible to turn too slowly. Hard and irregular wood like a dried cherry burl I recently turned, required about 700 rpm before my chisels would cut well. Unfortunately, one of the blanks was about 12" in diameter and not that well balanced, so I had to spin it at 200-400 rpm. The gouges just seemed to bounce off of the high spots (chatter). Another, smaller blank of the same wood turned much more easily at 800 rpm. 42-7 General Technique The books and videos that I found most useful are listed at the end of this article. Among these, the discussion by Frid and the booklet published by Sears, Roebuck are the most concise. The books and DVD by Raffan and Rowley go into more detail. 1. Presentation of the chisel to the wood is a major issue in successful, enjoyable turning. The subject is described at length in books and videos, but it is, finally, a skill that can be acquired only by much trial and error. Presentation of a gouge to a spindle turning is summarized in Figure 1. (See Item 8, below.) The A-B-C sequence must be followed, very deliberately at first, and the chisel kept well away from the work at other times. Don’t let the chisel wave around in the air as you change speed, etc. a. Address the work by putting the chisel on the tool rest with the handle depressed. The tool must be on the rest before it touches the spinning workpiece, and must remain so whenever the tool is in contact with the wood. (Rowley’s Second Law) b. Hold the tool on the rest with your off (left) hand and lift the handle with your other hand until the bevel rubs the work. The bevel (grinding angle) of the chisel must rub the wood behind the cut (Rowley’s Third Law). 257 Table of Contents 42-8 Turning Wood Figure 1. Presentation of Turning Gouge (From The Wood Lathe – An Illustrated Manual). Table 1. Recommended Workpiece R.P.M. (Assumes sound, balanced, evenly grained stock) Workpiece Diameter Under 2" 2-4" 4-6" 6-8" 8-10" Over 10" Roughing 900 - 1300 600 - 1000 600 - 800 400 - 600 300 - 400 300 Shaping 2400 - 2800 1800 - 2400 1200 - 1800 800 - 1200 600 - 800 300 - 600 Finishing 3000 - 4000 2400 - 3000 1800 - 2400 1200 - 1800 900 - 1200 600 - 900 The “target” is inches per second of surface travel. So desired rpm is proportional to circumference (πD). Slower rotation is necessary for larger gouges (deeper cuts), unbalanced pieces and for longer spindles, depending on whether a steady rest is used. For an unsupported spindle over 18" long or more than 4" in diameter, 700-900 rpm is recommended. Table 2. Pulley Combinations for a 1725 RPM Drive Motor Lathe RPM Motor Pulley Lathe Pulley 6900 4" 1" 3450 3" 1½" 1380 2" 2½ " 575 1" 3" The ratio of rpm equals the ratio of corresponding pulley diameters rpm1/rpm2 = d1/d2 258 Table of Contents Notes and Reflections While Shaving Wood c. Keep lifting the handle until the chisel starts to cut. This is much easier to learn with a sharp gouge. The skew is the most difficult type of chisel to master. The skew is presented so that the middle of the cutting edge will touch the round turning. It also is guided by rubbing the bevel against the work, but the handle of the chisel is rotated outward slightly until the tool begins to cut. If the top of the cutting edge touches the work, it will probably catch. Cutting action in spindle turning is a sort of peeling action, similar to planing wood. Cutting across the long grain (as in face turning) is fundamentally different – it can be done with a gouge or a scraper, but it is a scraping or shearing action. If a gouge is used, the gouge is turned on its side so that the flute faces inward toward the axis of rotation. 2. Until you are well experienced, keep the tool rest set slightly above the center of rotation, and ½" from the workpiece. The cutting edge of the chisel should not need to protrude more than about ½" beyond the tool rest before it reaches the workpiece. 3. Sharp tools are essential. Dull tools will catch when they should cut, i.e., when you are doing everything else correctly. It’s a bit like using hand planes. You may think that your turning technique is wrong when actually, it’s your sharpening technique. (See Chapter 26, Sharpening Hand and Turning Tools, Part II) 4. The only part of the tool that should be in contact with the wood is the cutting edge vertically aligned with the tool rest. That is, if you would have to keep the tool from spinning by holding on to the handle of the chisel you would have violated this rule. This 42-9 is Rowley’s Fourth Law of turning. This was difficult for me to understand and difficult to do. It was, however, the key to proper tool presentation after I mastered the ABC sequence. 5. Always cut “downhill” or with the grain. (Rowley’s Fifth Law) This rule is applied by cutting from thicker to thinner on a spindle, from smaller to larger diameter on the profile (outside) of a bowl and from larger to smaller diameter inside of a bowl. This is easy to remember if you understand the rationale. You always want wood behind the pressure point. Otherwise, the wood may splinter (tear) rather than cut. With a spindle, the grain runs parallel to the axis of rotation and cutting from larger diameter to smaller will always provide support behind the cut. Just think of trying to whittle a stick – it is much easier to whittle from thicker to thinner diameter, and if you tried to whittle in the opposite direction you would tend to dig into the grain and break off wood instead of cut it. With a face turning (vessel) however, the grain typically runs at a right angle to the axis of rotation. This is pictured in Figure 2. When cutting the convex profile of a bowl, cut from smaller diameter (usually the bottom end) toward large diameter (the mouth) so that the wood being subjected to pressure from the Figure 2. Cutting chisel is supported by Downhill in Face Grain 259 Table of Contents 42-10 Turning Wood as a beginner. Take your sweet time and develop good technique. Speed may come after good technique, but it won’t help you to develop good technique. the larger diameter behind it. The rim of a bowl is an extreme case. If you cut the convex profile of the bowl away from the rim and toward the smaller end, the wood at the rim may just crack or splinter instead of cut. Likewise, when cutting the concave inside of a bowl, you cut from larger diameter to smaller. So, you always attack the inside rim from the mouth of the vessel, pushing the chisel toward the center of the bowl. CENTER (SPINDLE) TURNING 6. Scrapers must be kept perfectly flat on the tool rest and presented in the “trailing mode,” i.e., with the tool handle somewhat higher than the tool edge. That’s because the cutter on the scraper is actually a burr – a ridge along the top edge. Like a chisel, a scraper must always be supported by the tool rest before it touches the work piece. (Rowley’s Sixth Law) 1. The easiest way to find the center of the blank is to draw two diagonals on a rectangular piece of wood or to use a 45E machinist square on a round piece. Two lines bisecting a circle will mark the center. Just be sure to center the workpiece well at both ends, unless you are changing centers to make an elliptical cross-section. Make a little mark so that you can remount the workpiece in the same location if necessary. 7. I find that center turning is easier to learn than face turning. It may be well to master center turning before you try to hollow out a bowl blank. 8. Don’t plan to make anything useful at first. That is, don’t buy an expensive bowl blank or start right in on mahogany table legs. Poplar seems like an excellent choice for practice. It’s inexpensive, even-grained and soft enough to turn easily. 9. Professional woodturners place great emphasis on quick work. This is especially true for demonstration videos – “Look how fast I can hollow this bowl.” I admire the sight of long shavings streaming over the craftsman’s shoulder, and the fast, efficient work of accomplished turners – but the “need for speed” attitude was very distracting to me In a center turning, the grain runs parallel to the axis of the lathe, and the work is usually longer than its diameter. Examples are table legs, chair spindles, etc. 2. Normally, a spindle is driven by a spur center in the headstock. If possible, cut a shallow groove for the spurs with a hand saw. This is preferable to pounding the spur center into the workpiece. A. Do not use the tail stock to force the work onto the drive center – you may damage the thrust bearing at the end of the headstock. B. If the end of the spindle at the headstock is narrower than the spur center, you can leave the spindle wide at the end and trim it off later, put it in a chuck and trim it later if necessary, or use a screw center. 3. A fixed (“dead”) center in the tail stock can get pretty hot. You can judiciously wax it with candle wax or soap. A live center in the tailstock (one that rotates) is a useful 260 Table of Contents Notes and Reflections While Shaving Wood 42-11 accessory. FACE TURNING 4. A long or thin spindle may flex, which will interfere with cutting your design, and the spindle may break. You can support a thin spindle with your fingers, but you may need to use a steady rest. You can buy one or make a birds-mouth rest out of wood. (See Lathe Steady Rest) In face turning (faceplate turning) the grain of the workpiece runs perpendicular to the axis of rotation. Normally, the diameter of the work is greater than its length. The exception is when end grain is being hollowed out from a spindle, as for a goblet or mug. Bowls and plates are typical face turnings. Bowl blanks can be glued up in horizontal layers of contrasting woods or made with inlay effects. (Figure 3) See Gluing Up a Large Spindle, below.) Frid describes a number of clever methods of gluing-up bowl blanks, e.g., segmented bowls. (See Chapter 43, Blanks for Segmented Turnings.) 5. If I am making duplicate spindles, e.g., table legs that need to match, I will make a story stick marked at least with the locations of beads, pommels, and other features. A “negative” template cut out of a piece of ¼" plywood or Masonite is even better. It also should be marked with key locations. Transfer the marks from the story stick or template to the roughed cylindrical blank with a pencil (lathe running) and then use a parting tool to establish the diameter of each feature, e.g., “high” and “low” points. 7. Remove wood with a gouge, smooth out gouge marks with a skew, and smooth the turning with a scraper. A heavy roughing gouge (say, ¾") is very useful for rounding the blank into a cylinder and for removing wood to approximate design. 8. As you turn, the shape of the spindle may be obscured by your hand, the tool rest, etc. You can more easily see the shape you are making on the side of the spindle opposite the chisel. 9. Expect to repeatedly adjust the tool rest as the diameter of the spindle changes. Remember that the bevel must ride on the workpiece before the edge begins to cut. (Figure 1, panel 5) 1. Turning the outside of a bowl (its “profile”) is somewhat similar to spindle turning, except for grain direction. In fact, the blank may be held between centers while the profile is cut. Learning how to cut a long convex radius is challenging, however. 2. Hollowing a bowl is not like spindle turning at all. In my opinion, it is much more difficult. (More about this later.) 3. Gouges for hollowing out bowls need to be “long and strong.” The flute of a bowl gouge is deeper with much higher sides or wings. A fingernail grind is more important for a bowl gouge than for a spindle gouge. In my limited experience, deeper bowls are quite a bit more difficult than shallow ones. The authorities I used say that a d" bowl gouge is the workhorse of bowl turning. 261 Table of Contents 42-12 Turning Wood 4. Skew chisels are rarely used for face work, while scrapers are used more commonly for hollowing a bowl where a gouge cannot reach. A big heavy scraper is a great help. 5. Examine every bowl blank to make sure that it is structurally sound. Glued-up bowl blanks are often used. They seem strong enough to withstand being turned, and may provide inexpensive blanks (often from scrap wood) that yield interesting bowls. Think about the grain pattern when orienting the pieces for glue-up, but normally the grain has to be parallel. Cut a solid blank to approximate cylindrical shape on a bandsaw. The orientation of a bowl blank may be important if the blank will undergo significant changes in moisture content, especially if the blank is still somewhat green. From the perspective of wood movement, orient a bowl blank so that the bark side of the blank is toward the bottom of the bowl. When it dries, a bowl blank may seem to bend toward the bark side of the wood, as if the annular rings were straightening out. If the bark side is at the top, the base of the bowl may become convex, and the bowl may not rest steadily on a table. If the bark side is down, the base of the bowl may become concave, which may be much less noticeable. 6. Begin by mounting the top of the solid bowl blank to the headstock. You can use a variety of methods to mount it. See Mounting the Work, above. 7. You can stabilize the mounted blank with the tail center while you turn the profile. This will leave a dimple in the bottom of the bowl, but that should not be too unsightly. Also, if you will later mount the blank with a screw chuck, you should use the tail center to mark the exact center of the blank. This is where you will later insert the screw chuck. 8. Cut the profile. On a “new” blank, if you will secure the bowl with a four-jaw chuck, cut a mortise or tenon into the bottom of the bowl. It should be the correct size for the fourjaw chuck to make good contact all around, as described above under Mounting the Blank. If you will secure the bowl with a screw chuck or faceplate, the bottom should be thick enough to take the screws. You can cut away the wood with the screw holes later, if you want. Smooth the surface of the profile. Pay attention to the appearance of the bottom of the bowl. People will almost always turn the bowl over to look at the bottom, so do what you can at this stage to make it smooth and attractive. Figure 3. Bowls from Laminated Dry Blanks (Maple and Mahogany) 9. Turn the bowl around and fix the bottom to the headstock with a chuck, screw chuck, faceplate, dowel tenon, etc. as described above under Mounting the Blank. I find that a four-jaw chuck holding a shallow 3¼" diameter shoulder is best all around. 10. Decide on the inside depth of the bowl. 262 Table of Contents Notes and Reflections While Shaving Wood Measure the outside depth of the bowl and deduct for the thickness of the bottom and any part of the base that you plan to remove. Mark a drill bit (¼"-d") or narrow gouge with this depth. Drill a hole through the axis of rotation to the chosen depth. (I find that a drill bit held by a Jacobs chuck in the tailstock is easier than the chisel and more accurate than a hand-held drill.) This accomplishes two things: it clearly marks the depth of the bowl so that you will not have to stop and measure while you are hollowing the bowl, and it eliminates having to remove wood from the axial area. I might use a Forstner bit to reduce the amount of wood that has to be removed with a gouge or scraper. 11. Put the tool rest parallel to the end of the workpiece, about ½" away from it and a bit below center. You want the cutting edge to be just at the center. If the cutting edge is above the center, the spinning work will tend to push the chisel toward the perimeter. This can mar or even ruin the turning. If the cutting edge is slightly below center, the spinning work will tend to push the chisel toward the center, which is very hard to control. 12. First, true up the edge of the bowl using light scraping cuts. This can require some patience on a roughed out green blank that has dried crooked. Use light slicing cuts with the point of the bowl gouge (rolled over on its side) and work from the outside of the rim inward. Then cut a groove to establish the wall thickness at the mouth of the bowl. 13. On a solid blank, begin hollowing at the hole you drilled in the center. Turn the flute of the bowl gouge toward the center of the bowl, so that the gouge is on its side, rub the bevel, 42-13 and rotate the handle away from you until the chisel begins to cut. Raffan and Rowley (and others) recommend using a gouge in this way. As you gain experience in hollowing face turnings, you will begin to appreciate the benefits of various bevel angles and wing grinds. This is a complicated and somewhat subjective subject. Suffice it to say that a much sharper angle is needed on a gouge for it to rub its bevel at the bottom of a bowl than the angle needed at the sides. Hollowing is a difficult skill to master. Even Tage Frid avoided hollowing with a gouge – he said that he tried it once and that it almost killed him. (Frid, p.81) He preferred to use scrapers for this operation. 14. Eventually, further hollowing with the bowl gouge will become too difficult. Switch to a scraper. The scraper should be used with a light touch to produce thin shavings, not powder. 15. Use a calipers to get the wall thickness even, in particular, to avoid turning away too much wood. You can use your thumb and forefinger for rough work, but a caliper becomes necessary as you approach final thickness. I prefer a wall thickness of about 5/16" for a utilitarian bowl made of good sound hardwood; d" is OK, just thin enough to make a light bowl that is pleasant to handle, but still safe; ¼" is a bit nicer but seems riskier. I have made bowls with 3/16" thick walls, but they are fragile and sometimes break on the lathe if I get a catch. For a thinner wall, work in small sections about an inch wide at a time. Starting at the rim, thin the wall to the final desired thickness before proceeding to the next section. Blend the transition between sections with a scraper before moving on to the next 263 Table of Contents 42-14 Turning Wood one. If you attempt to turn the whole bowl too thin in a single stage, you may find that the bowl wall oscillates uncontrollably, so that it is nearly impossible to take a clean cut. Working progressively one band at a time increases the support for the section you are cutting. Once you complete a section and are well started on the next one, don't go back for another cut. You will have already removed the support. A Note About Pre-turned Bowl Blanks A bowl blank that had been roughed out when green and then dried probably has acquired an elliptical cross-section, and usually the bottom and the rim will not be flat. For such blanks, the first step is to mount the bowl with the bottom end to the right (toward the tailstock). Jam the inside of the blank against a waste block mounted in chuck jaws, on a faceplate or screw chuck. Use a piece of sandpaper or a sheet of rubber to increase the friction between the waste block and the inside of the blank. Use the tail center to hold it in place. A pre-turned blank should have been left with a tenon or large spigot at the bottom. The first step is to turn the tenon round again, or to cut one. In addition, true up the bottom of the bowl. If you omit this step, the bowl will not be properly centered on the axis of rotation when you reverse the mount and may wobble on a screw chuck or faceplate. Because a blank that was partially turned before drying will be elliptical, you should not attempt to rough out the profile as if it were a solid cylinder. The bevel of the gouge cannot rub along the elliptical surface, and the gouge will just slam repeatedly into the wider parts of the circumference. These wider parts are endgrain, because wood shrinks mainly across the grain when it dries. This situation is unworkable – it may damage the wood, the chisel, the tool rest, the mounting, and maybe you. Start at the narrowest part (bottom) of the bowl. Roll a bowl gouge over on its side, with the flute facing the direction of the cut (normally, to the left) and use light slicing cuts over the narrowest inch or two until you have made a smooth, round band where the bevel can rub. Use a sharp gouge and light cuts because you are cutting uphill. Then, starting from there, make light cuts along the profile, cutting downhill from narrow to wide (usually, from bottom to top). Use slicing cuts with the gouge rolled over on its side and the bevel rubbing on the wood. If you continue these cuts to the rim of the bowl, be careful (gentle) because the wood at the rim is unsupported, and pressure from the gouge may break off a large chip from the rim. Finishing Most turnings require some sanding before finishing. Skillful use of a sharp scraper can leave a smooth surface that requires little sanding, especially on hard dense woods. Less skillful tool use can leave ridges and tear-out that have to be sanded away. Sanding is slow, tedious and creates a lot of dust. (If you don’t use adequate dust collection you will see how much sanding dust is produced.) I use a 1½ HP dust collector and a vacuum cleaner to collect as much of the dust as possible, but some still escapes into the air. Remove the tool rest before sanding, as a safety measure. Do not under any circumstances wrap sandpaper or a rag around your finger. The safest place to apply the 264 Table of Contents Notes and Reflections While Shaving Wood sandpaper, I find, is near the top of the turning (10:00 - 12:00). If you go past the bottom (6:00) the turning can lift your hand up, with painful results. Sand the piece to final smoothness by running through the grits up to 400. You are sanding across the grain, so finer grits are needed than with flat work; otherwise, scratches may show when you finish the work, especially if you will be staining it. Be careful not to sand off crisp details. When sanding wood of differing hardness, like oak or pine, or wood with inlays, back up the sandpaper with a cellulose sponge, blackboard eraser, cork block, etc. When sanding or buffing inside a bowl, be very careful when sanding near the bottom of the bowl. It is too easy to catch and twist your finger. Use a sanding pad on a stick and light pressure. Glued-up turnings of contrasting wood (e.g., ebony or bloodwood with holly or maple) should not be sanded because the darker wood contaminates the lighter wood and spoils the crispness. Finish these with a scraper. It is easiest to apply finish while the turning is still mounted, especially for face turnings. Put a piece of paper down between the turning and the lathe rails. Otherwise, you are almost sure to get finish on the rails. If the piece has a lot of figure, e.g., birdseye, quilting or burl, I pop the grain with boiled linseed oil and let it dry overnight. If I have used white wood, like holly or maple, and want to preserve colors, I use a water-based acrylic finish like Polycrylic. I often prefer a high-gloss finish, and French polishing is my preferred method. I make a polishing pad (called a fad) from absorbent cotton or wool wadded inside a 42-15 square piece of hard-woven fabric. The outside fabric has to be strong enough to withstand rubbing without shedding lint. A piece of Tee-shirt or old work shirt is good. I usually make it about 2-3" in diameter at the bottom, and tie it with a string, as if the outside layer were a small bag. I soak the pad in shellac, wring out some of the excess, and pad the shellac onto the slowly rotating workpiece. If I apply too much shellac too fast, it flies off the turning onto the walls, lathe, and me. As I form a layer of shellac and it starts to dry, I recharge the pad and may increase the speed slightly. When the pad begins to pull, and changing the amount of pressure and the position of the pad does not help, I add a healthy drop (~c tsp ) of boiled linseed oil to the bottom of the fad and continue. I stop when I have a thin glossy coat of shellac over the vessel. If I apply too much I have to let it dry overnight before final polishing. Varnish is food safe when cured, but I would use only 1 or 2 thinned coats to seal the wood so that it won’t peel or flake later. Lightly smooth it with 400 grit sandpaper and 0000 ScotchBrite, and follow with wax. Raffan seems to prefer beeswax or boiled linseed oil and beeswax. I have used beeswax followed by paste wax with satisfactory results. The melting point of beeswax is fairly high and it needs to be buffed well with a rag or piece of clean paper to melt into the pores. Then the excess must be removed or the bowl will feel sticky to the touch. I usually finish buffing with 0000 ScotchBrite (plastic steel wool) followed by folded typing paper or a piece of paper bag. Sometimes, I seal the wood with sanding sealer or shellac (because they dry fast) before 265 Table of Contents 42-16 Turning Wood waxing. Wiping varnish is also useful but takes a few hours or overnight to dry. If a piece has holes, bark, etc. it may be impossible or unsafe to finish with friction methods. I sometimes use spray shellac, but it’s probably better just to wipe on some drying oil and let it go. It is tricky to do this well. If you turn the piece by hand it is hard to get an even coat. If the lathe is spinning, it is hard to see how wet the surface is and it’s easy to apply too much, resulting in flying droplets of shellac and possibly an orange peel effect. In any case, put up dropcloths all around. A turning can be stained, of course, but you should probably use a prewash to fill end grain, otherwise it will appear darker than the long grain. Also, stain will greatly magnify any surface irregularities. After the finish is dry, reverse the mounting, using finger jaws or a jam chuck and finish the bottom. The appearance of the bottom is important, because people will often turn a bowl over and look at the bottom. The tenon on the bottom that was used to mount the bowl in the scroll chuck can be removed, or turned thinner, etc to suit your esthetic desires. Sign the bowl before you put the finger jaws away. the intended diameter of the blank. You can usually get a decent match if you lay up boards from the same plank. Figure 4. Vertical If the wood is Inlay Effect (From the same species, you Love) should think about color matching and grain orientation. It is best not to glue a flat-sawn face to a quarter-sawn face, because they may expand or contract by different amounts as ambient moisture changes. Better, think about inlay effects and use different species with contrasting colors. This takes a bit of planning. See the article by Edwin Love. You can produce a nice inlay effect by alternating wood of contrasting colors to produce stripes. (Figure 4) You can also use one species in the center of the blank and then use different species as you build up the laminations concentrically. (Figure 5) As you turn into the darker wood, a horizontal inlay will appear. Making the laminated block is a Lamps and Large Spindles I usually have to laminate a blank for a lamp from smaller pieces. Laminated (glued-up) turning blanks may sometimes be a good way to use odd cutoffs left over from other projects. You can’t just glue a bunch of sticks together and then turn them, however, if you want to make a neat, attractive lamp. Furthermore, piecing together small pieces can be fussy, time-consuming work. It’s best to use full-length boards, about half as wide as Figure 5. Lamp Blank of Laminated Boards. Concentric Inlay. (Note that inner boards are a darker colored species of wood.). Photo Andy Staiano 266 Table of Contents Notes and Reflections While Shaving Wood critical step, because all of the joints have to be perfect across nearly the whole glued area. A void or other imperfection that appears while you turn the profile may render the blank unuseable, at least for the original design. This is frustrating, given all of the effort required to get to that stage. Joint all surfaces well and glue them up carefully. The blank for a lamp base would be made of disks, and so could parts of a lamp standard. The end-grain joints might not be very strong, but if the lamp tube is holding the lamp together vertically, end-grain glue joints may be strong enough. You may want to put a lamp cord though the center of the lamp. (I don’t like to see a lamp cord run outside of the lamp directly to the socket, although this is sometimes done, especially if the lamp standard is re-purposed glass or pottery.) The most common size of lamp tube is c" NPT, which has an outside diameter of about d". Plan on a ½" or even e" pipe chase. You can either drill a hole through the center of the spindle (very difficult to get perfectly aligned through a long lamp standard) or you can cut a dado into the center board(s) before you glue up the blank. If you use an odd number of boards to make the blank, then you can cut a slot in the centermost board. Simply leaving a gap between the sticks does not work very well, because this complicates clamping the sticks together at glue-up. Ideally, cut a stopped dado to leave some wood for the centers to bite into, or if the ends of the spindle will not show, will be cut off, etc., glue in a short (23" long) plug at each end. The late Andy Staiano, who was a professional lamp turner from Rockport, ME, recommended putting the lamp aside after initial shaping and sanding with coarse 42-17 sandpaper. He wrote that a lamp standard should rest a few weeks to dissipate all of the water introduced by the glue – otherwise some joints may become uneven. A SIMPLE EXAMPLE 1. Design the lamp, including a cross-section of the laminations (Figure 6). The length of the lamp tube must equal, for example: the depth of the threaded recess in the socket, ½" plus the length of the collar or barrel, e.g., 1½" plus the length of the standard, 17¼" plus half of the thickness of the base, ¾" If you have a tap you can cut the lamp tube to length and re-tap the end. If you cut the threaded part of a lamp tube, run a nut onto the tube before cutting. This will clean off the threads disturbed by the hacksaw. 2. Cut out the parts and joint all surfaces that will receive glue 3. Laminate two pieces, e.g., the top and bottom layers in Figure 6. 4. When the glue is well cured (e.g., 24 hours) joint the surfaces to be glued and dry fit the top, middle and bottom layers. 5.Cut a plug about 4" long to fit snugly in the central opening at each end (Be careful that Figure 6. Cross Section of Laminated Blank for Lamp Standard (Maple and Mahogany) 267 Table of Contents 42-18 Turning Wood the plug is shorter than the maximum depth you can drill with your d" bit.) 6. Glue the remaining pieces to form the blank for the lamp standard. 7. Clean off squeeze-out and, square up the ends. After the glue is fully cured, the blank is ready to turn. 8. Cut out and glue up the blank for the base as necessary. The base needs to be thick enough to accommodate a washer and lock nut on the tube and space for the lamp cord to enter the tube. An overhand knot to secure the lamp cord is also desirable. The minimum thickness of the blank should be about 1". It can be tapered toward the edges so that it looks lighter. 9. If you will make more than one matching lamp, make a story stick or template before you turn the standard. 10. A tapered profile requires a lot of wood to be removed. If you made a template, it might be quicker to cut out the approximate profile on a bandsaw on each side (cut well outside the line). Cut two opposite sides and then stick them back in place with double-sided tape. Then cut out the remaining two sides and remove the cut-offs and tape. 11. Turn the standard and base. You may choose to let the moisture in the turning equilibrate for a while at this point. If you intend to remove the turning from the lathe, make sure that you will be able to remount it in exactly the same position. 12. Sand to 400 grit and finish. 13. Drill a centered 3/4"-1" diameter hole half way through the base from the bottom. This is for the nut that will go on the end of the lamp tube. 14. Drill a centered d" - ½" hole in the top and bottom of the standard. The holes have to line up exactly along the axis of rotation, or the tube will not slide through the standard when you assemble the lamp. This is most accurately done on the lathe using a Jacobs chuck in the headstock. Remove the tool rest. Set up the blank as if you were going to turn it, with the point of the drill where the spur center would be. Tighten up the tail center, then start the hole by rotating the standard by hand. After the drill bit has penetrated the wood by a half inch or so, you can turn on the power (low speed). The standard will turn, but you should be able to grip it with your hands to keep it from rotating. Just to be on the safe side, be sure the tool rest has been removed before doing this. 15. Drill a 3/16-1/4" hole through the side of the base, angled slightly downward, to intersect the 3/4"-1" hole in the bottom. 16. Assemble the lamp. (Figure 7) You can glue up long stave segments into a spindle blank. For example, the lamps in Figure 7 have hollow standards, so they are relatively light and waste less stock. Be sure to make the walls thick enough that you can turn the profile you want. Chapter 43 on segmented blanks discusses how to make these. They require jam chucks to mount the hollow blanks on the lathe before the base and top cap can be glued on. CHOOSING A HARP AND A LAMPSHADE Typically, a lampshade should be 2/3 the height of the standard and base, so that the lamp won’t l o o k t o p - o r Figure 7. Laminated Lamps 268 Table of Contents Notes and Reflections While Shaving Wood bottom-heavy. The top of the turned standard shown in Figure 7 is 18" above the tabletop, The shade should be 12" high. Choose a harp that will allow the lampshade to hide the socket, neck or barrel (the piece between the base and the bottom of the harp). The widest part of the lampshade should be at least a half-inch wider on each side than the widest point in the base, in this case about 8-8½". However, also consider the location of the lamp: if it’s going on a side table or bedside table, you won’t want to bump it whenever you get up. (Consider a shade that is nearly rectangular in cross-section.) Likewise, the shade should not stick out into the room so far that you bump it when you walk by. Segmented Turnings (Introduction) Segmented turnings are made from glued-together disks, comprising small flat segments in which the end grain has been cut at an angle, or compound-angle wedges, glued together. The long grain shows on both the inner and outer surfaces of the turning. (Figure 8) This can make a very attractive surface with some woods that have showy grain, like quarter-sawn oak flakes or fiddleback maple, and allows interesting patterns and inlay effects using wood of contrasting colors. It requires excellent joinery technique in addition to woodturning skill. Although ring construction is the most common, variations exist. Open segmented turning is similar to ring construction, but small gaps are left between the segments. Successive rings are offset so the segment overlaps with the ring above and below. This type of segmentation can appear delicate but is generally quite strong. 42-19 Bowls and hollow vessels are the most frequent example of segmented turning, but longer forms like lamps, t a b l e Figure 8. Two p e d e s t a l s , a n d Segmented Bowls columns can also be made this way. Stave and compound segment constructions are variations of segmented construction. Staves and compound segments are pie-shaped, with compound angles along their edges. Such constructions are assembled like barrels from pieces in which the grain runs parallel to the axis of rotation (usually the long dimension). Figure 9 shows two examples of compound segment construction (the bowl and the plinth), made for practice out of pine. Note that the grain runs radially (horizontally). In a staved construction the grain would run parallel to the center axis of the bowl. A segmented turning can combine ring construction, stave construction, and solid, non-segmented wood in a single piece. However, wood expands and contracts according to ambient humidity. To design and assemble a durable segmented turning, the turner has to be take this into a c c o u n t . Cross-grain con- Figure 9. Compond s t r u c t i o n s , e s - Segment Bowl and pecially of thicker Plinth rings or solid pieces, may fall apart if the piece is exposed to large changes in humidity, e.g., summer 269 Table of Contents 42-20 Turning Wood humidity vs winter dryness indoors. A final example of variety in segmented constructions is called "bowl from a board". The fundamental idea is to cut rings out of a flat, square board. Sometimes, the board is cut in half to create two identical rectangles, and then semi-circular rings are cut from each of the two boards. One half-ring is the mirror image of the other. The half-rings are typically cut with a bandsaw. The blade may be set at an angle from the vertical, typically 45-60 , so that when each pair is glued together into a ring, the ring will be shaped like a cross section of a cone. Successive rings are cut as concentric semi-circles, so that when the rings are stacked, they form a cone-shaped blank, somewhat resembling a blank formed by segmented rings. A bowl made up in this way would save wood, but it would look similar to a bowl turned from a solid blank, except for the layers formed by the semi-rings. It would lack the continual long-grain effect all around the turning that one gets from a segmented blank. The fun comes in the composition of the "board" we started with. A variety of fantastic effects can be achieved, depending on how the boards are constructed before the concentric rings or staves are cut. For example, the board can be made up of strips of contrasting woods to form a checkerboard design. The position of the semi-circular disks can be rotated slightly at each layer to create a spiral effect, and so on. Each of these will be covered in Chapter 43, Making Blanks for Segmented Turnings. References Raffan, Richard. Turning Wood With Richard Raffan. Newtown, CT. The Taunton Press, 2001. Raffan, Richard. The New Turning Wood with Richard Raffan (DVD). Newtown, CT. The Taunton Press, 2008. Rowley, Keith. Woodturning: A foundation course. New Edition, Lewes, East Sussex, England. Guild of Master Craftsmen Publications, Ltd. 1999 Sears, Roebuck & Co. The Wood Lathe – An Illustrated Manual of Operation. 1969. (Accessed from OWWM.COM) 270 Table of Contents Notes and Reflections While Shaving Wood Chapter 43 – Making Blanks for Segmented Turnings A discarded. Strictly speaking, the simplest form of segmented blank is made by gluing large pieces of wood together, vertically or horizontally, to form a solid blank or, in some cases, a solid blank with a small hollow core, as for a lamp blank. These are relatively simple to make. Interesting patterns can result after turning these on a lathe. These are not usually called segmented turnings, however. If contrasting woods were used, they are sometimes called inlaid turnings, although they are not inlays, either. s described in Chapter 42, a segmented turning is made from a blank composed of many pieces of wood systematically glued together, in contrast to a blank made of a single piece (or a few pieces) of wood. Segmented blanks can be made by four basic methods, alone or in combinations. Most segmented blanks require exacting design and joinery skills. Perhaps the best introduction to this topic is the series by Jim Rogers (See References.) Tage Frid offers a brief introduction. (He calls it “bricklaying.”) Woodturning Online has a section devoted to segmented turning, which offers many useful discussions and some beginner projects.3 One interesting beginner project includes a combination of inlay and segmented turning. This is the "polka dot bowl" in which a ring is laminated horizontally from contrasting woods. (See Daudelin reference.) The main purpose of making segmented turning blanks is to create interesting patterns of grain and color beyond the possibilities of solid wood. A great variety of patterns is possible, depending on the composition of the “boards” used to make the segments, especially with concentric rings or compound segments. For example, the boards can be made up of strips of contrasting woods to form checkerboard effects or the position of the rings can be rotated slightly at each layer to create a spiral effect. A mosaic picture can be created. Segmented blanks also use wood somewhat more economically than do solid blanks, because less wood is turned away and Flat Segments, Compound Segments and Staves One common style of segmented turning, uses flat rings. Each ring is made up of narrow (e.g., 1¼" wide) trapezoidal pieces in which the end grain has been Figure 1. Flat m i t e r e d a t t h e Segmented Ring appropriate angle and glued together end to end. Then the rings are glued together in layers to form a turning blank. The long grain shows all around the surfaces of the turning. Another style uses compound segments cut from a board and glued along the end grain edges. So, where a flat segment has ends cut parallel to the axis of rotation, a compound Figure 2. Bowl and segment has its ends Plinth made from cut at an angle to the Compount Segments 3 http://www.woodturningonline.com/Turnin g/segmented_turning/index.html 43-1 Table of Contents 271 43-2 Making Segmented Turning Blanks axis of rotation. A compound segment is meant to be sloped in the finished turning. (See Figure 2) A third style is a staved construction. Staves are similar to flat and compound segments, except that the long grain of a stave is parallel to the axis of rotation. Normally a stave comprises the whole height of the blank. The difference between long segments (flat or compound) is important when you think about stock and glue-up. A flat or compound segment can be only as long as the stock board is wide. A stave can be as long as the stock board is long. The joints between adjacent segments in a ring are end-grain joints. The joints between adjacent staves are long-grain joints. These three styles require careful design and excellent joinery technique. The wider the segments (inside to outside) and the higher the number of segments, the more accuracy is needed. My first segmented turnings were octagons, which was a good choice. As you gain experience, confidence, and patience you can work your way up to finer work, e.g., 12 or 24 sides. Bowl From A Board Blanks made of half-rings or solid rings cut from flat stock are sometimes called “bowl from a board”. These are glued together to form a conical blank. Choosing Wood for Segmented Blanks The ability to use wood species of contrasting colors is an important advantage of segmented turning. Furthermore, when small quantities are required, one can consider rare and exotic species that otherwise might be too expensive. Malcom Tibbets classifies the wood species that he has used in his career into three groups, as follows. (The species that I have used and appreciated are italicized.) Tibbetts’ personal favorites include apple, blackwood, bloodwood, bocote, bubinga, carob, ebony (Gabon or Macassar), holly, jarrah, figured maple, hard maple, spalted maple, mesquite, mountain mahogany, myrtlewood, persimmon, pink ivory, purpleheart, rosewoods (pau ferro, East Indian, Honduras, Brazilian, cocobolo, flamewood, tulipwood), yellowheart, and ziricote. He prefers these species because of color and figure. Except for the rosewoods they all glue well and most of them are tight-grained, which means they will polish nicely. Minor species (“limited-use”) include alder, ash, aspen, birch, lacewood, madrone, mahogany, soft maple, both red and white oak, padauk, poplar, redwood, teak, and zebrawood. Acceptable (“OK”) species include spalted beech, spalted birch, cherry, ironwood, limba, Osage-orange, walnut, and wenge. Highly contrasting colors can add a lot of visual interest to a turning. One complication, however, is that the woods should be about the same hardness and should finish well on the lathe with just scrapers. For example, a blank containing bloodwood and poplar is difficult to finish because the poplar is soft and does not finish well with scraping – it tends to have small chipouts and bits of rough grain. It sands easily, but sanding dust from the bloodwood will smear across the poplar. You can work around this by sealing the wood, sanding, sealing, etc. but this is tedious. Also, care is needed in turning such a blank lest the chisel cut too deeply into the poplar and then hit the bloodwood. Some of the woods listed above are 272 Table of Contents Notes and Reflections While Shaving Wood irritants or sensitizers. These include blackwood, cocbolo, ebony, myrtle, olivewood, paduak, rosewoods, teak, and wenge. Surfacing segmented rings, especially, can produce lots of dust, so proper dust control, including a facemask, may be a necessary precaution. MOISTURE CONTENT. Simply put, wood used to make segmented blanks should be dry, certainly not green. About 10% moisture content (MC) or lower should be OK if all the wood in the piece is at about the same MC. Because many different woods may be mixed in a segmented turning, it is important to ensure that you do not mix wood of very different MC, especially green and dry woods. Wood whose MC is in equilibrium with the ambient air should be OK even if the MC is a bit higher, except that ambient humidity can change with the seasons. The orientation of different layers also may present stability issues if the wood dries or expands to different extents. If the wood is not as dry as you would like, consider two points: (1) you can dry small amounts of wood quickly in an oven (180EF) or microwave (2) dimensional changes due to moisture content might only amount to 1/32" or less over the small distances involved. Radial (across the rings) dimensional change with moisture content is much greater than change tangential to the rings. Therefore, from this perspective it makes sense to cut flat segments so that the radial face of a segment is toward the inside and outside, leaving the tangential faces for the tops and bottoms of the rings. Appearance matters too, of course. Sometimes the radial (quartersawn) face is prettier. 43-3 Designing Segmented Blanks Flat segment constructions allow the greatest flexibility in designing a profile. For example, very bulbous bottle shapes, even hollow closed forms, can be laid up by assembling portions of the vessel, turning the interior, and then assembling the whole vessel before finishing the exterior. Compound segment and stave constructions are much more limited. The straight sides of the blank limit the amount of curvature possible, and the overall geometry fixes the ratio of diameter to depth for a given angle (jig makeup). Bowl from a board (BfB) constructions are limited somewhat by the number of rings that can be cut from a given diameter. Each is described further below Software is available that assists in the design process on a personal computer. Depending on what aspect of a segmented turning I wanted to draw, I have used Woodturner Pro, 3D Design Pro, and Google Sketchup. Each has its uses but also significant limitations. Woodturner Pro is useful for deciding on the number of rings, number of segments per ring, and segment widths; calculating angles and board length; and producing a cut list for each ring. For designing rings with many features, where placement of contrasting colors is important, I make a page of blank segmented rings which I can then number and use to show the placement of the various colored pieces. 3-D Design Pro is complementary to Woodturner Pro. It draws a 3-D rotated view of a profile that you select from a library or draw on-screen. Designing a segmented blank is straightforward, but getting rings of the right appearance, diameter and width can be complicated. A sketch, at least, is necessary. 273 Table of Contents 43-4 Making Segmented Turning Blanks My philosophy is that the design determines the required stock, although some compromises are possible to increase the efficient use of stock, especially use of scarce or expensive woods, without negating the design. For example, it is more efficient to locate features where the diameter of the form is more nearly constant (where the profile is more nearly cylindrical). Often, a cylindrical surface is best to show a design, anyway. In contrast, a wide ring is needed where the diameter of the form changes sharply. Most of that ring will be turned away, so this may not be an efficient use of expensive wood. Secondly, it is wasteful, for example, to specify a ¾" thick ring when your stock is 1" thick. You will have to trim off a 1/32" 1/16" piece, for which you may never find a worthy use. Why not try to use it in your design as a 1" thick piece? In the same vein, it may seem to make sense to buy short pieces of exotic wood for making flat segmented rings, but usually the many little cutoffs that result are useless. 1. Draw the object in vertical cross-section on graph paper or a computer, showing the profile and wall thickness. Establish the vertical centerline. 2. Draw rectangles, which represent the segments in cross-section, along the profile. For example, if the bottom ring will be ¾" thick after it has been assembled and flattened, draw a rectangle ¾" high on the profile. That will show you how wide the segment has to be to enclose the curved profile over that height. Depending on how many sides your polygons will have, allow extra (at least d") on each side for turning the rings into circles and smoothing them. For compound segments, the procedure is essentially the same, except that you have to draw the parallelograms representing the compound segments at the correct vertical angle. This will help you decide how thick the stock needs to be to get the curvature you want. Each ring must cover its part of the profile, including the thickness of the wall. 3. Consider the species and colors of the segments, including feature effects 4. Calculate the segment length and amount of stock needed to make each ring. When designing a bowl, recognize that a circle inscribed within a polygon must have a smaller diameter that the diagonal of the polygon. For example, an octagon with diagonal 8" can be turned into a circle with a diameter no larger than about 7d". Nonetheless, when mounted on the lathe, the diagonal of the polygon will be the limiting size of the blank, unless the corners are first cut off. If a lathe can swing 9", the largest polygon it can turn has a 9" diagonal, which corresponds to a circle with about an 8½" diameter. This, and the earlier points, may seem boringly obvious. I’ll just say that they will not seem so obvious if you ever hapen to glue up a blank that will not turn on your lathe or that comes out much smaller than you expected. Making Segmented Blanks The three fundamental issues in making segmented blanks are how to make segments of precise angle and length, how to flatten rings and build them into blanks, and how to mount a hollow blank on the lathe. 274 Table of Contents Notes and Reflections While Shaving Wood CUTTING SEGMENTS AND MAKING RINGS Segments have to fit together perfectly to form rings and the rings must be flat and well joined. Otherwise, the finished turning will be unattractive at best and may fall apart. Each segment in a ring must be of equal length, and each angle must be precisely correct. For example, the segments for making octagonal rings must be mitered at precisely 22.5E (I suppose, within 0.05 degree), or the ring will not “close” without gaps. Few miter 43-5 guides, even expensive post-market ones, would be precise enough. I prefer to use a miter sled on a table saw to cut segments or a compound miter saw. (See Chapter 45, Making a Sled for Cutting Segments) The stock from which segments will be cut should be surfaced to consistent thickness, in order to reduce the effort required later to flatten the rings. Table 1. Number of Sides, Miter Angle, Tangent, and Segment Length by Ring Diameter Sides (s) Miter Angle (a) * Tangent 6 30.0 8 Segment Length L (in.) by Ring Diameter (in) to the nearest 1/16" 5 5.5 6 6.5 7 7.5 8 8.5 0.577 2-f 3-3/16 3-7/16 3-¾ 4-1/16 4-5/16 4-e 4-15/16 22.5 0.414 2-1/16 2-¼ 2-½ 2-11/16 2-f 3-c 3-5/16 3-½ 10 18.0 0.325 1-e 1-13/16 1-15/16 2-c 2-¼ 2-7/16 2-e 2-¾ 12 15.0 0.268 1-5/16 1-½ 1-e 1-¾ 1-f 2 2-c 2-¼ 16 11.25 0.199 1 1-c 1-3/16 1-5/16 1-d 1-½ 1-9/16 1-11/16 24 7.50 0.132 11/16 ¾ 13/16 f 15/16 1 1-1/16 1-c * equals 180E/s Measured relative to the saw blade, i.e., as a miter gauge would read. Length of stock needed equals the perimeter of the ring (the number of segments times the length of a side) plus an allowance for saw kerf, (roughly 1.5" for a 12-sided ring). If the stock can be flipped over between cuts, however, the length of stock needed is less than that. Every other segment would require [L- 2*h*tan(a)] inches, where h is the width of the segment. For example, when making a 7½" diameter, 1" wide, 12-sided ring, each odd numbered segment (1,3, . . .,11) would be 2" long (plus a saw kerf). That’s about 6 x 2c = 12¾". Each even numbered segment would require only [2 - (2 x 1" x .268)] = 1½", (6 x 1½" = 9"). So, the 7½" 12-sided ring would require about 21¾" of stock, not 26½". 275 Table of Contents 43-6 Making Segmented Turning Blanks Most bowl profiles require rings of different diameters. The length of the segment is proportional to the diameter of the ring, given the number of segments. The constant of proportionality is the tangent of the miter angle. For example, if the ring is an octagon, the miter angle is 180/8 = 22.5E. The tangent of 22.5E is 0.414, and the relationship is segment length . tan (a) * diameter segment length . 0.414 * diameter So, an octagonal ring with an 8" diagonal requires a segment that is about 3 5/16" long. Tables are available with segment sizes precalculated (See Table 1). CUTTING SEGMENTS Cut and surface the supply pieces for the segments. Adjust the sled and the saw (See Chapter 45). After that, cutting segments is straightforward. If the stock is uniform on both sides, it can be flipped between cuts. This minimizes the number of necessary cuts and the amount of wood wasted. Consider keeping the segments in the order cut, so that color or grain changes will be gradual. The long grain running horizontally aroud the bowl can be very attractive. If a board has a markedly different appearance on one edge than the other, you may choose to cut the segments individually, so that one edge is always toward the outside. Also, very hard-to-plane wood should have the grain direction (rising or falling) the same for each segment for easier flattening (see below). One trick to economize on wood and to speed cutting is to make rings in pairs. As you cut a segment and flip the board over, make two piles. Then glue up rings from each pile. Until you are certain that the fence angle is dead on, dry assemble each ring, put it in a hose clamp (band clamp), and inspect the ring for gaps by holding it up to a bright light. If the segments fit together with significant gaps, the best solution is to adjust the saw or the sled. However, for tiny cutting errors that cannot be adjusted away, there is a workaround. Any ring comprising an even number of segments can be glued up as two semicircles. The arc described by the two semicircles should be 180E. Therefore, you can fit the segments together so that the gaps between segments disappear. The error will accumulate as a deviation from 180E, which can be sanded away after you glue up the two semicircles. You do this by placing two dowels or, preferably, shims between the semicircles on the side where the gap is. Re-locate the shims or dowels until all the gaps close under clamp pressure. Then, after gluing up sand the two semicircles until they mate perfectly. Then glue them together. (See Figure 3 and Rogers, Part 1.) This workaround is a normal part of assembly as long as a small amount of wood (say, 1/16" or less) has to be Figure 3. Split Ring removed. If very much more than that is sanded away, the ring will be elliptical. The ring will eventually be turned round, but this complicates assembly of the blank. It may make the ring a bit more difficult to center. It will require some caution when you begin to turn it round, and may distort some designs that depend on an equal amount of wood being removed all 276 Table of Contents Notes and Reflections While Shaving Wood around the ring, e.g., a polka dot design made from a lamination. (Of course, the latter can also enter into design, if you want a contrasting layer to appear only on opposite sides of the turning.) Tibbetts, the authority on segmented blanks, recommends sanding each glue surface. I prefer to use a blade that gives me a smooth cut. I have not found that sanding has been helpful. I have made “perfect” joints without sanding, but I have not been able to make a jig for a disk sander that could give me the precise angles I get on my table saw. Inspect each cut to make sure that it is free of scratches, crumbs, and broken fibers that might get in the glue joint, etc. A light, even rub on a piece of sandpaper resting on a dead flat surface might be helpful. After each segment is cut and the fit is ascertained, spread each bevel with PVA adhesive (Titebond II or III), lay the segments in a ring on a dead flat platen with a non-stick surface, and clamp them with a hose clamp. Figure 4. Octagonal Ring and Bench HookPlane Stop. The triangles are just screwed to the bench hook. The right-hand one is easily adjustable for various sized rings. I prefer a platen of melamine-coated MDF. I apply and buff off a coat of paste wax 43-7 to keep squeeze-out from gluing the ring to the surface. You can also lay down a piece of waxed paper. Apply glue evenly to both mating bevels and rub the surfaces together to help spread the glue and (possibly) improve penetration. Push each segment down against the platen so that one side of the ring (at least) will be flat, or nearly so. When you tighten the hose clamp (screw-band clamp) you should see squeeze-out at every joint. Recheck that each segment is against the platen. Sand the semicircles as necessary and glue them together. FLATTENING RINGS When the adhesive has cured (overnight) the rings must be flattened. That is, the segments must be trued up – all in the same plane and perfectly square to the axis of rotation. This step is important because it determines both the strength and the appearance of the glue joints between rings in the finished bowl. Flattening rings is a bit tedious and different people recommend a variety of methods. I normally use a different method for each side of the ring. First, I flatten one side of the ring using either a sanding disk or a bench plane. I use a 12" diameter sanding disk and a homemade table that mounts on the rails of my lathe for this purpose. If I can flatten the whole diameter of the ring with my sanding disk, I use that. The ring must rest on the sanding table, so the diameter of the ring has to be equal to or less than the radius of the disk. For rings with larger diameters, I flatten the rings with a bench plane. I use a modified bench hook to hold the ring secure while planing and to reduce the pressure on the glue joints (Figure 4). Work your way 277 Table of Contents 43-8 Making Segmented Turning Blanks around the ring, planing with the grain (tangentially) using a deliberate slicing action rather than repetitive forward and back planing motions. Use a very sharp plane iron, and be careful that the toe of the plane does not strike the corner of an adjacent segment and splinter a piece off. (Even if this were to happen, however, the chip on the inside of the ring would be turned away.) At first, you can use a block plane to roughly level off any proud segments, but final flattening should be done with a plane that is long enough that the toe and heel span the ring while you are cutting, e.g., a jointer plane. No light should be visible between a straightedge and the ring at any position. When the ring is nearly flat (within a few thousands of an inch) you can finish flattening by rubbing the ring on a piece of 80 or 90 grit sandpaper laying on a flat surface, e.g., ¾" MDF. I have used a thickness planer (the blades were very sharp) but this is dangerous. If one side of the ring is not flat, the planer rollers may break it apart. Also, since the grain runs in many directions, the planer may cause severe chip-out. If you try this, use a carrier board, hem in the ring well with waste blocks, and take very light (1/64") cuts. Others recommend a drum sander or belt sander. I feel that a hand-held belt sander would be much too difficult to control. Some use a so-called safety planer on a drill press. So far, the idea does not appeal to me. At this point in the process, one side of the ring is flat. You can either repeat the process to flatten the other side off the lathe, or flatten the other side on the lathe, as described in the following section. MAKING AND MOUNTING THE BLANK Obviously, the blank must have a closed bottom to become a bowl or vase, and may need a top as well. The two issues here are (a) to find a way to mount the rings (or the assembled hollow turning blank) on the lathe and (b) to glue the rings concentrically. They must be precisely concentric for a deep bowl or vase. One method for mounting a segmented blank is to cut out a solid bottom and mount it between centers. Turn the bottom to size and turn a tenon on one side that will fit a chuck. Or, you can use a glue block attached to a faceplate. The glue block can be fairly long, to provide working space between the headstock and the turning. Glue the solid bottom to the glue block and turn the bottom into rough shape at least. After the bowl is turned, you can cut it off of the glue block, reverse the mount using Cole jaws, and clean up the bottom. FLATTEN THE SECOND SIDE ON THE LATHE When one surface is perfectly flat, I normally glue it on the blank, and flatten the other side on the lathe. This is the fastest way to flatten a ring, but it takes some care. The surface of the ring must not only be trued (i.e., flat) it must be perpendicular to the axis of rotation. For example, if the ring is hollowed out, i.e., thicker toward the edges, the glue joint may look OK when the blank is glued up but then the gap will show as the profile is turned. If I have a lot of wood to remove, i.e., if the segments were not nearly the same thickness, I start with a gouge rolled over on its side, using a slicing cut. Then I move to a small square-end scraper, and then to a scraper at least as wide as the segment. I finally use a sanding block long enough to span the 278 Table of Contents Notes and Reflections While Shaving Wood diameter of the ring. If the ring is thick enough, I may mount it by itself on Cole jaws. Then I dismount the ring and check it for flatness using a straightedge. When one side is flat I turn the ring around and repeat the process. Mounting one ring at a time on the lathe is recommended by Rodgers and other experts in segment turning. One advantage of this approach is that each ring can be centered exactly on the lathe because it can be spun slowly by hand next to the tool rest. Any offcenteredness will be obvious, and can be corrected before the glue tacks up. The disadvantage of this method is that it takes much longer than gluing the rings up all at once, because the adhesive should be allowed to cure at least an hour before turning each layer. Also, although the surface of each ring will be trued up easily enough on the lathe, it is too easy to turn the ring out of square relative to the axis of rotation. You need to check it frequently with a straightedge laid across the rim, and this is best done off the lathe so that you can hold the ring in front of a bright light. Finally, you can get much more clamp pressure using a gluing jig than you can (or should) using the tailstock. An alternative method is to glue up many rings at one time. This requires that both sides of each ring be flattened off the lathe, but all of the rings can cure overnight under pressure from clamps or a glue press. By careful measuring and letting the Titebond II “grab” before applying clamp pressure, I can get the rings fairly concentric, and the blank well balanced. As a compromise, of course, you can glue up part of the blank. For example, if you leave it wide at the top, the inside will be easier to rough out. Rogers describes making a vessel in two halves, building up the top half separately from the 43-9 bottom half. That would be a way to make a closed vessel that would be impossible to hollow out if made in one piece, or even a bowl with a partially closed top. You can use the tailstock to press a flat plywood disk against the right side of the blank to support it while you roughed out the exterior profile. Then just remove the plywood to turn the inside of the bowl. Roughing the inside of the polygonal rings into circles is best done with a parting tool, or a gouge rolled over on its side for a shearing cut, coring out the corners, i.e., cutting “down” from the top surface through the ring, just inside the polygon. OPEN SEGMENTED BLANKS Open segmented turning is similar to ring construction, but small gaps are left between the segments. Successive rings are offset so the segment overlaps with the ring above and below. This type of segmentation can appear delicate but is generally quite strong. Bowls and hollow vessels are the most frequent example of segmented turning, but longer forms like lamps can also be made this way. Compound Segments and Staves As described above, the grain in a compound segment runs perpendicular to the axis of rotation (the same as a normal segment). A stave is like a compound segment, except the grain in a stave runs parallel to the axis of rotation. A compound segment is crosscut, while a stave is ripped, so the two types require different kinds of sleds and (ideally) blades. Compound angles are calculated by 279 Table of Contents 43-10 Making Segmented Turning Blanks trigonometry (See Appendix). Higher-end construction calculators with trig functions often have built-in functions for calculating compound angles. There are calculators on the internet. See the book by Malcom Tibbetts and the article by Bob Pritchard. Compound segments should be made with a sled on a table saw, or with a compound miter saw. Extreme accuracy is required for both the miter angle and the blade angle. The easiest way to use a table saw to cut compound angles is to tilt the arbor. I normally prefer to leave my blade arbor set at 90E, so I tried a slanted cutting sled. This was successful for a very shallow 30E angle to the horizontal, but for a deeper bowl it is necessary to tilt the saw blade. For example, to make a blank whose side has a slope of 60E, I would need a miter angle of 7.6E and a “tilt” of 12.8E. 4 A sled to produce a 12.8E tilt is impractical because it would put the workpiece too far above the table for the blade to cut it. It is possible to cut compound segments from individual small billets, but this is awkward and wastes the stock that is trimmed off each piece when it is cut. (This may be a significant amount of wood.) For bowls and other blanks that have compound segments less than, say, 8-10" long, It is better to cut many compound segments from a long board that is as wide as the length of a segment. Like flat segments, compound seg4 . Miter angle as it would be shown on a normal miter gauge, i.e., 0E is at a right angle to the path of the saw blade. The calculated tilt angle is measured from the normal vertical position of the table saw blade. ments can usually be cut by flipping the stock over after each cut. Since compound segments show a lot of grain and figure, however, you have to consider the appearance of both sides of the board. Sometimes, the two sides do not match very well and the most pleasing effect will come from using every other segment. So, as you flip the board over after each cut, make two stacks marked 1, 2, 3, . . . from one side of the board and one marked A, B, C, . . . from the other side. Keep them in the order that they came off the board. The two stacks of segments will make two blanks out of the same piece of stock. As described above, the width of the segment (at its widest end) approximately equals the tangent of the bevel angle multiplied by the desired diameter of the bowl (largest diameter). So, for a 12-sided blank, with a 7" diameter, the segment width W would be approximated by W = tan(15E) * 7" = .268 * 7 = 1.87", or about 1f. This ignores compound angle calculations, but is accurate enough for the largest diameter of the cone. The actual segment width by compound miter calculations (see Appendix) is 21/16". The amount of stock needed (the perimeter of the blank) depends on W plus an allowance for saw kerf (c"), multiplied by the number of segments. That equals 24" in this example. Moreover, if you will use every other segment, A, B, C, . . . additional stock will be required for the “extra” segments, i.e., 1, 2 3, . . . , i.e., the short side of each adjoining trapezoid. You can estimate the length of the short side by laying out two adjacent segments on a piece of paper (highly recommended) or by the calculations in the next section. 280 Table of Contents Notes and Reflections While Shaving Wood Compound segments are glued together end grain to end grain, while staves are glued together long grain to long grain. This affects the strength of the joints significantly. The strength of a flat segment or compound segment vessel comes from the long grain joints as the rings are stacked on each other to make up the blank. A vessel made up of one ring of long compound segments (see Figure 2) is likely to be weak. The end-grain glue joints may fail while the blank is still on the lathe, especially if the design calls for thin walls. It can be strengthened by gluing in a tenoned bottom. Staves and long compound segments are difficult to clamp during glue-up. It is best to glue up adjoining segments two at a time. Then when the glue has set, glue the pairs into into half-cones.(See below.) Too Much Math If you prefer a numerical solution instead of a sketch, here is the math. You can estimate the length of the short side by calculating the quantity W-2*h*tan(x) where h is the height of the segment and x is the miter angle given by compound miter calculations. In this case, a 12-sided blank with sides that are 60E to the horizontal, the miter angle equals 7.6E). So, for example, if the compound segments will be 3" high (long), then the narrow end of each trapezoid, will be about 1 f - 13/16" = 11/16" wide. Since there are 12 of them, we need to add about 12¾" to the stock requirement. The total length of stock is 24" + 12 ¾" = 33¾" The quantity 2*h*tan(x) is important in itself. It has to be equal to or less than W, the width of the compound segment. If it 43-11 equals W, the compound segment will come to a sharp point at the bottom outside. For example, in the bowl described above, 2*h*tan(x) must be less than or equal to 1.87" h #1.87/2*tan(7.6) h #1.87/ 0.27 = 7" Mathematically, the outside surfaces of the blank will converge at a point when the compound segments are 7" long. Of course, the segments have thickness. The insides of 1" segments will meet at the bottom if the segments are about 4½" - 5" long. This is too much math for most people. The point is, you should at least make a sketch of your bowl. If you had planned to make compound segments longer than about 4½" in this bowl, you would have been disappointed and you would have wasted stock. (As the saying goes, please do not ask me how I noticed this fact.) ASSEMBLING A COMPOUND BLANK Once the segments are cut, they can be assembled into turning blanks. Assembling them is awkward, because they cannot easily be clamped. Assembling compound segments in pairs is the simplest way to start, and works fairly well, although I prefer clamp pressure (see below). To assemble compound segments by pairs, evenly spread on adhesive (Titebond II) and let the two faces sit apart for a minute or two. If the glue joints are end-grain, the adhesive may soak in, in which case you should apply just a bit more. Soon, the pieces will resist when rubbed together. Push them together, carefully align the edges, and set them aside for an hour or so. You can add clamp pressure by tightly wrapping them with 281 Table of Contents 43-12 Making Segmented Turning Blanks strong rubber bands. Assemble all the pairs this way, then repeat the process with the pairs, until you have made two halves, e.g., two pieces of six segments each. Let the two halves cure overnight. Like the simple miters cut in flat segments, any even number of compound segments must form semi-circular sections (semi-cones) if they will finally fit together into a circle. Therefore, we can again use the trick of assembling half cones and sanding them against a flat surface until they fit. Longer staves, especially, are difficult to control during glue-up. Holding them together with stout tape is very helpful. 1. Lay out the staves on a flat table top with the wide side up and the edges touching all along the length of each stave. 2. Apply strong tape – masking tape is too weak – to hold the staves together. 3. Roll the dry staves into position to form the half-cones and check that they will fit together properly and tightly. 4. Open up the set with the tape (wide) sides down. Apply glue to each mating surface. 5. Then roll the segments together into a halfcone and clamp them or secure them temporarily with more tape. For more clamp pressure (which I tend to think is a good idea), put the two half-cones face to face (without glue). Use spacers as necessary to force the pieces together (similarly to the trick shown in Figure 3) and clamp them over the tape until the glue sets. The fit of my compound segments and staves is occasionally(?) unsatisfactory, even though I cut them with great care, and even though the error (average gap) is in the tenths or even hundredths of degrees. The gaps can be sanded away from the half-cones and the adjustment is not apparent to the eye. I initially sand smaller (10" or less) half-cones on a sanding disk on my lathe, and then finish them by rubbing them on a piece of sandpaper laying on a dead flat board. They then fit together with no gaps. When you glue the half-cones together, use the same methods as described above. Be careful, however, to support the half-cones if necessary so that they cannot rotate apart and open a gap before the adhesive has cured. CLAMPING METHODS. The appropriate method for clamping compound segment constructions depends in large part on how steeply the sides taper. Long, gradual tapers, e.g., a lamp standard, can be clamped with hose clamps, perhaps with a strip of duct tape underneath them to keep them from slipping or denting the wood. Hose clamps will not stay in place on steeper sides. Malcom Tibbetts recommends using an assembly & clamping jig. There are two kinds: 1. Cut circular holes of different diameters into three pieces of plywood. Size the holes so that they can fit part way down the cone-shaped blank, e.g. at both ends and near the middle. Then stack them and drill 3-4 mated ¼" bolt holes around the perimeter. Fit long bolts or threaded rods through the bolt holes. Tighten each bolt down on the segments. This method works best if the slope of the sides is 60E or more and if the segments dry-fit without any visible gaps, so they can be assembled all at once. (I assume that you could use it to glue half cones if you did not apply glue to the half-cone joint surfaces.) 282 Table of Contents Notes and Reflections While Shaving Wood 2. Use glue blocks, cut at the complementary angle to the slope of the sides, with hose clamps, as is done when gluing up segmented rings. For example, glue blocks with a 30E angle will provide parallel clamping surfaces when fitted against a 60E sloping side. A tab is needed at the bottom to keep the caul from sliding up the blank under clamp pressure. MOUNTING A COMPOUND SEGMENTED BLANK Like segmented rings, compound segmented blanks are often open at each end, so normally they would be glued to a solid piece, either another part of the blank or a glue block before turning. (See the lower blank in Figure 5.) The ends of compound segment blanks, however, usually do not have a flat surface suitable for gluing, since each segment is at angle to the axis of rotation. Therefore, you need a way to mount the truncated hollow cone so that you can flatten the end. For a bowl blank, say 5" deep or less, you can mount the blank with Cole (finger) jaws on a scroll chuck. (You could hand sand the end of the segments on a flat surface, but this would be extremely tedious.) If the outside diameter of the blank is too large to fit the Cole jaws, use a jam chuck, something along the following lines. 1. Turn a 4" diameter by 4" long cylinder out of hardwood scrap. (Softwood is too prone to split.) Mount it in a scroll chuck and turn one end to a cone, approximately matching the inside angle of the blank (e.g., 60E.) You want a fair amount of contact between the cone and the inside of the blank. 2. Then make a ¾-1" thick disk with a slightly larger diameter than the opening in 43-13 the small end of the blank. Mount it on the lathe and chamfer the edges of one side to form a steep cone. 3. Return the tapered cylinder from step 1 to the lathe and apply a piece of nonskid rubber foam (or double sided tape – something to increase friction). 4. Jam the blank against the covered cylinder. Put the smaller cone against the small end of the blank and hold everything in place with the tailstock. 5. Rotate this by hand and adjust it until the blank is well centered, and everything turns well. If a lot of sanding was needed to make the half-cones fit together well, the blank will be elliptical in cross-section and may wobble on a conical jam chuck. It’s not safe to tighten the jam chuck too much to overcome the wobble. In such cases, I have made tapered, irregular jam chucks out of hardwood (not pine!) to fit inside the blank so that I could true up the end. Figure 5. Two Compound Segmented Bowl Blanks. Note that the raw ends of the segments are sharp arrises, unsuitable for gluing to the base. The bottom of the larger blank has been planed off on a lathe to make a flat surface for attaching the bottom disc. However you mounted the blank, you now should use very light cuts to plane off the 283 Table of Contents 43-14 Making Segmented Turning Blanks ends of the segments until you have a flat glue surface for the bottom disk. The glue joint has to be a flat and true as possible because you will be mounting the blank from the bottom end, and because the glue joint will show for the life of the bowl. There is no point to shaping the blank further at this stage, unless you can round off the corners of the polygonal blank a bit so that it will fit in the Cole jaws. It will have to be re-balanced later anyway. When the bottom edge of the compound segment blank is flat and smooth, fashion a bottom section to close the blank. The bottom of a compound segmented blank can be solid, but it will look better if you make a polygon with the same number of sides as the rest of the blank. It can be flat (usually) or shallow, say 30E (See Figure 5.) For a flat bottom, the length of each section equals the radius of the circle, so the bottom resembles a pie with, say, twelve slices. This is what I did in the bowls shown in Figure 5. If the rest of the blank contains contrasting woods, the bottom can match that pattern. It is theoretically more correct to cut the segments for the end ring so that the grain direction matches the grain direction of the vertical segments. If you do that, wood expansion or contraction will not be an issue. If the grain in the staves runs lengthwise, while the grain in the bottom runs sideways, wood movement with changes in moisture content could break the bowl apart. Prepare (flatten) the bottom and glue it to the main part of the blank. Finally, either glue on or turn a tenon on the bottom for a scroll chuck. Mount the scroll jaws, mount the blank, and turn the bowl as normal. Well, almost as normal. Compound segments that are ¾" thick require some careful planning and execution for the bowl profile — there is not much wood to allow catches, etc to be turned out of the bowl. Note that the full blank is assembled before it is turned. Flatten and glue the end disk(s) and then finish turning the blank. This is especially important for a long object like a lamp base. If you turn them separately, it is possible that the vertical part will be noticeably out of square. Bowl From A Board The fundamental idea of “bowl from a board” (BfB) is to cut solid or semicircular rings out of flat stock and glue them back together into a bowl blank. (Technically, segment constructions also are ways to make a bowl blank from a board, i.e., flat stock, but are not usually called by that term.) This is normally a very efficient use of wood. Flattening the rings is much easier than with segmented blanks. One of the major appeals of BfB designs is that the board can be a lamination of contrasting wood colors and grains. Very interesting “modern art" patterns can be obtained, depending on how the board is made up and how the rings are arranged. On the other hand, the shape (profile) of a BfB is the most constrained, and requires the most careful planning. BfB bowls tend to have straight sides, even more so than compound segment-constructed bowls. In my opinion, BfB is the most difficult to design and execute well of the four construction methods described here. This is mainly because of the geometrical constraints, the difficulty of cutting accurate half-rings with a band saw or whole rings on a lathe, and the fragility of the blank once it is glued up and ready to be turned. Even a well-made blank is easily shattered by a minor catch. But, in my experience, design and 284 Table of Contents Notes and Reflections While Shaving Wood planning are the keys to success. Bowl dimensions and shapes are more constrained with BfB construction than with other kinds of blanks because the radius and depth of the blank are related. In a BfB design, the depth is somewhat determined by the number of rings, and the number of rings is somewhat limited by the radius of the blank. Design and Layout 1. Decide the approximate dimensions of the bowl. For example, suppose that you would like to make a bowl blank that is 9" in diameter at the top, 3" in diameter at the bottom, and 5" high. These are typical dimensions that correspond to a wall angle of about 60E and the so-called “golden ratio” of height to width equal to 1.67. 2. Decide how thick the walls of the blank should be. The nominal wall thickness (width of the ring) is the theoretical amount of overlap between rings. The minimum design width should probably be about ½", but this leaves very little margin for error in cutting or in centering the rings when you glue them up. In my experience ¾" wide rings are wide enough to accommodate the small errors that are likely to occur in cutting the rings and in stacking them to form the blank. These errors can only make the walls thinner. Thicker walls will allow you to shape the sides somewhat. Also, the sides tend to be more fragile when you are initially roughing them out, compared to other constructions. Making the walls thin may be a false economy. Note that because the rings are parallelogram in cross section they are in a 43-15 sense wider than the wall width. For example, cut at a 60E angle, a ¾" wide ring actually is almost 1½ " wide overall. This may affect how many rings you can cut from a board. Here is a very brief summary of the geometry of a BfB blank: Number of rings N = (r1-r2)/w Height of blank H = N x thickness of board Approximate wall angle = arctangent (H/N) where r1 is the radius of the largest ring; r2 is the radius of the smallest ring; and w is the width of the rings. Now, using a ¾" thick board (after joining, planing, etc) a height of 5" will require 7 rings. In order to cut 7 rings from a 9" square board, leaving a 3" center, each ring can be a maximum of 0.42" wide [(4.5"-1.5")/7], including saw kerf. That’s thinner than the ¾” we wanted. Assuming a 1/16" kerf, and that you can cut and stack rings precisely, you may get walls that are 0.37" wide. That will be tight. If you want a ¼" wall width in the finished bowl, you only have c" to smooth the inside and outside surfaces of the wall, and essentially no extra wall thickness to shape the profile. It may not be possible to make this bowl from one board. Furthermore, if your supply board is thinner than you intended after it has been laminated, flattened and smoothed, you will need more rings. If the finished board happened to be ½" thick after planing, you would need 10 rings to achieve 5" in height, so the wall width would now be 0.3" (minus saw kerf, cutting and stacking errors, etc.) That’s probably not achievable. Now, you can change your specification (accept a shallower blank) or you can find a way to make the blank deeper. There are five ways to make a deeper blank, to avoid the rigid constraints of BfB geometry. The best, in my opinion is by making two boards and cutting them to 285 Table of Contents 43-16 Making Segmented Turning Blanks complement each other. You can also make a BfB blank deeper by using thicker stock. Decreasing the width of the rings (so that you get more rings from a given radius) or increasing the angle of the sides are also theoretical possibilities but they are not consistent with our original specification. . The last two approaches require the walls to be narrower. Of course, you can always tweak the depth by adding non solid or segmented rings to the top or bottom. 3. A this point we have discovered that our original plan was not feasible. We have decided to use two supply boards. (We could have decided to use thicker stock.) Now is the time to make a cross-sectional sketch of the Figure 6. Sketch of Example Blank Using Two Boards blank and the two supply boards. The wall angle is, according to our specification, 60E. The approximate wall angle is given by the arctangent of rise (5 ¾ Figure 7. Four BfB Designs. The upper left and lower right bowls were made from the same board. The lower right bowl was finished with shellac; the other three were finished with acrylic varnish. ") over run (2 27/32") about 63E This calculation is more precise than you need. You can get an equivalent result from the sketch. Eventually, you will lay out the angle on the stock itself. The question at this point is whether you like this angle, whether your bandsaw will cut it, etc. 4. Prepare the boards per your design, e.g., glue up strips of contrasting wood species such as maple and walnut. If you will cut halfrings on a bandsaw, it’s a good idea to put a dark board in the center to help conceal the eventual glue line. Also, keep in mind that the thickness of the rings is very important, and that surfacing the glued-up board may reduce its thickness significantly. (1/16 or so). If you want a 1" thick piece of stock that is flat on both sides, start with 1¼" thick stock and be careful to push them down against a straight flat board as you make up the lamination. Designing the board itself involves deciding where to put the laminations relative to where the rings will be cut. For example, do you want a symmetrical board with a center stripe? (Lower right of Figure 7.) Also, if rings will be cut every ¾” at a 45E angle do you want the cut to be made through a 286 Table of Contents Notes and Reflections While Shaving Wood contrast strip or between them? Will you lay the rings back in their original orientation, so that the contrast stripes remain in line or will you rotate each successive ring to produce a random or “woven” pattern? (Lower left and upper right of Figure 7.) 5. Surface the glued-up stock board on both sides to obtain smooth glue surfaces and uniform ring thickness for when you stack and glue the rings. 6. Use a compass to draw circles on the stock. The circles will be guidelines for laying out and cutting rings. The distance between them should equal the nominal width of the ring, e.g., ¾" plus an allowance for saw kerf or parting tool. They do not have to be evenly spaced. If you widen them near the outside, the bowl will flare somewhat. Make sure that the compass point leaves a center mark on the stock. (See Figure 8) If you will use two complementary boards, the rings on the second board must overlap, as shown in Figure 6. The most direct way to do this is to cut the outermost ring for board#1 (the uppermost board), lay it on board#2, and trace the inside circle. That will mark the outside circle for board #2, and so forth. If you are cutting on a lathe, you will have whole rings. If you are cutting half rings with a band saw you can do the same with half-rings. To calculate the offset, note that each cross sectional parallelogram can be thought of as a rectangle and two triangles. The offset is the base of either of the triangles. This is given by the thickness of the board (¾" in this example) divided by the tangent of the wall angle (63E in this example), that about 3/8" According to Figure 6, board #1 would have a circle drawn with radius 4 5/32" and 43-17 board #2 would have a circle drawn with radius about 3¾". The difference between them would continue across the board. This is for purposes of example. It is not necessary to be this exact, because the wall is thick enough to allow some wood to be removed in smoothing and shaping the walls. Cutting Rings and Half-Rings Rings can be cut on a lathe with a parting tool, on a band saw, or on a scroll saw. Cutting with a normal “diamond” parting tool requires a wide kerf (c"), heats up the tool and the wood, and tends to tear out the back of the board. There is a product called Ringmaster that is available to mount on a lathe or as a free-standing tool. The ringmaster cuts both sides of the board, so tearout is less of a problem. The band saw makes much cleaner cuts and the kerf is 1/16" or less. HALF RINGS ON A BAND SAW The stock is cut in half lengthwise to create two rectangles, and then matching semicircular rings are cut from each of the two halves. Pairs of half-rings are jointed and glued together into rings. Usually, the sides of the rings are cut at the wall angle, to maximize the use of wood. When each pair is glued together into a ring, the ring will be shaped like a short, truncated cone. Alternatively, the rings can be cut at 90E. It is much easier to cut half-rings with the band saw table flat instead of tilted. It is a somewhat less efficient use of wood. 287 Table of Contents 43-18 Making Segmented Turning Blanks The rings are cut from two different pieces of stock so that the rings overlap, similar to the two-board design described above. For example, if the top ring is 8" in diameter and 1" wide, and cut from the first board, the next ring would be 7" in diameter and 1" wide, and cut from the second board. This would give a ½" overlap between the rings, and so forth. Cutting rings on a bandsaw is theoretically straightforward but somewhat challenging in practice. The main difficulty is Figure 9. Laying Out Half-Ring Cuts. Looking down from above: the two halves are side by side in a vise. The concentric circles have been continued across the edges and cut lines have been drawn between the “corners”. aligning the edges, and then planing or block sanding the edges until they joint perfectly– no light shines between the boards when you hold them up to the light. These edges show more prominently in BfB construction than in flat segmented construction. Figure 8. Concentric Circles Scribed On a Built-up Board that the work is difficult to hold in position while the bandsaw table is tilted. Also, the half-rings are cut separately but need to mate perfectly when they are re-joined. Here is the procedure that works well for me: 1. Cut the board in half through the center of the circles. The cut should be smooth and straight. If not, joint the cut edges. An excellent way to do this is by folding the two halves as if they were covers of a book, 2. With the two half-pieces side by side, with their edges aligned and the circles showing on the outside, mark the cutlines for the rings. Accuracy at this step is very important. Because the two half-rings are cut separately, the cut marks have to be as symmetrical as possible. The idea is that the cut lines match the circle lines and that they are symmetrical around the center point. Otherwise, the rings will not match up correctly when you rejoin the rings. (See Figure 11.) a. Mark off right angles across the edges corresponding to the circles you drew in step 4. b. Mark the cut lines from corner to corner. This should form a chevron pattern if the marked sides of the boards are both facing outward (Figure 9). 288 Table of Contents Notes and Reflections While Shaving Wood 43-19 3. Set up the bandsaw using a bevel gauge and cut the half-rings. The semicircles must be cut accurately. A jig is very useful. (See Figure 10 and Chapter 44, Circle Cutting Jig for a Bandsaw.) 4. Check the fit, and re-assemble the half rings on a dead flat glue-resistant surface, e.g., waxed melamine or plywood covered with waxed paper, using PVA glue. Normally, I just rub them together and let then rest undisturbed until the glue has cured. If both ends of the ring do not match perfectly, the ring is not ready to be mounted. When you try to turn a ring in which the end of one half-ring protrudes even a little bit, you risk a bad catch that can shatter the blank. It may be best to sand, file, or saw the edges until they are smoother, with no part of the ring jutting out. CUTTING RINGS ON A LATHE Another method is to cut whole rings on a lathe. This method leaves out the step of rejoining the half-rings. You should take the thickness of the “kerf” into account when you are designing the blank. Cutter and Tool Rest A standard, c" diamond parting tool works OK, but wastes wood. A thin parting tool (e.g., 1/16" thick) is better, but may be a bit too wide to fit into the circular groove you are cutting. I ground one off so that it was only about 1" wide at the end so that it would bind less. I made a cutter out of a small Sawzall blade glued with epoxy into a long handle. (I tried a knife handle but the tool was too hard to control. It tended to tip into the cut.) The Figure 10. Matching Half-Rings From Bandsaw tool is easier to control with the leverage provided by a longer handle, about as long as the handle on a bowl scraper. I mad a cutter out of a small (3/8") spade bit mountd in a long handle like the Sawzall blade. I simply ground the end at about 60E, like any scraper. It worked well. Cutting rings this way requires a fixed tool holder. For a tool rest, I cut about ¼" deep angled grooves in a block of hardwood. The grooves should be about the same width as the parting tool. Mount this on a solid, flat toolrest. Mounting the Stock Cut out a disk from your board the same size as the largest outside diameter of the intended blank and mount it on a glue block. The glue block must have a smaller diameter than the smallest ring you plan to cut, or you will have to cut the blank off the glue block to free the smallest ring(s) and then re-glue it back on. Parting a board into rings at a 45E requires a cut that is 1½ times as long as the board is thick. The parting tool will generate heat. Pause often, so that the wood will not get too hot. PVA adhesive is thermoplastic, 289 Table of Contents 43-20 Making Segmented Turning Blanks and if the wood gets hot enough the board may come apart at a glue joint. (I know this from experience). This is not a disaster, because you can sand off the old adhesive and re-glue it. ASSEMBLING THE BLANK Whether you cut rings on a bandsaw or a lathe, at this point you should have a set of bevelled rings. Glue the rings together in a centered stack, approximately as described above for segmented rings. You can get some very interesting patterns by rotating the rings relative to each other. I sometimes prefer to use a separate solid disk as the bottom of the blank. That disk can be glued to a waste block or held in a four-jaw chuck, etc. Flattening the rings is still necessary, but only light sanding on a flat surface should be required if the board was surfaced before the rings were cut. Arrange the rings in the correct order and orient them according to your design. Accurate centering is often more critical when mounting BfB rings than when mounting segmented rings. Turn the profile and interior of the bowl. Keep in mind that a BfB blank may be more fragile than blanks of other constructions. For example, it is more subject to splitting along the grain if you catch the end of a half-ring. Also, be careful when removing wood – measure more, cut less. It seems very easy to turn through the wall of a BfB blank, because the wall thickness is not necessarily reflected by either the inside or outside surface of the blank, depending on how accurately you cut and centered the rings. Also, the maximum wall thickness may be less than with segmented blanks. If, for whatever reason, there is a hole Figure 11. Re-assembled Half Rings, ready to be flattened and Mounted. in the bottom of the bowl, you can drill out a neat hole, say d" diameter, and cut a matching plug from a piece of scrap left over from the board. If you used a solid disk for a bottom, and turned away most of the original bottom of the blank, this is not necessary. Laminating Stock As mentioned above, the ability to use laminated stock is a main attraction of segmented turning. It is, however, considerably harder to design and execute than it may appear in a magazine article. (And segmented blanks are difficult and fussy enough, even without laminated stock.) As will become apparent below, I have had at best limited success with laminated stock. Designing laminations to produce a desired figure on the surface of a turning requires either experience or an excellent graphical imagination. It is much easier (for me, at least) to follow instructions to make a pattern designed by somebody else. One easily overlooked but important consideration in making laminations is the work-ability of the wood. Layers will normally be cut on a band saw to maximize the number of useful pieces of expensive or 290 Table of Contents Notes and Reflections While Shaving Wood scarce wood. Then, the saw marks must be planed or sanded off. In any case, each layer in a lamination must have consistent thickness and be jointed (surfaced) well if it is to make a strong and invisible glue line. Unfortunately, some woods that turn well are actually quite difficult to plane, while their sawdust can be allergenic or even toxic. Secondly, gluing up laminations from short pieces can actually reduce the number of useful pieces and is often a nuisance. Its better to order longer pieces of more easily worked stock. Save the short pieces of highly figured wood.for another use, e.g., segments. POLKA-DOT FEATURE Dennis Daudelin has described a “polka-dot” feature ring that is made up of two woods of contrasting colors laminated together horizontally. The wood that will appear as the polka dot is on the outside, and the background wood is on the inside. When the polygon (with straight sides) is turned into a circle, some of the background wood is exposed, producing the polka-dot effect. This sounds simple, but the laminated ring has to be centered perfectly, or the dots will not be uniform. It is very difficult for me to lay up ring segments accurately enough. Actually, if the ring is off center by the thickness of the feature wood layer, say c", one side will have dots and the other side will have no dots, showing either the feature wood or the background wood. I have not solved this problem, yet. For a six-sided ring, Daudelin recommends that the outer layer be about ½" wide. For a 12-sided ring, it appears to me that the layer should be about ¼" - d" wide. The procedure is simply to laminate a feature 43-21 wood, say bloodwood, yellowheart, or Osage orange, to a strip of background wood such as holly or maple. The layers should be well jointed, e.g., sanded smooth, and well clamped, because the glue line will be exposed when the wood is turned away. The lamination should be normal width, e.g., about 1" - 1 ½" wide, depending on how thick the ring will be and how much change in diameter it will have to support. So, for example, the lamination would comprise a stick of feature wood ¾” thick by d" wide by 35" long, glued to a piece of background wood ¾" thick by 1" wide by 35" long. (It’s necessary to make plenty of laminated stock because of the way that it must be cut.) After the glue has cured, clean up, smooth, and square up the laminated stock and cut segments. With this kind of lamination, however, the stock cannot be flipped after each cut. The feature wood must Figure 12. Sled for Cutting Zig-Zag Pieces from Laminated Stock (See arranged pieces in upper left) always be on the outside, so each segment must be cut individually. This will waste a bit of laminated stock, which is why you will need to make extra. For a 8" diameter ring, I 291 Table of Contents 43-22 Making Segmented Turning Blanks estimate that you would need about 35" of stock. ZIG-ZAG FEATURE Even a simple zig-zag pattern can be a challenge, as I discovered when I made one. I followed (somewhat) the description given by Tibbetts on p. 101 of The Art of Segmented Wood Turning, or should I say that his description lead me down the garden path. Like the rest of segmented turning, making laminations requires planning and precision. My first decision was the diameter of the zig-zag ring, because that would determine the length of each segment. An 8" diameter ring with 12 segments, would be composed of 2" long segments. I decided on a 45Eangle, which lead me to four pieces per segment. The length of lamination needed for each zig would be the hypotenuse of a right triangle whose base and height were ½", so I’d need 11/16" per zig. Allowing c" for saw kerf, I would need 13/16" for each piece. So, 13/16" per piece x 4 pieces/segment x 12 segments = 39" of laminated stock. It is very important that each segment in the zig-zag ring be identical. The easiest way to do that is to make them all from one piece of laminated stock. For this learning exercise, I chose Radiata pine as the light wood and cedar as the dark wood. I ripped a piece of cedar about 1½" wide and d" thick and then planed it to ¼" thick. I cut two pieces of 1½" W x ¾" thick pine and jointed them for gluing. I glued the three pieces together between cauls, with clamps about every 8". It is necessary to make sure that the lamination is clamped evenly along its length and width, and that the layers don’t slide sideways under clamp pressure, because this will limit the width of the finished segments. When the glue was cured, I cleaned up the lamination with the table saw and hand plane so that it was smooth and square in cross section. Then, using a crosscut sled fitted with a 45E guide and a stop set for 11/16", I cut diagonal slices. (Figure 12.) I kept the pieces in the same order that they were cut from the stock. After the pieces were cut and arranged (alternate pieces flipped over), I glued them together into sets of four or eight. The saw cuts were very smooth, with no saw marks, etc., but I sanded them by rubbing them against 100 grit sandpaper on a flat surface. If I had glued them into larger sets, e.g., sets of 24, it might have been easier to trim them. When arranging the pieces, I focused on the appearance of the zig-zag line, trying to make it line up as well as possible. Any tiny error in the cutting angle will tend to make a long string of pieces wander off at an angle, but this can be corrected in the next step. 292 Table of Contents Notes and Reflections While Shaving Wood 43-23 difficult part of making a zig-zag feature ring. (See Figure 13) Alternatively, you could wait to trim the segments until the rings were made. Then you could glue the segments into the rings so that the zig-zags lined up, but the rings might be very difficult to flatten. References Figure 13. Segmented Bowl with Zig-Zag Feature Ring. Note irregularities in the zigzag design To trim the pieces, I marked guidelines along the points of the pattern and then ¼" on each side, for a border. The outer lines were the guidelines showing where I wanted to cut the pieces. I cut a small strip of Masonite to act as a sled. I put a piece of double-sided tape on the Masonite and carefully arranged the rows of segments on the sled, aligning the guidelines so that the pattern would be uniformly centered along all of the pieces. Then, I lined up the row on the table saw, using an angle guide against the rip fence when necessary, and trimmed the pieces to the same size. The small chips that were removed are shown in the picture to underline the point that a zero-clearance throat plate is necessary for this operation; otherwise, the pieces may jam in the throat of the saw. When cutting the segments on the miter sled, the joint has to exactly bisect the saw kerf. That is, sawing has to remove equal amounts from each side. Otherwise, the “zig” will not line up with the “zag” when the rings are glued together. This was, for me, the most Daudelin, Dennis. Making a Polka Dot Bowl, http://www.woodturningonline.com/assets/t urning_articles/polka_dot_bowl.pdf Frid, Tage. Tage Frid Teaches Woodworking Book 2: Shaping, Veneering, Finishing. Newtown, CT. Taunton Press 1993. Love, Edwin M. Tricks of Inlay Turning. Popular Science January, 1946 (accessed through Google Books) Pritchard, Bob. Building a bowl from a board. http://www.squidoo.com/building-a-bowl-fr om-a-board Raffan, Richard. Turning Wood With Richard Raffan. Newtown, CT. The Taunton Press, 2001. Raffan, Richard. The New Turning Wood with Richard Raffan (DVD). Newtown, CT. The Taunton Press, 2008. Rogers, Jim. Segment Edge Estimate Table http://www.jlrodgers.com/pdf/segment_edge _table.pdf Rogers, Jim. Segmented Turning School. Part 1 - Cut Accurate Segments. American Woodturners, Dec. 2005. 293 Table of Contents 43-24 Making Segmented Turning Blanks http://www.jlrodgers.com/pdf/part-1-cut-acc urate-segments.pdf __________ Part 2 - Planning a Segmented Vessel. American Woodturners, March 2006 __________ Part 3 - Source of Errors. American Woodturners, Sept. 2006. Rowley, Keith. Woodturning: A foundation course. New Edition, Lewes, East Sussex, England. Guild of Master Craftsmen Publications, Ltd. 1999 Sears, Roebuck & Co. The Wood Lathe – An Illustrated Manual of Operation. 1969. (Accessed from OWWM.COM) Segmented Turning. http://en.wikipedia.org/wiki/Segmented_tur ning Tibbetts, M. The Art of Segmented Wood Turning – A Step-by-Step Guide. Fresno, CA: Linden Publishing, 2003. Arc tan (.875) = 41.19E 2. Let the number of sides be represented by s. The vertical angle of the sides is b. a=360/s x=arctan((cos b)*tan(a/2)) y=arcsin((sin b)*sin(a/2)) The "x" value will be the desired angle of the miter (cross cut). The "y" value will be the desired (tilt) angle of the saw blade relative to the saw table (horizontally). For example, to make a octagonal barrel with parallel sides the angle has to be 22.5E In the example, given 12 sides, a = 360/12 = 30E x= arctan((cos(41.19)*tan(30/2)) = arctan((.753)*(.268)) = arctan(.202) = 11.41E y= Appendix: Compound Miter Calculation The following formula will give the miter gauge setting, and the head (tilt) angle for any number of sides on any angle you choose. 1. Calculate the vertical angle of the side pieces, measured from the horizontal. For example, if the sides were vertical the angle would be 90E. Suppose, for a bowl, the bottom diameter is 2½", the top diameter is 6" (rise= 3½") and the bowl is 4" high. The tangent of the angle is rise/run = 3.5/4= .875. arcsin((sin(41.19)*sin(30/2)) = arcsin (.659*.259) = arcsin(.171) 9.83E By the way, for a 45E vertical angle and 12 sides x= 10.7 and y = 10.5. For a bowl with 12 straight sides, x=0E and y=15E Trig formulas from Robert Smith, http://groups.google.com/group/rec.woodwo rking/msg/ebffc02b598816cd, accessed September 5, 2010 Calculator available at many sites on the web. 294 Table of Contents Notes and Reflections While Shaving Wood Chapter 44 – Circle Cutting Jig for a Bandsaw A jig for cutting circles is fairly easy to make and greatly increases the speed or accuracy of cutting. Sure, you can trace a circle on the stock and cut it freehand. You cut 1/32"-1/16" outside the line and then cut to the line as you remove the saw marks with a spokeshave or block plane. Using a jig is quicker, although you still have to fair the cut. Also, practical difficulties often intervene, especially if the workpiece is difficult to control, e.g., mitered, as in the rings for Bowl from a Board constructions. A main use of a small circle jig in my shop is for cutting turning blanks, especially the rings for “bowl from a board” (BfB) turning blanks. These semi-circular rings must be cut at an angle with good accuracy. Because the table is tilted, this is very difficult to do without a jig, because the work tends to slide downward slightly and bind the blade. The simplest circle jig comprises a nail through a piece of plywood, which is appropriately positioned and clamped to the bandsaw table. The plywood should be approximately the same size as the right-hand half of the bandsaw table. You make a small hole at the center of the circle you intend to cut in the workpiece, put it over the nail, and rotate the workpiece against the bandsaw blade. This works if the nail is positioned along a line perpendicular to the plane of the blade, and lined up with the cutting edge. That is, the blade should be tangent to the circle you are cutting. That means that the supply piece should be square. A much nicer jig would be adjustable, have a smooth, low friction surface of shellac or Melamine, and offer a choice of fine and thicker pivots, e.g., a ¼" dowel in addition to a nail or sharp screw. Adjustability is easily 44-1 Table of Contents Figure 1. Circle Jig on 10" band saw accomplished with a slider set perpendicular to the saw blade. The slider can run in a groove routed in the auxiliary table or can be dovetailed. The slider can be clamped in position with a clothespin clamp. If it s dovetailed in it can be locked by means of a 1/4” stud running in a threaded insert. Figure 1 shows an example. The jig is made up of an auxiliary table (two layers of ¼" oak plywood) a dovetailed slider, clamps and positioning blocks as shown. The pivot is a sheet metal screw sharpened to a fine point. The correct way to use this jig is to adjust the pivot to the desired radius and to fix it in place. Then put the workpiece on the pivot. Then slide the workpiece into position, until the positioning block is up against the table. If the workpiece is not square, the saw should be turned on so that the blade cuts across the board until it is tangent to the circle. Use mild pressure forward and to the left to keep the cleat up against the saw table. Then you can rotate the workpiece to cut the disk. I made a larger jig using the same principles, with two refinements. The 295 44-2 Circle Cutting Jig for a Bandsaw positioning block has an adjustment screw in it for fine-tuning the position of the blade tangent to the circle. A strip of hardwood is attached to the bottom of the jig to run in the miter slot on the saw table. That eliminated the need to hold the jig against the table during the cut. Half-Circle Jig My half-circle jig is an attachment for the circle jig. It takes advantage of the fact that the rings for a BfB blank can be cut in order, from larger diameter to smaller, and so they can be cut around a pivot point. I drill a ¼" hole though the center of the stock. Then I make a simple disk with a dowel through the center. It looks a bit like a toy top, except that the bottom side is flat. I jam the center of the dowel onto the sharp pivot point of the circle cutting jig, and then firmly register the workpiece against the dowel. The disk serves as a turntable for the ring while I am cutting it. I can cut larger rings (e.g., 4" radius) without the “turntable”. I just use a ¼" dowel as a pivot. Smaller rings Figure 2. Ring Cutting Attachment for Circle Cutting Jig are very difficult to hold in position and the large dowel seems to help. I can safely cut rings as small as 1½" radius with this setup. I don’t think smaller rings would be safe to cut while being held with the fingers. I don’t need them for my work, but if I someday needed to cut smaller rings, I could easily attach a small cam clamp to the turntable. 296 Table of Contents Notes and Reflections While Shaving Wood Chapter 45 – Making a Sled for Cutting Flat Segments C utting segments for ring or stave turning blanks is repetitive. For example, a medium-sized project angle of the fence are easier with a wide sled, as explained further below. There should be a bridge at the back of the sled to hold the base together after a kerf is cut through it. The bridge can be a solid rectangle, but an opening in the center, as shown in the illustration, would allow wider pieces to be cut. A bridge at the front would also help the sled stay flat. The workpiece would be positioned toward the middle of the sled, so it is not likely that the sled would be pushed too far forward, such that the saw blade would reach the rear of the sled. Nonetheless, the operator must be able to slide the sled without placing his hands in the path of the sawblade. The cam clamp, as shown in the figure, makes a useful handle for pushing the sled. Note the two 2x4 pieces and the red “x” marks in Figure 1. The fence should be about 1½" wide Baltic Birch or maple, something that will not split when the mounting screws are tightened down. (I used pine once and regretted it when it split.) The sled requires two holddowns, one to hold the stock while cutting, and the other to hold a stop block in place. It will not be safe to Figure 1. Example of a sled for cutting flat segments. might comprise eight rings of 12 segments each. Also, the segments must be cut very accurately. The angles of the segments must add up to 360E or the ring will close with a gap. This calls for a well-made sled. The sled for a table saw (See Figure 1) should be made of ½"-¾" Baltic Birch plywood or the equivalent, with two phenolic or hardwood runners for the miter guides on the saw table. (For more on making a crosscut sled, please see Chapter 16.) The main part of the sled should be about 24" wide by 12" deep. Although the sled can be larger or smaller, precise adjustments to the 45-1 Table of Contents 297 45-2 Sled for Cutting Flat Segments cut the last few segments while holding the work with your fingers. I prefer to position the hold down so that it holds the piece itself (which will be cut off) rather than the feed stock. The foot of the hold-down may be too wide for very small segments, however, and it may be necessary to hold the feed stock instead. A stop block is, of course, essential if you are to cut the segments to uniform length, and a hold down will make it easier for you to adjust its position. A sled for cutting compound segments is basically the same, except that the saw blade is tilted at the appropriate angle. See Chapter 43, Blanks for Segmented Turnings. Longer staves cannot be crosscut and must be ripped. See Chapter 46, Sled for Cutting Staves on a Table Saw. 1. Make the runners to the theoretical dimension, e.g., ¾" wide, and then lightly plane or sand them until they just slide easily with no side play. It will be important that the sled has no side-to-side slippage in the miter slots. Place the runners in the miter slots, apply double-sided tape to them, and carefully lay the sled base on top of them. Use the rip fence to square up the sled base to the runners. You could use a light coating of PVA glue instead of tape, but, if you do that, protect the saw table with a coat of wax to prevent glue from sticking to it. When the runners are securely bonded to the sled, slide the sled off the saw table, flip it over, and screw the runners to the sled. Be sure to trim off any extra tape or clean off the dried glue. 2. Attach the right-hand end of the fence to the sled with a flathead machine screw, countersunk and passed from under the sled, through the fence. 3. You can closely approximate the angle of the fence with a protractor or with trigonometry, but the final angle must be found by trial and error. 4. Lay out a right triangle on the sled. The front edge of the fence will Figure 2. Layout of Fence. The imaginary right triangle described in the text is shown in white. be the hypotenuse, and one apex will be near the pivot. Measure a distance from the apex along an imaginary line parallel to the edge, e.g., 16". (See Figure 2.) 5. The angle needed is given by 298 Table of Contents Notes and Reflections While Shaving Wood 45-3 180 divided by the number of sides your ring will have, e.g., 180/8=22.5E for an octagonal ring. The ratio of “rise” over “run.” gives the tangent of the angle. For the 22.5E angle needed to make octagonal rings, the tangent equals 0.4142. Therefore, the distance from this imaginary line to the front of the fence should be .4142 x 16 = 6.627", i.e., 6 e". (See Figure 2.) There is an alternative to calculating with trigonometry. You can buy setup gauges that are cut to the precise angle you need. For example, see Seg-Easy a An additional advantage to using set-up gauges is that one sled can be used for a variety of angles. 6. Tack down the left end of the fence. You may need to adjust the fence by trial and error, based on the test cuts. See Adjusting the Fence, below. A stop block along the fence is necessary to make every segment in a ring the same length.b (Segment length does not need to be exactly as calculated.) If rings of different sizes will be needed, an easily adjustable stop is very convenient. A cam clamp is one way to hold an adjustable stop securely. A hold-down located between the saw blade and the stop is mandatory. You can safely hold longer pieces of stock against the sled fence on the left-hand “feed” side, but you should never allow your fingers to get closer than 3" from the blade. Therefore, the final segments cut from each piece of stock will be too short to hold safely with your fingers. Another disadvantage of holding the supply piece with your fingers is the tendency for it to slide into the saw blade at the end of the cut. This may ruin the edge needed for the next piece. A stop on the saw table is very useful in preventing the fence from traveling too far as you repetitively cut segments. The easiest way is to clamp a block of wood to the rip fence of the table saw. Alternatively, you could clamp a block of wood to the rear of the saw table itself to serve as a stop for the sled. a Adjusting the Fence (http://www.segeasy.com/segeasy.htm) b Actually, segments of different lengths can be used in repeating patterns. Then, each set of segments, e.g., pair, triplet, must total the same length. Note that an even number of segments, e.g., 8, can be glued up into two semicircular rings. For a rough, 299 Table of Contents 45-4 Sled for Cutting Flat Segments initial sled adjustment, cut half a set of segments out of sound scrap wood (like a piece of sound pine or poplar) and assemble them into a half-ring. Lay the side of the half-ring against a straight edge and measure any gap between the side and the straight edge. Or, dry-fit a complete ring in a band clamp (large hose clamp). Push the segments together tightly and then measure the remaining gaps between the semicircles, or measure each gap with a feeler gauge and add up the gaps. Hold the ring up to the light to reveal any gaps. If gaps appear at the outside of the ring, the angle of the fence was too acute (too narrow) and needs to be increased (widened). If they appear at the inside of the ring, the angle is too wide. If the gaps are not symmetrical, e.g., if the gap is to the outside on one side and to the inside on the opposite side (or if the ring does not close at all, showing one side with a small gap while the other has a rectangular gap), it simply means that you have flipped one of the half-rings over. Just flip it again. This following approach can significantly reduce the number of test cuts needed to get the fence to the exact setting, so consider it even if you hate math. Wiggle each segment to make it fit tightly against the adjoining segments. You are trying to accumulate all the angular error in one place, at the end of the semicircle. Hold the half-ring against a straightedge and measure the gap. (Or, assuming all gaps are on the same side, measure each gap and add them all together.) Calculate the average gap by dividing the total of all the gaps by the number of segments. Then, divide that number by the width of a segment. This gives you the ratio of rise/run that you need to adjust the sled. (The angular error is the arc tangent of this number, if you are a trigonometry enthusiast.) Then repeat steps 5 and 6. Move the left end of the fence a tiny amount. For example, my first attempt to make an octagonal ring had five gaps (three joints were tight). The gaps were about .006" each. The average gap was therefore (5 x .006")/8 = .0045" The “rise” is .0045, and the run is the width of the sled from the pivot to the end of the fence, in this example, 16". (Note that the fence, corresponds to the hypotenuse – see Figure 2.). The segments were 1c" wide, so the ratio of rise/run (tangent) of the average gap angle is .0045"/1.125" = 300 Table of Contents Notes and Reflections While Shaving Wood .004. The rise was 16". I used this to estimate how far to move the end of the fence. 0.004 x 16" = 0.064", i.e., 1/16". I rotated the fence by moving its left end 1/16" and tacked it down again. I cut another eight segments. When dryfit together, no gaps were visible – no light could shine through any joint. “Perfect” accuracy (no light visible anywhere around the ring) is seldom achievable, except by luck. Because an even number of segments resolves into two semicircular rings, small (1/32"-1/16") gaps between the semicircular rings can be sanded away. Glue up the half rings and then sand them against a flat surface or a flat sanding disk until the two semicircles mate perfectly. This may distort the ring, and does take time, however, so it is usually worth the time and trouble to adjust the sled to cut as accurately as possible. As the number of segments increases, or the width of the segments increases, the accuracy needed increases. For example, an error of 0.1E in an octagonal ring would yield a gap of about 1/64", while the same error in a ring with 24 sides would yield a gap of about 1/16". The one is nearly insignificant, 45-5 and easily sanded away, while the other is significant. Therefore, a sled for cutting more acute (shallower) angles probably should be wider, e.g., 24", than one made for cutting more obtuse angles. Small adjustments are difficult to make accurately. They are easier to do if you use a spacer as thick as the amount you want to adjust the end of the fence, in this example, I used a shim 1/16" thick. Clamp a block of wood to the side of the sled so that it is against the shim, which is against the end of the fence. Then remove the shim and move the fence until it is tight against the block of wood. Then secure the fence. Compound Miter Saw Flat segments can be cut on a miter saw almost as accurately as they can on a table saw. The angle of the blade, rather than a secondary fence, is adjusted. The equivalent device for a miter saw is not a sled, because it does not move. It is simply a fixture to safely hold the work piece in place. (Figure 3) 301 Table of Contents 45-6 Sled for Cutting Flat Segments Figure 3. Fixture for a Miter Saw The same principles apply, with some changes in emphasis. The miter saw will cut though the rear fence of the fixture. The fixture is attached to the built-in fence of the miter saw on both sides of the saw cut, but it need a bridge at the front to hold it together. Therefore, the fixture must be deep enough to accommodate that bridge. A means of holding both the supply piece and the cutoff piece (the segment) is essential because the miter saw blade is normally lowered or slid into the cut. If it is allowed to rise back up after the cut, that gives the blade two opportinities to contact the wood. Some people recommend waiting for the blade to stop rotating before releasing it to rise back up. I recommend fixing both side of the cut. Reference Neely, K. Plans for a Frame (flat) Miter Sled. http://www.turnedwood.com/framesl ed.html 302 Table of Contents Notes and Reflections While Shaving Wood Chapter 46 – Sled for Cutting Staves on a Table Saw L ong beveled staves are useful for making turning blanks for lamp standards, table legs, buckets, and other deep narrow vessels. I will call a simple stave one that is beveled along both of its long sides but not tapered. When you fit simple staves together you get a cylinder. A compound stave (compound angle stave) is beveled along both of its long sides and tapered. When you fit these staves together, you get a cone or truncated cone. The trigonometry for these staves is the same as for flat and compound segments. In particular, the trig calculations for compound staves simultaneously involve both bevel and taper, just as with compound segments. The additional complications for cutting compound staves are: (1) you have to rip the long narrow stock, so the accustomed compound crosscut miter sled won’t work; and (2) the angle of the reference edge changes after the first cut, so the angle of the taper guide (secondary fence) has to change in order for the stave to be symmetrical. Regarding point 2, what I call the first cut is made on a rectangular board or billet: the reference edge is square to the ends. You set up the taper jig for the taper you want. To make the second cut, however, the reference edge is now tapered. The cutting angle has to rotate from a few degrees left of 0Eto a few degrees right. The main issues in cutting beveled staves are to get the tilt of the saw blade exactly correct so that the staves will close into a circle with no gaps and to avoid binding, burning, and kickback. The first issue is ameliorated somewhat by the fact that any even number of staves can be assembled into half cylinders (or half cones). The edges of the half cylinders should be flat, i.e., coplanar, so should rest solidly on a flat surface with no gaps. If they are not flat, they can be sanded on a flat surface until they are. (See Chapter 45, Sled for Cutting Flat Segments.) Two half cylinders prepared in this way will fit perfectly. Long half cylinders, however, can be difficult to sand perfectly flat. The second issue is resolved by using a sled under the main body of the work piece. (Figure 1) As the bevel is cut off, the thickness of the 46-1 Table of Contents 303 46-2 Sled for Staves sled base gives it space to fall away from the saw blade instead of binding. Sled for Simple Staves Figure 1. Straight Bevel Jig A sled for cutting beveled, nontapered staves on a table saw (Figure 1) is straightforward to make. The sled simply consists of a base made of plywood or MDF that slides along the rip fence, with a hold down to keep your fingers away from the saw blade. The hold down should be robust enough to keep the work from sliding away from the blade, rotating, or lifting up during the cut. The knob or cam clamp used for the hold-down can be made to be the handle for pushing the sled along to make the cut. The secondary fence is meant just to rub against the rip fence on the table saw and to support the workpiece. It should be taller than the thickest workpiece you expect to cut with the sled so that a beveled edge will bear against it without riding up. If you were cutting staves without a sled, you would use the rip fence to gauge the width of the cut and to guide the work. Since you are using a sled, you need another way to gauge and guide the workpiece. The most direct approach is to make the sled and then trim it to the desired width of cut. (This approach is shown in Figure 1.) That way the right side of the workpiece rides against the secondary fence. The disadvantage is that the sled might not be useful for much wider staves. It would be too complicated to make an adjustable secondary fence (even for me!) You can make a wider sled somewhat adjustable simply by cutting a spacer to go between the secondary fence and the workpiece. Or, just make another sled whenever you need to cut wider staves. The sled should have a push tab or stop block – a piece of wood protruding to the left that will push the 304 Table of Contents Notes and Reflections While Shaving Wood workpiece along with the sled. Attach a handle and push tab (stop block) at one end of the sled. The handle should allow you to slide the sled firmly against the rip fence without your fingers coming within 3" of the saw blade at any point. A piece of Figure 2. Compound 2x4 on edge that Stave. (Light color shows s p a n n e d t h e wood to be removed to width of the sled make stave) from the saw path to the secondary fence worked OK. Screw it to the base. Keep the screws well away from the saw path. Mark a reminder on the handle about where to hold the sled and/or where not to put your fingers. To cut the stave, simply put the work piece on the sled and cut the first bevel. Then flip it end for end horizontally and cut the second bevel. An alternative way to cut these long bevels is on a router table with a miter bit, but if you have many to do, I feel that the table saw method will yield better quality cuts. 46-3 Compound Taper Jig A compound stave is tapered in two dimensions, as illustrated in Figure 2. The sleds to cut these staves are somewhat complicated to make and set up. The typical commercial taper guide can’t produce repeatable tapers on both sides of the stave, and can’t hold a narrow piece securely. Holding it with fingers would be foolhardy. A sled allows a hold down to clamp each piece in place to provide repeatably accurate cuts. I took this up as an interesting challenge, but you may not find it so. These sleds can be enough trouble to make that you should consider making simple staves on the table saw (as described in the previous section) and then tapering the assembled piece (e.g., with a lathe, table saw or jointer). This would work well for a solid form like a lamp standard or table leg, but not as well for a large hollow form like a conical bucket. Furthermore, I had a lot of staves to cut and so I made a refined (and complicated) sled. After you read this (if you do read this) you use the ideas to improvise a sled that suits your needs. A sled for cutting compound staves is a variation on the sled 305 Table of Contents 46-4 Sled for Staves described in the first section. A pivoting taper cutting guide has been added to the sled shown in Figure 1. compound angle calculations are given in the Appendix to Chapter 43. Higher-end construction calculators have the function built in. The design (compound angle) calculations should have given the following specs: 1. Length of stave L 2. Width W at widest part after cutting 3. Width N at the widest side of the narrow end 4. Bevel angle (saw tilt) B 5. Taper angle A Table 1. Layout Dimensions for Guidelines Figure 3. Compound Taper Jig in Position 1 The taper guide should be a bit thicker than the workpiece so that a beveled edge on the workpiece will not lift up when you press it against the guide. The hold down should prevent this, also. The hold down should be sturdy because strong clamp pressure may be needed to hold the workpiece against the narrow shelf between the taper guide and the edge of the sled. Setting Up A very accurate setup is essential to cut accurate staves. Formulas for Guide Distance from Saw Line Front of Sled (Top of Figure 3) Back of Sled (Botto m of Figure 3) Tangent (Ratio to lay out guideline) Line 1 ½(W-N) W (W+N)/(2L) Line 2 W N (N-W)/L Line 3 N W (W-N)/L For an example, here is the setup to cut staves that are 16½" long, 1-7/32" wide at their widest end, and 21/32" wide at the narrow end (wider side). The theoretical blade angle is 14.97E and the miter angle is 0.93E 306 Table of Contents Notes and Reflections While Shaving Wood 46-5 Figure 4. Diagram of Adjustable Sled (theoretically) or .976E if you round off to the nearest 1/32". The geometry involved is straightforward, although I found it confusing for a while. (See Figure 4 and the Table). There are three imaginary right triangles involved, all with “run” L. Note that the cutting guide (which is the hypotenuse of the triangles) swings though an angle from a few degrees left to a few degrees right of zero (negative to positive, if you like). It might be simpler to make 2-3 tapered cutting guides that fit between the secondary fence and the workpiece, as shown in Figure 5. The description that follows, however, is for a pivoting guide with spacer blocks. Figure 5. Compound Taper Jig in Position 2 1. Layout Line for First Cut Line 1 is the guideline for the first cut into untapered stock. The widest cut you will make is W (e.g., 17/32"). From looking at Figures 1 and 2, we can see that the first cut from a 307 Table of Contents 46-6 Sled for Staves square end will remove ½(W-N) (e.g., 9/32") from one end of the blank, tapering down to zero wood removed at the other end. Mark a point 1- 7/32" from the saw track, at the stop block. Then measure up 16½“ along the saw track and ½(W-N) units to the left. To check, you can lay out the cut marks on the blank, as if you were going to cut it by hand. If you mark the workpiece and lay it against the saw track, the opposite edge will give you the position and angle of the taper guide for that cut. The tangent of the angle equals “rise” divided by “run”, e.g., 15/16"/16½“ = .0568, corresponding to an angle of 3.25E. 2. Layout Line for Second Cuts Line 2 is the guideline for cutting a piece that already has a taper cut on one side. The second cut leaves a width of W at the front end and a width of N at the other end, so the “rise” is (N-W), e.g., -9/16". The tangent is (-9/16" /16½”) = .0341 and the corresponding angle is -1.95E. The angle slopes in the opposite direction from line 1 because the board is rotated end-for-end. Alternatively, measure ½(W-N) units (e.g., 9/32") to the right of position 3 and measure (W-N) units (e.g., 9/16") to the left of position 2. (Check: position 3 is now W units from the saw path so that the saw blade will not cut anything from the corner. Position 1 is (W-N) units from the path, so the saw blade will cut the taper to final width N.) The two layout lines will intersect. 3. Third Layout Line The third guideline is not needed in the one-step procedure (see below). It would come into use only for cutting the bevel on one side of a pre-cut taper. It does not remove any width, so it has to be W at one end and N at the other. Therefore its base is (W-N) and the angle is +1.95. Unfortunately, this angle can’t be achieved using the pivoting guide described below. It would require a separate sled. 3. Pivot Especially with the one-step procedure, it may be necessary to switch frequently from line 1 to line 2 and back. A pivot is very convenient for this, but it has to be in the correct location. a. Measure the distance from the stop block (at the bottom of 308 Table of Contents Notes and Reflections While Shaving Wood Figure 3) to the point where the two layout lines intersect. (Call this distance d.) b. With the taper guide held down in position along the first layout line, make a mark near the center of the guide, exactly d from the stop block. c. Drill a 3/16"-¼" hole through the guide at this point. d. Return the guide to the position it was in step d and mark the sled for a corresponding hole. e. Drill that hole with a countersink on the bottom of the sled and screw the taper guide to the sled. Now the guide should pivot between the two layout lines. The edge of the guide does not have to be on the layout lines but it has to be exactly parallel to each of them. That is, it has to provide two accurate cutting angles. 46-7 guide to the secondary fence at each end of the fence. They do not need to be any particular width, but the maximum width is the distance from the rear of the guide to the secondary fence (See Figure 4.) Then slide these blocks until they are firm against the guide and the guide is in the correct position, i.e., lined up with layout line 1. Draw a mark across the spacer and the secondary fence and label it. Move the guide to layout line 2 and repeat. 4. Hold Down – Install a sturdy hold down near the pivot. A 5/16" threaded insert into a 1" thick hardwood guide seems strong enough. 5. Spacers Cut small blocks of wood to fit between the guide and the secondary fence. Set the guide to layout line 1 and measure from the back of the Figure 6. Actual Sled on Table Saw in Position 1 The spacers and the pivot screw 309 Table of Contents 46-8 Sled for Staves will hold the guide securely in position and allow precisely repeatable cuts. See Figures 6 and 7. Cutting the Staves ONE STEP OR TWO? Some authorities recommend that beveled tapers be cut in two steps: taper first, then bevel. Some one-step procedures can burn the work if the saw is slightly mis-adjusted or if the workpiece is able to bind the saw blade during the cut. The sled described here can cut beveled tapers in one step, at least on my table saw. The one-step procedure cannot, however, efficiently use wider boards. In the example we have been using, two 1 7/32" staves require 2" from a wide stock board with a twostep process (allowing c” for kerf.). Using a one-step process, two staves require 2 9/16", more than 25% more. That’s because cutting the tapers before the bevels allows the tapers to be cut in an alternating sequence that leaves less cutoff (waste) stock. So, the one-step process saves time but may waste wood if the stock lumber is in wide boards. The twostep process takes longer (and may require two sleds) but can save wood. Figure 7. Actual Sled on Table Saw in Position 2 TWO-STEP PROCEDURE The two-step procedure would usually begin with a wide board of length L. The saw blade is not tilted when the tapers are cut. a. Make the first cut on the square end of the board with the guide set in position 1. b. Rotate the guide to position 2. Cut the taper on the other side of the first stave. c. Flip the stock board end over end and cut the next stave. d. Cut the remaining staves 310 Table of Contents Notes and Reflections While Shaving Wood from the stock board adjacent to each other using position 2, flipping the board after each cut.. This alternates wide and narrow ends and yields more staves from a given width of stock. After all the tapers are cut, a. Mark the bevels on the wide end of one of the tapers. b.. Tilt the blade to angle B (e.g., 14.97Eleft of vertical or 75.03 right of horizontal). I use a digital tilt gauge. A triangular gauge cut out of plywood or a sliding bevel gauge (laid out with trigonometry) should be a bit more accurate, but in reality it is a bit difficult to see the gap between the saw blade and the gauge. c. The taper guide should be at position 2. Position the sled according to the layout line. Adjust the saw path to bevel the taper without removing any wood from the widest side. (When you make this adjustment, you will see that the tilted blade makes it difficult to measure from the wide side of the stave. It is more accurate to set the saw line at the bottom (narrow) side.) 46-9 d. Cut one bevel on each taper. e. After the first edges are beveled, you can either use another sled with a guide set up to line 3 (as described above). If you used cutouts for your taper guides, you can use the third cutout. Or, you can move the rip fence to the other side of the blade and rotate the sled 180E. Another sled with just one layout line should be easy enough to make and is the preferred way. It must have a taper guide (fence), stop block, hold down and handle. Set the guide to cut N units at the front end and W units at the rear end. Rotating the sled to the left of the saw blade seems easier, of course, but is an unreliable and potentially unsafe method for ripping a bevel because the saw blade is tilted into the work. The blade may bind and burn the workpiece, or even kick it back. (Note that up until now, the blade was always tilted away from the work and the cutoff was free to fall away from the blade.) If you were planning to sand the glue surfaces before joining the staves into half-cones, perhaps a small amount of burning would not be a problem, but think about the whole sled being kicked back at you and you 311 Table of Contents 46-10 may decide not to cut this way. If you decide to cut this way, despite the risks, you will need to add a safe way to push the sled and hold it down. The sled will be rotated 180E, and run along the right hand side of the rip fence. What was the front of the sled is now the back and you will be moving it in the “opposite” direction. Also, you must add a stop block, because the blade will now be pushing in the opposite direction. You can just screw a block on the edge of the sled. ONE-STEP PROCEDURE The one-step procedure begins with rectangular billets of length L and width W or wider (but not wider than W+N+ saw kerf). To cut the staves in one step, tilt the blade to angle b. (See above) Setting the sled for width of cut W is a bit tricky because the blade is tilted. Probably the easiest way is to lay out the cut lines, including the bevel angle, on one of the billets and then set the sled (in position 1) to cut those lines. After the first cut, move the guide to position 2, rotate the workpiece 180E horizontally, and cut the second side. Don’t flip it over. The wide side stays “up.” Check that the bevel angle will be in the right direction. You need a trapezoidal cross section, not a parallelogram. You will again cut off ½(W-N) units from the other side of the narrower end, but now the narrower end will be toward the front of the saw. (See Figures.) 312 Table of Contents Notes and Reflections While Shaving Wood Chapter 47 – Shaker-Style Ladder-back Chairs C only one spokeshave, with a convex sole, for fairing concave curves on back slats. I could have used a block plane to smooth convex shapes and sandpaper to shave off the tenon end of a rung. But I indulged myself with three spokeshaves: flat, convex, and concave, and I was happy to have them by the time I had finished my third chair. My first lathe, which is quite adequate for turning table legs, was limited to about 32" between centers. That’s 10-12" too short for turning the rear post of a full-sized ladder back chair. You may face the same issue. I had been persuaded by some “biggeris-better” lathe enthusiasts that a shop-made wooden extension would not be acceptable to turn longer spindles. A commercial bed extension was out of the question (the lathe was made in the late 1930's). But I considered that people can (and do) turn wood on almost any sort of tool that will rotate wood. The refinements are nice, but just a matter of taste. I made a wooden bed extension for the tailstock out of a block of oak and screwed it to the heavy wooden lathe table top. The tailstock of the lathe mounts on a suitably-positioned lag bolt. I made a tool rest for the extension, which also mounted to the table top. The tool rest is as long as the bed extension. I set the height to slightly above the centerline of the turning, anticipating that I would be turning 1½" diameter posts. The height is adjustable only if I remove and re-install the oak rail that is the actual tool rest. Turning a 42-44" long by 1½" dowel also requires a steady rest. I made a simple birdsmouth steady rest as described by Ernie Conover in a video on FWW.com. An adjustable commercial steady rest using hairmaking is quite specialized. Many chairs, Shaker-style in particular, require turned parts, so add spindle turning to the necessary repertoire. In addition, chairmaking requires some tools, jigs, knowledge and skills beyond those needed for other kinds of woodworking. When I started making chairs, I felt like a novice, a bit like I did when I started woodworking itself, and again when I was learning how to turn on a lathe. I was doubtful about getting into chairmaking. I wondered whether I would be able to do it well enough that I would want to continue, and whether I really wanted to invest significant time and a bit of money making a steam box, bending jigs, drilling jigs, etc., just to get started. Learning to make chairs sparked my interest and enthusiasm, however, just as woodworking had, and each step drew me further in. A necessary skill in making ladderback chairs is the ability to visualize the geometry of the chair as you carry out individual tasks -- in particular, getting mortise holes drilled on the correct side at the correct angle. Round chair posts do not provide sufficient visual clues as to where a hole should be and the correct angle. I made three chairs to teach myself the rudiments of this sub-specialty: the “Simple Chair” described below, the same chair in a slightly larger size, with curved back slats, and finally a rocker. New Tools and Jigs I needed spokeshaves to fair back slats after I cut them out on the bandsaw and to trim rungs to exact dimensions. Perhaps I really needed 47-1 Table of Contents 313 47-2 Ladderback Chairs rollerblade-style wheels would cost about $150 including the stand. The list of new jigs is extensive: drilling and mortising jigs for dowels, a simple wooden quick-release vise for holding rungs while I shave or sandpaper them to fit, slat bending and post bending forms, a steambox with steam source, a flat wooden “needle” for weaving the tape seats, and maybe drying racks if I wanted to bend slats or posts for more than one chair at a time. rail to the front post is 81E then the angle of the same rail to the rear post has to be 99E. So you can make a 9E wedge to get the correct angles on a drill press – turned one way it gives an 81E angle and turned the other, a 99E angle. (See Figure 1. DRILLING FOR RUNGS A Simple Chair While researching the topic of chairmaking I found Chairmaking and Design by Jeff Miller. He has one Shaker style ladder-back design, a child’s chair. (See Figures 5 and 9.) It’s fairly simple, but typically elegant, in the style of Shaker furniture. The posts are straight, and it can be made from furniture-grade dowels. I made one for practice out of birch. Birch is nice hard wood, it finishes well, and birch dowels are quite inexpensive. This was a valuable exercise. Miller’s directions on how to build this chair and how to make the necessary jigs are very helpful. That chapter, and the exercise of building this chair, provided me with a decent introduction to the craft. If you want to build this chair, I highly recommend that you read that chapter. Rather than repeat all of the instructions here. I’ll just add some comments that I think are important for making these chairs. First, the concept that Miller calls the “master angle.” In a trapezoidal chair seat, since the front and back rails are parallel, the angle between the side rails and the front posts and the side rails and the back posts must equal 90E plus or minus the same amount. For example, if the angle of the side Figure 1. The Master Angle The drilling jig Miller recommended worked well. (Figure 2.) It’s just a piece of plywood with a 45E groove cut in the center, with a clamp to hold the post in place. The strips along the sides act as stops. They prevent rotation of the post beyond 90E. They should run the length of the jig, or at least run on both sides of the clamp. The short piece of dowel sticking up is the index pin, mentioned below. At first, I tried to mark the positions of all side, front and back rung mortise holes on the posts. The posts are round and the holes are not at 90E to each other. It’s too easy to rotate the post and mix up directions. I think the right way to do this is as follows: 314 Table of Contents Notes and Reflections While Shaving Wood 1. Cut the posts to length. (Actually, cut out posts and rungs and put the rungs in a hot sand bath as described below.) Since the posts on this chair are cylindrical, this is a good time to add any finials, tapers, roundovers, etc. to the top of each post. They will help to show the orientation of each post. 2. Holding all four posts in their relative positions, mark the bottom of each with its position, e.g., LF (left front), RF, LB, RB. Make a pencil mark showing the direction of the front and back rungs, i.e., connecting LF to RF and LB to RB. (Remember, the legs are upside down at this point so either F/B or L/R is reversed.) 3. Put each front post in the drilling jig. 4. Use a story pole to mark the front rung positions (consistent with the mark on the bottom of the rung.) 5. Drill the holes for the front rungs in each front post. The depth should be about half the diameter of the post. 6. Repeat 3-5 for back posts 7. For each post, put an indexing dowel in one of the holes, rotate the post 90E, so that the side where the side rungs should go is up. Lock the clamp. 8. Use a story pole to mark the vertical positions of the side rung holes. 9. Put the “master angle” wedges under the drilling jig. Re-center the drill bit & re-adjust the depth stop on the drill press. The wedges change the relative position of the post and the drill bit, so these have to be checked before drilling the second set of holes 10. Drill the holes for the side rungs. Pay close attention to which side and which angle when you set up the chair posts in the drilling jig and the “master angle” wedges. It is easy to get confused and drill in the wrong position or at the wrong angle. 47-3 Figure 2 may help to keep track of this. If the index pin (side rung hole) is pointing to the thick side of the wedge, the angle of the hole is less than 90Eand you Figure 2. Dowel Drilling Jig and “Master Angle” Wedge. (There is a groove cut down the center. The strips near the clamp are 5/16" thick, so that when the index pin is resting on them the post has rotated exactly 90E.) should be drilling front posts. If the index pin is pointing to the narrow side of the wedge, the angle is more than 90E and you should be drilling rear posts. Also – you do not have to flip the jig and reset the depth and position of the drill press for each post. If the LF post is Figure 3. Dowel Mortising Jig 315 Table of Contents 47-4 Ladderback Chairs drilled, say, with the bottom of the post to your left, then you put the RF post in the jig with the bottom of the post to your right. Alternatively, you can drill for the front rungs, assemble the front ladder, and then drill for the side rungs. Figure 4. Quick-release Dowel Vise PREPARING THE RUNGS Miller recommends heating the end of the rungs in hot (160E - 180E) sand, to dry them and to shrink them. Then, as the rails reabsorb moisture over the years, the joints will tighten. A 750 watt hotplate set on lowmedium has more than enough heat to do this. Use a thermometer. Make sure that you do not overheat (char) the dowels. As it happened, the e" dowels I bought for rung stock were very slightly undersize, even before being dried in the sand bath. The holes drilled with a 19/32" brad point drill were actually a bit too small, nonetheless. I had to reduce the diameter of the rungs slightly to get a good fit. Also, a hole gauge showed that they were not quite round. I made a dowel clamp as shown in Figure 4. I clamped it in my face vise and then clamped dowels in it. Although the range of dowel diameters it will accept is limited, it was much more convenient for holding rungs than my face vise. I made a hole gauge by drilling three 19/32" holes in a block of oak. I blackened the inside with a soft pencil. Then I shaved and sanded each end of each rung very slightly and tried it in the hole gauge. The graphite showed me where the pressure points were. I removed a bit from the blackened parts of the dowel end and tried the fit again. (See p. 237 of Lonnie Bird’s book.) This resulted in well-fitting rungs. I found, however, that the rails can fit too well, and I could not drive some of them all the way into the mortise. According to Kerry Pierce, a properly fitting rung should slide into the mortise with just a bit of resistance when dry-fitting, but, once in, should not wobble from side to side. By the way, the graphite from the hole gauge will make the tenon slide more easily into the hole, but the glue will wash it off, and the tenon may not slide in as easily at glue-up. Take your choice, but it is easier for me to fuss with getting a good fit than to have a difficult glueup. BACK SLATS I made flat slats that were flat on the bottom but arched across the top. The first one I cut became the template for the other two. These were mortised into the chair posts. I did not like the mortise block Miller recommended. Unfortunately, my first attempts at a mortise jig allowed the post to rotate while I was mortising for the back slats. I finally made two clamps with arched cauls that worked well. (See Figure 3.) The bad mortise cuts would have 316 Table of Contents Notes and Reflections While Shaving Wood ruined the posts, except that I decided to fill the mistakes with wood putty, and then paint the slats and the damaged areas of the posts. As they say, every mistake is a design opportunity. FINISH The clear finish comprised a washcoat, followed by NGR dye, tinted shellac, and a top coat of wipe-on poly. I regretted using tinted finish because it is too difficult to rub in an even color on a chair. I should have left the wood unstained or used a pigment stain over the wash-coat. I chose to paint the slats because I needed to hide my errors in cutting mortises. Wipe-on finish seems definitely the way to finish a chair like this. WEAVING WITH SHAKER TAPE The seat of the child’s chair is upholstered with Shaker tape. This is quite acceptable. Although it is not a rustic as rush, wood strips, etc, it is much more practical and it is attractive. Shaker tape is densely woven cotton. It is available in a variety of colors and striped patterns, in 1" and e" widths. (I have used only the 1" width. I understand that synthetic fabric is also available, but I have not noticed a supplier for it. The cotton is fine with me.) At the time I made the chair, VI Reed & Cane had the best prices while Shaker Workshops had a wider selection1. 47-5 The reference by Carlson is a useful guide for learning how to weave a seat or a chair back with Shaker tape. Miller’s chapter on a child’s chair also has decent instructions. A Shaker tape seat is actually a double seat. The tape is wrapped around the seat rungs and woven on both the top and bottom. Some sort of cushion or filler is usually placed between the strips. I use 1" polyurethane foam. On the rocking chair (below) I used 2" NuFoam but I think that 1" polyurethane is preferable. Others use a bag of cedar chips or just layers of cloth. Vendor websites give formulas for calculating how much tape is needed. Wrap a tape measure around the seat rungs fore and aft. (This is the length of tape in inches needed for one warp row). Multiply this by the width of the seat in inches and divide by 36 to get the linear yards of 1" tape needed for the warp. (For e” tape, divide by 22.2) You will need the same amount for the weft. If you are using just one color, just double the length needed for the warp. (Order about 10% extra for a margin of error.) The process of weaving a seat is straightforward, but there are a few details. The first detail is that the front of the chair is wider than the rear. The experts recommend that the warp be rectangular, i.e., that each row of tape be exactly square to the front and back rungs. Then the triangular spaces at both sides of the front is filled by nailing one or more extra warp strips to each side rung, wrapping it around the front rung, and returning it to beside its origin. 1 (http://www.shakerworkshops.com/catalog/shaker_ch air_tape.php/tape) 317 Table of Contents 47-6 Ladderback Chairs I overlap each warp strip at the rear a little bit, so that the strips fan out slightly toward the front. That fills the extra space at the front without the extra tacking and fiddling. That’s how I did the warp on the rocker (below). I find a flat oak “needle” (about 1" wide by c" thick by 6" long) with a slot cut in one end for an “eye” is very useful. The distance from the end to the slot should be a bit longer than an inch, not only for strength but to make it easier to push the needle under the warp. A hemostat is very handy for fishing the tape through at the end. A Simple Chair with Curved Slats Next, I wanted to make a slightly larger version of the same basic chair with curved back slats. This chair was based on the Miller Child’s Chair, but is a bit larger – approximately the dimensions of a #1 Mt. Lebanon chair. (See Figures 9 and 13 and Handberg, p.56 of Vol. 1). I used a e" roundover router bit to create a dome on the tops of the posts. The best way to do this cleanly is to cut the full roundover in one sector of the post, withdraw it, rotate it slightly, and repeat. Rotating the post while pushing it slowly into the spinning cutter did not produce clean roundovers. Figure 6. Geometry of Curved Chair Slat BENDING BACK SLATS Curved back slats add an interesting complication in building a ladder-back chair. You have to decide how much to bend the slat. The slat piece, as you cut it out (before it is bent) has to be longer than the distance between the rear posts. The angle of the mortises is no longer zero relative to the back rails. And you have to bend the slats. Figure 5. Simple Child’s Chair 318 Table of Contents Notes and Reflections While Shaving Wood 1. Decide on the radius of the slat curve. The idea is to approximate the shape of the sitter’s back. A chair comfortable for a child or smallstatured adult would have a narrower back, e.g., 11" and the slats would have a smaller radius, e.g., 15". A chair comfortable for a larger person would be wider, say 18", and would have a radius of 22" or so. Laying out the bent slat requires us to know the height of the arc (also called the sagitta), the length of the chord, and the length of the arc. The height of the arc (saggita) is the distance between the arc and the chord. It is 1¾" in Figure 6. The chord is the straight line drawn between the rear posts. The length of arc is, of course, the length of the slat. Table 1 and the text in the Appendix show examples of the relationship between radius, chord and the height of the arc. 2. Mark the distance between the rear posts on a piece of plywood or Masonite. Mark the height of the arc. Inscribe the arc and cut out 47-7 2.) or measure it from a tape wrapped around the arc. 4. Place the bending template on the jig with a ¼" spacer underneath it (representing the slat). Position the rails so that they touch the template and the spacer. Screw them to the base. Make sure that they are parallel to each other and to the pressure arm. (See my separate description of the bending jig.) 5. Cut out the blanks (arc length plus twice the tenon length). It is better, if you may make the chair again, to prepare a cutting template for the slats. Cut out the slats on a bandsaw. Fair the edges and relieve the arrises. 6. Steam the blanks and bend them, secundum artem.** I steamed ¼" thick slats on top of a stove, in a baking pan tightly covered with aluminum foil, for 20 minutes. (I punched a Figure 8. Slat Bending Jig in Use Figure 7. Slat Bending Jig Setup a bending template with that arc. (This is not the same as the cutting template for the slat.) 3. Calculate the length of the arc (See Table few holes in the foil to let the steam escape ** Secundem artem is a pharmacist’s term meaning, “according to the art”. This is an ironic hint. Steam bending is a whole ‘nother thing. At least, slats are easier to bend than posts. See the next chapter. 319 Table of Contents 47-8 Ladderback Chairs evenly.) The slats bent readily, but were slightly discolored by a trace of iron in the water. I left the slats in the bending jig and let them dry in the sunlight for two days. (Figures 7-8.) MORTISING FOR THE SLATS The bent back slats require the mortises to be cut at the correct horizontal angle. Experienced chair makers (e.g., Conover, Pierce) say that they eyeball the angle and cut the mortises freehand, with drill and chisel. I prefer to use my router. That requires me to calculate the angle (Table 3) or read it off a drawing or the bending template. It is the angle between the chord and a tangent to the arc where it meets the chord. (See Fig. 6). For this chair, the angle I read from the bending jig is 20E. (Figure 6 says 30E.) Mark the angle on a perfectly rectangular piece of plywood, perhaps with a notch cut in it to clear the post. Put an indexing jig in one of the rear rung holes of a rear post. Rotate the rung forward (toward where the front of the chair will be) until the index pin lines up with the mark. The front of the mortise should be at the midline of the post. Lock the clamps and cut the mortise. The second post is the mirror image of the first, not a copy. It is all too easy to make a mistake by cutting the mortises on the front of the post. (You needn’t ask why I know this.) Suppose that you cut both posts with their bottoms to your left. The index pin should be rotated away from you for one post and toward you for the other. FINISH I finished this chair with two coats of shellac (the second coat was de-waxed shellac) and one coat of wipe-on polyurethane. Rocking Chair This chair is close to the chair described by Handberg (p. 54 of vol. 1), but it has an additional curved back slat like a similar chair described by Conover. Its rear posts are bent at the bottom. The bent rear posts, bent back rungs, and woven back represent increased complexity from the larger child’s chair described above. (See Figure 11.) According to the Mt. Lebanon catalog, this chair is sized for a adult woman. By modern standards, it is sized for a short woman. It is wide enough, but the seat would be an inch or so too low for most adult women. The arm also should be an inch higher, about 5-5 ½" above the seat. TURNING AND BENDING THE POSTS Figure 9. Larger Child’s Chair with Curved Back Slats The first step, after making a detailed drawing, was to turn the posts. I used commercial oak dowels in ¾" and 1¼" 320 Table of Contents Notes and Reflections While Shaving Wood diameter, but I wanted a flame finial on the rear posts and a bit of profile on the front posts. I also turned the ¾" tenons at the tops of the front posts. Second, I steamed and bent the rear posts, and then the back rungs and back slat. The results of the bending operations were barely acceptable, perhaps because I used commercial dowels that had been kiln dried, or perhaps I simply did not steam them long enough. I was barely able to bend the 1¼" posts. I had some splintering (tension failure on the outside of the bend). This occurred on the posts and even on the ¾" rungs. I left the bent wood in the forms for over a week. There was significant springback in the 1¼" dowels when I took them out of the bending form. I was able to repair the splintering but this would be a real concern for the future. I conclude that commercial furniture grade dowels are acceptable for straight work but not for bent work. In the future I will look for air dried, preferably riven, stock for posts that I plan to bend. MAKING THE FRONT LADDER I prepared the front rungs and front posts. I used a e" plug cutter to reduce the diameter of the ¾" dowels. The fit into a e" hole was too loose, however, so I drilled 19/32" mortises, and then sanded the dowels to a precise fit. I marked positions for the front rungs using a story stick, drilled the mortises in the front posts for front rungs, marked the positions for the side rungs, again with a story stick, and then drilled the mortises for the side rungs. I glued up the front ladder. MAKING THE REAR LADDER A minor mystery while I was drawing the 47-9 design had been how to measure for the mortises and slats in the bent rear posts. The angle of the back would affect the position, length and vertical angle of the side-rung mortises. What is the “angle” of a bent post, you may ask? I clamped a long straight board to the rear post so that it touched the post at the very bottom and at the very top. I considered a 95E angle, as a bit more comfortable, but I think I eventually chose a 90E angle. I doubt it matters much, since this was to be a rocking chair. Without rockers, however, the angle would have to be a compromise between comfort and stability. The chair might be unstable if the top were too far behind the bottom. I would worry that the chair could tip over backwards too easily. After the angle of the rear post to the “floor” (ground plane) had been set, I measured from the floor directly up the same distance as the mortises on the front posts and marked the rear posts. The next mystery was how to drill mortises at the correct vertical angle to the bent posts. Because of springback, the bends might not be identical, but I needed the side rungs to be horizontal. Conover drills these by eye, but this was not something I wanted to try, since spoiling the post would sent me weeping back to the lathe and then to the steambox. I clamped the two rear posts between two boards, and clamped two more boards at the ends. In effect, the two rear posts were clamped in a box. (See Figure 10.) I made sure that the bottom edges of the long sides of this jig were parallel so that the jig would lie flat on the workbench with no rocking. Then, when I took the jig to the drill press, the mortises for the side rungs would be drilled at the correct vertical angle, regardless of the 321 Table of Contents 47-10 Ladderback Chairs remembered that the rear posts are mirror images, and got the mortises cut on the correct side of the post, at the right angle. When I prepared the bent rungs and the backslat, I took pains with the fit of the tenons. I anticipated that I might not be able to force the bent pieces into their mortises if the tenons fit too tightly: they might bend more or even break rather than slide into the mortises. Even so, I had to reinforce the bent rails by clamping the radius between two straight pieces, to prevent them from bending when I applied clamp pressure. This might be a good application for an adhesive that fills gaps and is strong even in a loose-fitting joint (e.g., epoxy). I dry-fit the rear ladder carefully. I glued the slat and rungs to one post and then and then glued on the second post. SIDE RUNGS, ROCKERS AND ARMS Figure 10. Drilling Jig for Rear Posts bend in the rear posts. I chose to drill the mortises for the side rungs in the same plane as the bends. That is, the bend, the side rungs and the rockers were parallel, and 98E to the rear rungs. This should have caused the rear posts to “toe in” a little, but it was not really noticeable. (To make the plane of the bent rungs square to rear rungs, I could have laid the rear posts on their sides and drilled the mortises for the rear rungs.) I drilled the mortises for the rear rungs at a 98E angle to the side rungs, using the index pin and “master angle” wedges as described above. (See Figures 1 and 2). Then I cut the mortises for the two bent back rungs and the back slat, as described above. I After the glue in the rear ladder had set well, I glued in the side rungs. Then, I measured the distance for the arms, cut out the arms from ¾" stock, and rounded over the arrises with a spokeshave and sandpaper. (Conover recommended ½" thick stock for his maple chair, but I thought that I’d need thicker stock for my chair.) Unfortunately, I cut the arms to the distance between the rear and front posts, plus 2" for overhang on the front. After I drilled the ¾" hole to fit over the tenon on the front post, I realized that I had not allowed extra length to turn a tenon at the back of the arm, which is supposed to attach the arm to the rear post. I could have made the arms again, but instead I drilled d" mortises through the back of the post and into the arm, to accept d" “loose tenons” (i.e., dowels). This was OK, but I would not wish to do it again. Also, I tapered the arm to about ¾" square at the rear. 322 Table of Contents Notes and Reflections While Shaving Wood I think that the rear of the arm should be wider than ¾", perhaps as wide as the diameter of the rear post. The design of this joint will need more thought. The rockers are joined to the posts with bridle joints, which I cut by hand. (This kind of work really makes me appreciate my little saddle square and my Japanese pull saw.) I copied the pattern for the rockers from Conover. He specifies ½" thick rockers. I chose to make mine e" thick. I cut the pattern out of 8/4 oak (because I had a nice, straightgrained piece available) on the bandsaw, faired them nicely with spokeshaves and sandpaper, and the resawed off two e" thick rockers. The posts were 1¼" diameter. If I cut 5/8" wide mortises into the ends, this would leave only 5/16" on each side of the rockers. I cut ½" wide mortises and cut away c" inlets on the rockers. When I make this design again, I will use 1½" - 19/16" diameter or thicker posts and will not need to inlet the rockers. I was very careful not to split the bottom of the post while fitting the bridle joints. At this late stage it would be a disaster. I clamped the posts just below the joint while I was fitting the joints and when I glued them up, to be sure that they would not split apart. I did a decent enough job cutting the mortises for the bridle joints, and I smoothed off the insides with a rasp and sandpaper until the fit was snug but not too tight. They did not fit quite as well as I would normally want an M&T joints to fit, however. I glued them together with slow-setting epoxy. I added ¼" pins after the epoxy had set. The rockers as designed by Conover seemed OK, except that they are not shaped properly where they should fit flat against the post bottoms. Another necessary design 47-11 Figure 11. Rocking Chair modification, perhaps, or the rocker profile could be adjusted as part of the fitting operation. I finished the chair with one coat of shellac (wiped on) one coat of dewaxed shellac (sprayed on from an aerosol can) and one coat of wipe-on polyurethane. References Lonnie Bird, J Jewitt, T Lie-Nielsen, et al. Complete Illustrated Guide to Woodworking. Newtown, CT. Taunton Press 2005, p. 237 Glenn A Carlson. Weaving Shaker Tape Seats. Fine Woodworking Nov/Dec 1966 323 Table of Contents 47-12 Ladderback Chairs Ernie Conover. Shaker Rocker: Jigs and story sticks ensure accurate joinery. Fine Woodworking May/June 1995 The Countryman Press, 1991. p. 56 Ejner Handberg. Shop Drawings of Shaker Furniture and Woodenware. Woodstock, VT. Kerry Pierce. Chairmaking Simplified, 2008: Cincinnati, Ohio, Popular Woodworking Bo ok s Jeff Miller. Chairmaking and Design. Newtown, CT. Taunton Press 2007 Figure 12. Child’s Chair Dimensions. Round or chamfer top ends of posts. 324 Table of Contents Notes and Reflections While Shaving Wood 47-13 Figure 13. Chair With Bent Slats (Mt. Lebanon #1 Chair.) 325 Table of Contents 47-14 Ladderback Chairs Figure 14. Rocking Chair (Mt. Lebanon #3 ). Rocker 326 Table of Contents Notes and Reflections While Shaving Wood 47-15 Figure 16. Seat Diagram for Rocker Appendix: Math for Chairmaking When I was making my first chair, I wanted to use a beam compass to lay out a back slat with an arched top. The arc of the top of the slat should hit the height requirement and the two ends of the slat where they touch the posts (not counting the tenons). I found that trying to guess the radius I needed by trial and error was a real pain. It is possible to calculate the radius of a circular segment from the chord length and the height of the segment (technically called the sagitta) by mean of the formula r = (c2/8s) + s/2, where "c" =chord length and "s" = sagitta (height of the circular section) I used a spreadsheet to calculate some values that I expected to need in making chairs. See Table 1. (These also will give a rough idea of the radius involved in bending legs.) For other calculations, most books and articles say to make a full-sized drawing of the chair. That’s probably a good idea, but just for fun I calculated some theoretical dimensions for curved back slats. For a curved back slat, we need to know the length to cut the slat before bending. The formula for calculating the length l of a circular arc is simply the radius of the arc times the value of the inscribed angle, l = r θ. The angle can be found by trigonometry to equal 2 * arcsin (c/2r) or 327 Table of Contents 47-16 Ladderback Chairs 2 * arc cosine((r-s)/r) and therefore the length l of the arc is given by l = 2*r*arc cosine((r-s)/r). I used a spreadsheet to calculate some common values of arc length for chair slats. Note that the numbers in Table 2 do not include the lengths of the tenons. Also, for curved back slats, we need to know the angle of the slat to the post, so that we can cut the mortises at approximately the correct angle. This angle is the angle of a tangent to the circle, at the chord. You can get this from a drawing, especially a good drawing program like Sketchup that will Figure 17. Sketchup Drawing of Curved Chair “read” the angle for you. (See Figure 14.) Slat. Chord=10.75", s=1", r=15", angle = 20E By trigonometry, the acute angle between the chord and the tangent line, i.e., between a line drawn between the rear posts and the slat as it enters the post. this is given by Angle (degrees) = (π/2 - arcsin ((r-s)/r) * 180/πi Some calculated values are in Table 3. Note that the chord length is the distance between the rear posts, not the chord of the curved back slat, including the tenons. 328 Table of Contents Notes and Reflections While Shaving Wood 47-17 Table 1. Radius Corresponding to Chord Length and Height of a Circular Section (inches) CHORD (in) 1 13       21 10/16 13  1/2  23  5/16 14       25       14  1/2  26 13/16 15       28 10/16 15  1/2  30  9/16 16       32  8/16 16  1/2  34  9/16 17       36 10/16 17  1/2  38 13/16 18       41       18  1/2  43  5/16 19       45 10/16 19  1/2  48  1/16 50 8/16 20       SAGITTA (Height of Circular Section) (in.) 1.5 2 2.5 3 3.5 14 13/16 11  9/16 9 11/16 8  9/16 7 13/16 15 15/16 12  6/16 10  6/16 9  2/16 8  4/16 17  1/16 13  4/16 11  1/16 9 11/16 8 12/16 18  4/16 14  2/16 11 12/16 10  4/16 9  4/16 19  8/16 15  1/16 12  8/16 10 14/16 9 13/16 20 12/16 16       13  4/16 11  8/16 10  5/16 22  1/16 17       14  1/16 12  3/16 10 14/16 23  7/16 18       14 14/16 12 14/16 11  8/16 24 13/16 19  1/16 15 11/16 13  9/16 12  1/16 26  4/16 20  2/16 16  9/16 14  4/16 12 11/16 27 12/16 21  4/16 17  7/16 15       13  5/16 29  4/16 22  6/16 18  6/16 15 12/16 14       30 13/16 23  9/16 19  5/16 16  9/16 14 10/16 32  7/16 24 12/16 20  4/16 17  6/16 15  5/16 34 1/16 26 21 4/16 18 3/16 16 1/16 4 7  5/16 7 11/16 8  2/16 8  9/16 9  1/16 9  8/16 10       10  8/16 11  1/16 11  9/16 12  2/16 12 11/16 13  5/16 13 14/16 14 8/16 radius of arc = (c2/8S) + S/2, where "c" =chord length and "s" = sagitta (height of the circular section) Example: A 2" high arc that intersects the ends of a 17" chord has a radius of 19 1/16" 329 Table of Contents 47-18 Ladderback Chairs Table 2. Length of Arc from Chord Length and Height (Sagitta) CHORD (in) 13       13  1/2  14       14  1/2  15       15  1/2  16       16  1/2  17       17  1/2  18       18  1/2  19       19  1/2  20       1 13  3/16 13 11/16 14  3/16 14 11/16 15  3/16 15 11/16 16  3/16 16 11/16 17  3/16 17 10/16 18  2/16 18 10/16 19  2/16 19 10/16 20  2/16 1.5 13  7/16 13 15/16 14  7/16 14 15/16 15  6/16 15 14/16 16  6/16 16 14/16 17  6/16 17 13/16 18  5/16 18 13/16 19  5/16 19 13/16 20  5/16 2 13 13/16 14  4/16 14 12/16 15  4/16 15 11/16 16  3/16 16 11/16 17  2/16 17 10/16 18  2/16 18  9/16 19  1/16 19  9/16 20  1/16 20  8/16 SAGITTA (in.) 2.5 14  4/16 14 11/16 15  3/16 15 10/16 16  1/16 16  9/16 17       17  8/16 17 15/16 18  7/16 18 15/16 19  6/16 19 14/16 20  6/16 20 13/16 3 14 12/16 15  3/16 15 10/16 16  2/16 16  9/16 17       17  7/16 17 15/16 18  6/16 18 13/16 19  5/16 19 12/16 20  4/16 20 11/16 21  3/16 3.5 15  6/16 15 13/16 16  4/16 16 11/16 17  1/16 17  8/16 18       18  7/16 18 14/16 19  5/16 19 12/16 20  3/16 20 11/16 21  2/16 21 10/16 4 16  1/16 16  7/16 16 14/16 17  5/16 17 11/16 18  2/16 18  9/16 19       19  7/16 19 14/16 20  5/16 20 12/16 21  3/16 21 10/16 22  1/16 length of arc = rθ = 2 r arc cosin((r-s)/r), where "r" = radius and "s" = sagitta Example: a chair slat with a 2" deep curve should be 17‐5/8" long, to fill a 17" distance between the rear posts (legs) plus 2 x tenon length 330 Table of Contents Notes and Reflections While Shaving Wood  Table 3.  Angle (degrees) of Tangent to Chord  SAGITTA (in.) CHORD (in) 1 1.5 2 2.5 13       17.5 26.0 34.2 42.1 13  1/2  16.9 25.1 33.0 40.6 14       16.3 24.2 31.9 39.3 14  1/2  15.7 23.4 30.8 38.1 15       15.2 22.6 29.9 36.9 15  1/2  14.7 21.9 28.9 35.8 16       14.3 21.2 28.1 34.7 16  1/2  13.8 20.6 27.3 33.7 17       13.4 20.0 26.5 32.8 17  1/2  13.0 19.5 25.8 31.9 18       12.7 18.9 25.1 31.0 18  1/2  12.3 18.4 24.4 30.2 19       12.0 17.9 23.8 29.5 19  1/2  11.7 17.5 23.2 28.8 20       11.4 17.1 22.6 28.1 47-19 3 3.5 49.6 47.9 46.4 45.0 43.6 42.3 41.1 40.0 38.9 37.8 36.9 35.9 35.1 34.2 33.4 4 56.6 54.8 53.1 51.5 50.0 48.6 47.3 46.0 44.8 43.6 42.5 41.5 40.4 39.5 38.6 63.2 61.3 59.5 57.8 56.1 54.6 53.1 51.7 50.4 49.1 47.9 46.8 45.7 44.6 43.6 Angle (degrees) = (pi/2 ‐ arcsin ((r‐s)/r) * 180/pi Example: 14" between posts, 1.5" sagitta, slats meet post at about 24 degrees 331 Table of Contents 332 Table of Contents Notes and Reflections While Shaving Wood Chapter 48 – Steambending Wood for Furniture B need a propane tank for any other purpose. ending wood opens up many additional design possibilities. The two main methods of making bent constructions in wood are steambending and laminating on a form. This chapter describes my limited experience with steambending. STEAM BOX My steam box is vertical so that the condensate can reflux. The two most common materials for a steam box are plywood and PVC pipe. I made a 6" x 6" x Generating Steam The prime requirement is, of course, a source of hot fog, the hotter and wetter the better.c When I started, I wanted a steam box approximately 6" x 6" x 50" long. I knew from experience that a 750 W hotplate would be completely inadequate to provide the necessary volume of steam, merely based on how long it takes one to heat a kettle. Other woodworkers report good success using an electrical wallpaper steamer rated at about 1500 watts. They should know, but I was skeptical, and 1500 W wallpaper steamers are expensive. Based on very approximate calculations (See Appendix), a steam generator for this box should be able to produce about 6000-11,000 BTU/hr and the water reservoir should hold about 2-3 gallons. I decided that a 1500 W electrical heater with a 3 gallon container would constitute the bare minimum requirement for a 6 x 6 x 50" steam box. I decided to buy a 54,000 BTU propane burner and a 10 qt. pot, because they cost about the same as a smaller propane burner and much less than a 1500 W wallpaper steamer. The disadvantage was that I did not Figure 1. Two Steam Boxes 50" box out of plywood (for steaming chair posts) and a 5½" x 30" one out of PVC (for steaming rungs and slats). They each fit snugly into a plywood kettle lid with a hole cut in the center, to support them over the kettle. (The hole is covered with hardware cloth so that nothing can fall into the boiler.) The plywood box is made with rabetted edges and angle brackets so it slides together and is easily disassembled for storage. The top is a loose fitting piece of plywood with a hole for the probe of a cooking thermometer. This arrangement produces billowing steam out of the top, with little or no visible leakage at the bottom. The temperature inside the 5½" pipe was about 200-210E F but the temperature inside the 6 x 6 box never got above 180-190E despite plenty of steam billowing out of the top. (In fact, the burner c Hot fog, because we don’t really maintain temperatures above 212E F. in the steam box Actually, temperatures above 212E may be counter-productive. 48-1 Table of Contents 333 48-2 Steambending was turned up too high and the water was nearly depleted at the end of 75 minutes.) Although hardware store PVC is not rated for high temperatures, and will soften when hot, the 30" piece did not distort noticeably in a 50 minute session. Next time, I shall replace the 6" x 6" plywood box with a 48" length of 5½" PVC and reinforce it with a stick or two if it starts to sag. I hung the wood pieces inside the box with string attached to some screw eyes around the opening at the top. Choice of Woods Probably wood of any species or state of previous drying will bend somewhat. Furthermore, success would depend a lot on the radius of the bend and whether a compression strap is used. Some authors highly recommend one, others do not. I have had difficulty in bending commercial furniture grade oak dowels, in both 1¼" diameter (very hard to bend, some internal separation, substantial springback) and ¾" (tendency to break out on the outside). The posts and the dowels were bent to a large radius, e.g., 15 - 20", which is not at all a tight bend. I suspect that these dowels might have been kiln dried. Also, I don’t think I steamed them long enough and I was too hurried to orient the dowels optimally. I should choose the orientation and mark the piece before putting it in the steambox. ¼" oak slats bend nicely. According to the Veritas Steam Bending Instruction Booklet, only air-dried wood of an appropriate species should be used. Kiln-dried wood must not be used; the lignin in the wood has been permanently set during the hot, dry kilning process. No amount of steaming or soaking will weaken the lignin bond sufficiently for successful bending. The same applies to air-dried wood that has been allowed to dry and stabilize below 10% moisture content; the lignin will only partially plasticize with steam, not enough for successful bending of anything beyond a shallow curve. Schleining disagrees with this advice: Many experts claim that you must use green or air-dried lumber for steambending, but . . . I've had no choice but to bend kiln dried material, and the results have been quite successful. Kiln-dried wood is not only easily available, even if you live in the city, but it's also dimensionally stable, dry and ready to be milled as soon as bending is complete and the piece has had time to cool. According to the US Forest Products Laboratory Wood Handbook Heat and moisture make certain species of wood sufficiently plastic for bending operations. Steaming at atmospheric . . . pressure, soaking in boiling or nearly boiling water, . . . are satisfactory methods of plasticizing wood. Wood at 20% to 25% moisture content needs to be heated without losing moisture; at a lower moisture content, heat and moisture must be added. As a consequence, the recommended plasticizing processes are steaming or boiling for about 15 min/cm (38 min/in) of thickness for wood at 20% to 25% moisture content and steaming or boiling for about 30 min/cm (75 min/in) of thickness for wood at lower moisture content levels. Wood for steam-bending should have straight 334 Table of Contents Notes and Reflections While Shaving Wood grain, which enables you to bend without stressing the areas already weakened by grain direction. Green, riven (split) billets would be ideal, if you can find or make them. Longgrained woods such as hickory, ash and oak bend easiest. Dense woods, such as cherry, hard maple and most exotics, are more difficult to bend. (See following table) 48-3 Chair-Slat Bending Jig This jig was developed from a photo in an article on ladder-back chair making by Ernie Conover. I wanted a jig that would bend ¼" slats of any length up to about 18". I made it large enough to accept 4, 3" slats. The jig bends slats that are suspended Bendability of Domestic Hardwoods Limiting Radii of Curvature (in inches for 1-in. stock) % Unbroken Pieces Supported By Strap Ash . . . . . . 67 Beech . . . . 75 Birch. . . . . 72 Elm, soft . 74 Hackberry . . . . 94 Hickory . . . 76 Magnolia . . . . . 85 Maple, hard . . . 57 Oak, red . . . 86 Oak, white . . . . 91 Pecan . . . . 78 Sweetgum . . . . . 67 Sycamore . . . . . 29 Tupelo. . . . .42 Walnut, blk 78 Afrormosia. 14.0 Alder . . . . .14.0 Ash . . . . . .4.5 Beech. . . . ..1.5 Birch. yellow.3.0 Douglas fir. 18.0 Ebony . . .10.0 Elm, white . .1.7 Hemlock .19.0 Hickory . . . 1.8 Mahogany 36.0 Oak. white . .0.5 Oak. red. . 1.0 Spruce. Sitka36.0 Teak . . . .18.0 U.S. Forest Products Laboratory. Wood Handbook: Wood as an Engineering Material 1974. Unsupported 29.0 18.0 13.0 13.0 17.0 33.0 15.0 13.5 36.0 15.0 32.0 13.0 11.5 32.0 35.0 W.C. Stevens and N. Turner. Wood Bending Handbook (Princes Risborough. England: Forest Products Research Laboratory. 1970) Figure 2. Slat Bending Jig, End View of Two Settings between two rails of the correct height and distance apart. (Figure 2) The distance between the rails can be calculated, as shown in the figure. (See also Chapter 47, LadderBack Chairs). The distance between the rails also can be fitted to a template with the desired radius, as shown in the photograph (Figure 4). Note the ¼" spacer under the template in that figure. That template represents the inside curve of the slat, and the spacer represents the thickness of the slat. 335 Table of Contents 48-4 Steambending A slight (e.g. 70E, bevel on top of the rails may spread out the pressure on the slat somewhat. The hinge is two oak blocks that are screwed to the base, reinforced with an inverted “L” shaped holder, also screwed to the base. (This is shown in Figure 4 but not the drawings.) The hinge pin is a 5/16" machine screw. The pressure arm is 1½” x 1½” x 15" oak. (Apparently Conover’s jig used a pine 2 x 4.) The pressure arm is tapered at the bottom of the hinge end (the bottom corner is cut off) so that it can rotate. Figure 3. Slat-Bending Jig significantly widen and shape the pressure arm, say 3-4" wide. If that did not work, I should consider using a bending jig with two mated surfaces (See Pierce, p. 56, Photo 3). I could use the jig shown in Figure 3 as a drying form. Post-Bending Form Bending chair posts requires great force and speed, so everything must be set up in advance. The posts will be hot when they come out of the box, so eye protection and gloves are essential. Both Pierce and Conover show Figure 4. Photograph of Jig, Showing Setup Template The base has 1 x 1" stiffeners all around. After using this jig, I found that the 1½” wide pressure arm is too narrow. It has a tendency to fold some slats rather than create an even (fair) arc, at least for a 15" radius. You might need a pad of wood to widen the contact surface between the pressure arm and the slat. One advantage of this jig is that it is open, and allows the slats to dry while still in pressure. If I were making this again, I should Figure 5. Rough Sketch of Chair Post Bending Form 336 Table of Contents Notes and Reflections While Shaving Wood essentially the same design for a bending form. (Figure 5). To make a form, shape a piece of wood, e.g., a 6' 2 x 6, with the curvature that you want in the post. If only a portion of the chair post will be bent, the form only needs to be long enough to secure the post to the form on both sides of the bend, but not necessarily the full length of the post. There will probably be some springback, so the form can slightly exaggerate the bend. I have not used a bending strap, but experts recommend their use, and they should be considered for sharper bends and thicker stock. As you set up the form and rehearse the motions, you may find that you do not have a clamp that can apply sufficient force through a sufficient distance. A “C” clamp may have the force needed but not the length. A band clamp may offer the distance but the ratchet may be very hard to tighten. Once a part is removed from the steambox, it begins to cool immediately. Ideally, it should be clamped in the form and bent within a minute. A rehearsal or two will reduce both the time it takes you and the anxiety it causes. References Conover, Ernie. Shaker Rocker: Jigs and story sticks ensure accurate joinery. Fine Woodworking May/June 2005 Forest Products Laboratory. 1999. Wood handbook—Wood as an engineering material. Gen. Tech. Rep. FPL–GTR–113. Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Laboratory. 463 p. Keyser, William A. , Jr. Steam Bending: Heat and moisture plasticine (sic) wood Fine 48-5 Woodworking #8 pp. 40 October 1, 1977 Pierce, Kerry . Chairmaking Simplified, 2008: Cincinnati, Ohio, Popular Woodworking Books Schleining, Lon. Take the Mystery out of Steambending. Fine Woodworking May/June 2001 Veritas Steam Bending Instruction Booklet http://www.leevalley.com/en/html/05F1501ie.pdf Appendix My reasoning about heat and water requirements is as follows. (1) At atmospheric pressure, the volume of steam is about 1600 times the volume of an equal weight of water. So, even though the steam condenses into fog, assume that it issues from the boiler at about that volume, and that it heats that volume of water vapor-air mixture to almost 212E F. A steam box measuring 6" x 6" x 50" long has a volume of 1800 cu. in. Therefore, vaporizing 1.125 cu in (18ml) of water would fill the steambox with steam/fog. (2) If the box is reasonably well insulated (e.g., plastic pipe) a turnover of, say, 2-4 times/min. (3600 - 7200 cu in/min) should keep the temperature near that level. So, arbitrarily the objective should be to vaporize 36-72 ml. of water per minute. None of the articles I consulted recommended a flow rate. One author recommended that we maintain a temperature of 210Ein the box. (3) The heat of vaporization of water is about 2.3 BTU/ml). (1 BTU is approximately equal to 1 KJ.) So (assuming 100% efficiency) we need a burner that can generate at least 36 x 2.3 = 83 BTU/minute = 5000 BTU/hr. 1 BTU/min = 17 watts so the minimum electrical draw should be 1400 Watts. Build in a 25% fudge factor: 6000-11,000 BTU/hr, or about 1800 watts. (4) One source stressed that the flow must be continuous (at least, not interrupted by the need to refill the boiler). According to these calculations a two-hour steam session should require 4.3 - 8.6 liters (1-2.5 gallons) of water. The water requirement would depend a lot on whether the condensate could return to the boiler. Of course, some steam will leak out so these are rough minimums. 337 Table of Contents 48-6 338 Table of Contents Notes and Reflections While Shaving Wood Chapter 49 – Inlay I nlay is another of those woodworking specialties, like chair making and turning, that is nearly another craft with respect to tools and techniques. It requires some usual skills and tools of woodworking, such as routing and finish planing, and some skills of woodcarving. Inlay is mainly hand work, with or without the use of an electric router. The inlay method that works best for me requires a template, from which I cut the inlay and the recess. For larger inlays, I use a router set as described below. Transfer the drawing to the template material using carbon paper. I recommend ¼" Baltic birch plywood or Masonite for the template because it has to be strong enough to guide the knife or router without breaking. If you will use a router and guide bushing, the template should be about as thick as the protrusion of the bushing (¼"). If you will cut the inlay with a knife, the template may be somewhat thinner. Cut out the template using a coping saw or scroll saw and fair (smooth) the edges with a spokeshave, wood rasp and/or sandpaper. (A small sanding cylinder on a drill press works well.) If the inlay will have more than one part, cut out a template for the background profile and all inner shapes. Some people don’t make a template. They simply cut out the inlay and then cut the recess from that. I have broken a few inlays when I tried that. If the inlay breaks when the recess is partially cut, you have the problem of trying to make another inlay that exactly matches the part of the recess you have already cut. This is not insurmountable, but it may require either that you make a slightly larger inlay or that you move its position slightly to start over. If an inlay were complicated, e.g., alphabet characters, the “direct” method would probably be necessary for the sake of efficiency. Perhaps I will develop a lighter touch as I gain skill. Template Inlay begins with designing and drawing a template. Draw it freehand or with a computer Figure 1. Drawing, Templates and Practice Inlay Inlay Material graphics program. Print it full size in black and white on a good piece of paper. It might be worthwhile to cover the drawing with clear plastic packing tape to make it more durable. (See Figure 1) Mark Laub, in a video available for viewing on www.finewoodworking.com, recommended contrasting wood veneers, mother-of-pearl, paua abalone shell, ivory (old piano keys) 49-1 Table of Contents 339 49-2 patinated copper, and solder. Wood veneer can be fragile. Cross-banded veneer (made by gluing two pieces together, with grain running at right angles) is more durable, but more difficult to cut with a knife. Paper-backed veneer might also be useful. Cutting by Hand Cutting an inlay by hand is flexible and straightforward, but requires skill and patience. It certainly meets Pye’s idea that craftsmanship is workmanship of risk. If the knife slips while cutting the recess, the mark remains on the workpiece, right beside the attention-getting inlay. Begin by sticking the template to the workpiece with rubber cement or double-stick tape (turner’s tape), and a few well-placed brads or tacks. If you use brads, pre-drill a wooden template so that you won’t split it. The holes left by the brads will not show on the finished inlay because you will be excavating the wood under the template. If the inlay has more than one part, begin with the background template, e.g., the circle in Figure 1. Mark one spot on the template for identification and mark the corresponding spot on the workpiece. No hand-cut inlay is perfectly symmetrical, and you need to be able to put the template and the workpiece together in the same orientation every time. It is essential that the template not move, despite lateral pressure from the knife. If the template moves while you are cutting the margin for the recess, the inlay and recess probably won’t match. This is frustrating, to say the least, and may ruin the workpiece. Lighter-weight (photo mounting) doublesided tape can be unreliable, especially if attached to a dusty surface or the “waffle” Inlay side of Masonite. Turners’ tape or doublesided fabric backed carpet tape is usually strong enough if you make sure that the surfaces are not dusty. To be extra safe, place a Masonite template smooth side down, or “prime” the waffle side with rubber cement. I find that the best knife to use is an X-Acto knife with a #16 blade. This Figure 2. No. shape is less likely to cut into 16 X-Acto the template as you trace Knife Blade around it. A hooked blade like a # 28 works well, too. Some people take a small scratch awl and sharpen the point from one side only, in effect making a tiny marking knife. The knife has to be razor sharp, not only to cut easily but also to cut the veneer without splitting it. 1. Trace around the template with a fine point pencil and then with the knife blade. Use light pressure on the first cut: aim for a shallow cut or deep Figure 3. No. scratch. Press the knife blade 28 X-Acto firmly enough to scratch the Hooked Knife wood, but not hard enough Blade that the knife is difficult to control precisely. A. Cut mainly with the point of the blade. B. Be careful not to let the grain of the wood guide the knife blade away from the edge of the template. Keep the blade rotated slightly toward the template. (This is why the #16 blade is preferable.) Trace it again with gradually more pressure until you feel that you have scribed at least as deep as the inlay will require. Some people put a reference mark on the blade so they 340 Table of Contents Notes and Reflections While Shaving Wood can see the depth of cut. Once the wood is scribed, the #28 blade may cut a bit faster. Its easy to forget which part you have cut, so be systematic. 2. Carefully pry the template off of the workpiece with a wide, thin putty knife or XActo chisel blade and set it aside. 3. Take a narrow (e.g., ¼") razor-sharp chisel or a #18 X-Acto chisel blade and backcut all around the line from about c” - 3/16" inside the line. Be careful not to under-cut the line. 4. Excavate all of the wood out of the recess to a depth just a hair less than the thickness of the inlay material. (You will want to be able to plane, scrape or sand the inlay flush with the surface of the workpiece.) The bottom of the recess should be smooth and flat. Use a small router plane (e.g., Veritas router plane) or a small powered router, e.g., a Dremel tool with a base. Dremel sells a router base, but its depth of cut setting can be a bit imprecise. A Veritas or Stewart-MacDonald router base is more expensive but gives a more predictable depth of cut. The hand plane is much easier to control. A powered router, even a small one, obscures the cut with sawdust. It is too easy to cut outside the line, which will ruin the workpiece, unless you can enlarge the inlay to hide the mistake. 5. Lay the template on the inlay material and cut out the inlay, as #1, above, except that you can’t use pins to hold the template in place. A. Mark the inlay to correspond to the mark on the template and the 49-3 workpiece. B. It is best to lay out the inlay so that the grain of the inlay is parallel to the grain of the workpiece. C. Use a cutting mat, smooth plywood, polycarbonate, melamine, etc. as a cutting board. D. Cut the inlay to the outside of the line, i.e., a hair larger than the recess. Some inlays can be cut with scissors. Look at the back of the inlay repeatedly to follow your progress and to find any places where the knife may have skipped a bit. An inlay thicker than 1/16" should perhaps be cut with a scroll saw or by the router method. Cutting With Router Inlay Kit Cutting inlays with a router is more limited than cutting by hand, but offers quicker work and –theoretically at least – machine-like precision for some jobs. The router method is probably better for larger inlays, multiple inlays of the same size, and those that are thicker than 1/16", especially butterfly inlays and Dutchmen. (See Figure 4.) Its particular disadvantages are that the guide bushing does not cut sharp corners very well, although you can sometimes touch these up by hand. Also, a router bit may fracture thin, fragile inlay material. This method uses a special matched inlay kit (There are many brands. Freud and Whiteside are brands that come to mind). The kit comprises a c" carbide bit, a guide bushing and a collar that fits over the guide bushing. The guide bushing will fit into any router base that can accept Porter Cable guide bushings. It is ¼" long. The collar, which is used to cut the recess, has a radius of 9/32". The template for routing with an inlay 341 Table of Contents 49-4 kit is a “negative” template, i.e., an enclosed space to guide the bushing. The template material should be ¼" thick, perhaps a bit more, or the bottom of the guide bushing will hit the workpiece. (You could file the guide bushing and collar to make them shorter. I prefer not to do that.) Also, the border around the negative template should be wide enough to support the router base, so that the router won’t tip while you are cutting. You can’t see the template as you cut using the guide bushing. You have to be able to feel your way along. This is difficult, but somewhat easier with a lighter-duty router like a Bosch Colt or other trim router that will accept guide bushings. You need a plunge router to properly position the bushing before beginning to cut the recess or inlay. This also involves lateral pressure on the template, which is why the template has to be securely attached to the inlay material (but easily removable). The router method may affect the design of the inlay. The guide bushing with collar has a radius of 9/32", so it cannot follow an inside curve smaller than that. Imagine the cutter as running along a 9/32" offset inside the template. For example, if a template includes an inside curve with a radius of 9/32", the bit will just cut a straight line c" wide, which is useless. As a rule of thumb, the template should not have an inside corner, radius, etc less than ½". Furthermore, the clearance required by the collar can change the proportion of a Figure 4. Offset Effect of Collar Inlay Figure 5. Table With Large Inlays (Router Method) design. See Figure 4 for a simple example. The dark outside of the figure represents the negative template for a triangle. The triangle is about 1:6 The blue portion represents the offset caused by the bushing and the collar. The yellow area inside represents the recess that would actually be cut in the workpiece. The recess is much shorter, and the proportion is slightly different, than the design. Also note that the recess ends in an arc with the same radius as the cutter. The inlay itself is cut without the collar in place, so it would be just a bit longer than the recess, and would come to the sharp point intended. Either the recess would have to be extended by hand or the inlay would have to be rounded off at the point. Either way, the proportion is not exactly what was what was intended when the design was made. This problem becomes less salient the larger the inlay is. 1. It may often be easier to draw a pattern that is ½" oversize in every direction and then cut out the negative template on a jig saw. Remember, when you are cutting, that the waste is on the inside of the negative 342 Table of Contents Notes and Reflections While Shaving Wood template. Theoretically, you could begin by cutting out a “positive” template that is the exact size and shape that you want, as described above, and then cut out the negative template using the guide bushing without the collar. I find that this is, in practice, more difficult that it would seem because the heavy negative template (e.g., ¼" Masonite) is difficult to cut by feel and you can’t see the positive template while you are cutting. It might still be practical to cut the negative template from the positive template if you are willing to cut it in many passes, say 3 passes for a ¼" Masonite template. Also, don’t include thin pieces, e.g., the stem of a leaf or flower in the positive template because it will be too weak. Plan to cut thin parts separately. 2. Position the negative template on the workpiece, as described in step 1 under Cutting By Hand. Adjust the depth of the router to a hair less than the thickness of the inlay materials. Put the collar on the guide bushing. Rout out the margin of the recess. Later, you can clean out the interior with a router plane, larger bit, etc. Carefully pry the template off of the workpiece with a wide putty knife and position it on the inlay material. 3. Remove the collar from the guide bushing. Cut the inlay. The path of the router may fill with sawdust and make it hard to see whether the inlay has been cut through completely. Trust the depth setting on the router bit. Repeat as necessary for inside inlay pieces. The inlay will be just a bit too large. Sand it cautiously until it fits. 4. As an alternative to step 5, cut a positive template as in step 3 out of plywood, 49-5 fit it to the recess, and then use it to cut the inlay by hand. This might be preferable for very thin or fragile inlay materials. Installing the Inlay Now, whichever way you have cut them, you should have an inlay and a recess cut in the workpiece. Carefully check the fit of the inlay, including the depth of the recess (feel for a ridge with your fingertip.) You may need to trim off any stringiness left on the inlay by the router bit. Because you will want to sand, scrape or plane the inlay flush with the surface of the workpiece, it is best for the grain of the inlay to run in the same direction as the grain of the workpiece. Line up the reference marks and begin to fit the inlay. Work your way around. A close-fitting inlay requires repeated, careful cutting and sanding. If the inlay is slightly big for the recess, carefully cut away the walls of the recess to fit the inlay or sand away the inlay – its an artistic decision. Make sure that the walls and floor of the recess are straight and clean of chips, etc. Almost any adhesive will work for installing inlay: white or yellow glue, hot-melt veneer adhesive, epoxy. Epoxy might be better if the fit is not perfect and you want to fill any tiny gaps, but it may be messy for this application. Apply adhesive to the recess and the inlay. Install it and clamp in place under a block of wood, using a piece of waxed paper to keep the block from sticking to the inlay (or press the inlay down with a hot iron). If you are making a two-part (multipart) inlay, after the adhesive has set, cut the recess for the second inlay, and so forth. There should be no gaps between the 343 Table of Contents 49-6 walls of the recess and the inlay. If there are, they can be filled with a wood putty made of wood flour and white glue or shellac. Or, you can rub wood flour into the gap and then slowly let some CA glue seep into the flourfilled gap. Do not use a lot of glue. If you use too much it will float away the dust. Dab it with a toothpick and let it flow into the gap. You will be able to see the dust get darker, as the glue works its way through When it has set, flatten the inlay flush with the workpiece. Straight Inlay Straight inlay, e.g., a decorative “frame” design around the border of a panel, can be done as above. In this case, use an edge guide or a straightedge instead of a template to cut the recess. I tried using a router with an edge guide to cut the recess, but the cut was not clean enough – there was some tearout along the edges of the recess. Also, it was difficult to cut the inlay exactly the right width. Scribe the edges of the recess (groove) with a mortise gauge. (If the mortise gauge has sharpened needles instead of knife blades, score the work very lightly at first to avoid tearing the wood when you cut across the grain.) Clean out the recess. The router plane is much more precise and controllable than a powered router, and you can see where you are cutting. The depth of the recess should be a hair less than the thickness of the veneer so that you can sand or scrape the inlay flush with the surface, clean off squeezeout, etc. This can be difficult if you are using commercial veneer (1/32" or less thick) because there is so little tolerance. Thicket veneer (3/64" or so) is considerably easier to Inlay use for straight inlay because a deeper recess allows a somewhat larger tolerance. Square up the ends of the grooves and dados with a chisel. The inlay (if it is thin enough) can be cut with a mortise gauge to exactly the right width to fit the recess. Stick the veneer to a piece of melamine or other smooth surface with double-sided cellophane tape (better for this than double-sided carpet tape). If the distance from the edge guide to the first cutter equals the distance between the cutters, you can get two strips of veneer from each cut with the mortise gauge. Alternatively, prepare a piece of wood that is exactly as thick as the inlay should be wide, e.g., d" thick. For example, slice 4/4 piece in half and lightly plane it smooth on a jointer to a thickness of d". Then cut off thin (3/64 - 1/16") strips using a thin ripping gauge on a table saw (See Chapter 25, Jigs). Fit one strip of veneer to one of the grooves. Tape it in place with blue painters tape. Trim the ends square. Fit an adjoining strip so that it overlaps the first strip at the end. Tape it in place. Double-cut a 45E miter with a razor-sharp chisel. Remove the two triangular pieces of waste, fit the miter joint and re-tape. Continue fitting the inlay around the margin. Then glue all the strips in place, & clamp using waxed paper between the glue block/caul and the inlay. Epoxy adhesive or liquid hide glue are great for 1/32" veneer because they get tacky enough to hold the veneer down without curling. Also, there will be squeezeout and either one is easier to sand off than yellow glue. Yellow glue is fine for thicker veneer. Carefully plane, scrape and sand the inlay flush with the surface of the workpiece. 344 Table of Contents Notes and Reflections While Shaving Wood Chapter 50 – Veneering I It is possible to make your own veneer by re-sawing planks on a band saw and then surfacing them. I have made some veneer this way, but it is much easier to buy veneer. Depending on the quality of the band saw and the skill of the operator, shop-made veneer can be as thin as 3/16". The resawn pieces need to be smoothed by hand planing or sanding, or with a thickness planer. Normally, they need to be stuck on a carrier board before planing. admit to a bias in favor of solid wood furniture, or combinations of solid wood and veneered hardwood plywood. However, as beautifully grained wood becomes scarcer, more of it is going to the veneer mill than to the lumber mill. Also, some grain patterns, such as crotch grain, are impossible to obtain or even to work as large pieces of solid wood. Veneer seems the only option for those. Finally, veneer allows curved surfaces that would be impractical in solid wood. This chapter is meant to be an introduction to the subject, notes from my limited experience with veneer. They may be enough to support a small project, but for a serious project, I highly recommend that you get a book on the subject. All of the references listed below are useful, even though they contradict each other in some respects. Obviously, veneering is open to personal preference. The book by Mike Burton is comprehensive and authoritative, but offers almost too many choices for a beginner. Substrate For quality work, Baltic birch is an excellent substrate. Solid wood, e.g., poplar, is traditional (in the sense of old-time), but the need for edge banding the end-grain presents a problem because solid wood will expand and contract with changes in environmental moisture. My impression is that professionals use MDF as a substrate. I think it’s too dense and too subject to moisture damage. You could use other hardwood plywood, but it is already veneered and is expensive. Fir plywood will telegraph its grain pattern through the veneer. I have bought raw veneer from Highland Woodworking and from Rockler. The Sauers & Co. Veneer from Highland was sequenced matched and quite good quality. I think that’s what Woodcraft sells as well. Types of Veneer Veneer is commercially available in three forms: raw, paper backed, and veneer backed (cross-banded). Raw veneer is the least expensive (everything else being equal) and offers the widest variety of wood species and grain patterns. Veneer backed veneers are probably easier to work with because they are thicker and more stable. I have a bias against paper backed veneer, for no solid reason except that it tends to have the thinnest wood layer, and I wonder whether the paper backing is as strong and permanent an interface as wood. Cutting Veneer: Use a piece of melamine coated MDF to act as a cutting surface for veneer. A scrap from Home Depot costs cents or a few dollars. A hooked “linoleum” knife works well for cutting across the grain. Hone it to a razor 50-1 Table of Contents 345 Veneer edge on a buffing wheel. Always cut from the outside inward, toward the center of the piece, and don’t try to cut even thin veneer in one pass. A properly sharpened veneer saw seems to be a bit safer for cutting with the grain, because it is less likely to get caught up in the grain, or to split the veneer. I prefer a wooden straight-edge to a metal one, to protect the knife or saw. I press down hard and clamp the straightedge to hold the veneer in place. Cut veneer about ¼" over-sized, to leave no more than c" on each edge. More than that is difficult to plane flush because it tends to bend or break with the grain, which is trouble. Jointing Veneer Veneer butt joints (seams) have to be jointed to look right, just as much as joints in solid wood. You can double cut long-grain seams if you can clamp them firmly in place. Jointing seams in veneer is difficult to get perfect. I made a shooting board out of plywood and melamine. I arrange the two pieces to be jointed like the covers of a book and clamp them in place with about c" protruding along the edges to be jointed. I’m not sure whether it is better to use a sanding block or a block plane. If you use a plane it has to be razor sharp. Also, it is even more important to read the grain when planing veneer than when planing thicker wood. If you cut against the grain, you will tear the veneer because there is nothing to support it but the jointing jig. Dry-fit each joint. When the joint fits perfectly, tape the joint together using veneer tape on the show side. Water-based brown paper packing tape or blue “painters tape” will work in a pinch. Remove water-based adhesive tape with moisture. Warm blue tape 50-2 slightly with an iron and pull it back against itself very gently. The adhesive also can be softened by wiping the tape with acetone, but be careful not to stain the wood. Tan masking tape is not acceptable. Adhesive Based on my limited experience, I prefer white or yellow glue. I clamp the surface using platens and cauls. Both white and yellow PVA glues are thermoplastic after they set, so you can re-stick a loose section with a hot iron. White glue is perhaps preferable because its melting point is lower, about 180E. Yellow glue melts at a higher temperature and is not thermoplastic after it cures fully (which may be an advantage). Always put a piece of typing paper between the iron and the veneer. Most professionals recommend hot hide glue or urea-formaldehyde resin. I feel that hot hide glue is too much trouble. Liquid hide glue works OK but does not tack up as fast as the hot version. I may try ureaformaldehyde one of these days. It is important that the platens apply even pressure across the whole surface, so large sheets may be needed to serve as a platen or clamping caul. Because of its nonstick surface, ¾" melamine would be a good choice. It is important to use slightly tapered cauls, or to place a piece of veneer under the center of the caul so that the center will receive about as much clamping force as the rest of the panel. Space the cauls about every 12" or so. Bleeding of adhesive through the veneer can be a problem. Apply a thin, even coat of adhesive with a paint pad or roller to both the veneer and the substrate, give it a minute to tack up, and apply the veneer to the substrate. You may need to apply a bit of 346 Table of Contents Notes and Reflections While Shaving Wood water to the other side of the veneer to keep it from curling too much, but do not overmoisten it. Hot-bond adhesive (this is basically white glue allowed to set) works OK for smaller parts but a wider (4-6") sheet may expand when moistened with the adhesive and then shrink after the heat is applied. This can split the veneer across the grain. Kirby does not think contact cement should be used, although Burton sometimes uses it. I don’t think I would use it for most applications. 50-3 radius hand plane. (See Chapter 37.) Smoothing A card scraper or cabinet scraper seems safe to use. I would not sand veneer much or touch it with a smoothing plane. It is important to use a block behind the sandpaper and to keep it flat so as not to round over the edge. References Burton, Mike. Veneering A Foundation Course. New York:Stirling Press 2006. Trimming I trim the edge of the veneer flush with the edge of the panel with sandpaper or a very sharp block plane, prior to edge banding. Edge Banding I prefer to use solid wood edge banding, but hot-melt tape will work as well. I apply the edge banding after the veneer, so that the edging covers the edge of the veneer. Then I ease the edges of the edge banding with a c" Kirby, Ian J. Laying Veneer: Meeting the small shop's pressing needs. Fine Woodworking #47 Kirby, Ian J. Veneering : Preparing substrates is the first step Fine Woodworking May/June 1984 Schrunk, Thomas R.. An Introduction to Veneering Simple techniques and common tools produce stunning panels for doors, box lids, and more. FineWoodworking January/February 2007 347 Table of Contents 50-4 348 Table of Contents Notes and Reflections While Shaving Wood Introduction to Projects, Et Cetera. T plans for your own needs anyway. Even if they suit you well, however, please do not follow them blindly. I have tried to make them accurate. I have built each of them, at least once. They are valid, and will guide you as you make the piece. Nevertheless, please check the drawings against the cut list and milling list. Don’t cut out all the pieces and then hope that they all fit. Check dimensions and other details as you go along. Also, as I gain experience, I find that I do not draw and describe as much detail as I once did. I now usually know where a mortise should be cut, for example. I documented the aspects of each project that seemed to need explanation at the time I was designing it or reviewing someone else's design. I might have left something out that was obvious to me, but it may not be obvious to you. Finally, these projects and methods may not suit your skill level, and you may not be able to follow them safely. I can't take responsibility for that. I do not wish to lead you into danger. With all due respect, you have to take responsibility for what you do in your shop. he following pages describe a few projects and other articles that especially interest or amuse me. The projects are listed roughly in the order that I made them, so the skill level they require (and the degree of detail in the descriptions) somewhat reflects my development as a woodworker. Each includes a drawing, parts list, and narrative. The “et cetera” are reflections, commentaries and such, sprinkled here and there to distract you from the serious business of building furniture. I think of them as Easter eggs, but you may liken them to other things that one finds in a lawn. Learning about woodworking has, for me, included learning some fundamentals of design. (See Chapters 7-9.) Some woodworkers claim that they don't make detailed designs, because the work tells them what to do next. I cannot work that way, but I do find that my designs often change in the midst of construction, for a variety of reasons. The point is that some changes I make in my shop don't make it back into my written descriptions. Sometimes, a wood-worker who is following plans he did not develop himself will complain that the plans contained one or more errors – usually incorrect dimensions of a component part. The user blames the plans, which seems pointless. The real problem is his own lack of preparation to build a piece of furniture. Blaming the plans will not save your project. Thinking critically and planning ahead will. These designs are probably not perfect, either. If you think that project plans are supposed to be perfectly correct and unambiguous, then please consider these as detailed concept descriptions instead of plans. Chances are, you will want to modify these Prospectus I modified the plans for the bedside table and the Adirondack chair from published plans. The others are more or less original designs, but each one is based on a bit of research into design and construction options. The bedside table (nightstand) was my first serious project. I have made three pairs of these and a three drawer chest based on this design. The second pair, which I describe, was -i- Table of Contents 349 Introduction to Projects, Etc. larger than the first, had double doors, and used knife hinges. The dropleaf end table was a step up in difficulty. It has rule joints and the legs attach with sliding dovetail joints. The square coffee table has four large drawers and is finished in black dye and stain. The kitchen table with drawer has turned legs and a drawer at the end. I cut the drawer opening into the apron. It is another exercise in using sliding dovetails to attach the legs to the aprons. I have made two versions of that table and look forward to a third. The writing desk has three drawers and a little removable bookstand or gallery at the rear. The front apron and the drawer fronts were cut from the same piece of wood. The Adirondack Chair (Jakes Chair, Amended) is my minor variation of a well-recommended design available on the web. -iiTable of Contents 350 Notes and Reflections While Shaving Wood Chapter 51 Workbench T he workbench is, to me, the most important tool in my shop. It can be as simple as a piece of plywood on sawhorses or screwed to base cabinets, or as complicated as a reproduction of an 18'th century English or French classic. The main criteria to me are that the workbench fit me, that it be sturdy and economical, and that it be fitted with the accessories I need. Fit Fit is especially important. I once had a temporary workbench that was too short for me. Consequently, I worked bent over. Before long my back went into spasm from my poor posture and I could not work for over a week. I had to put risers underneath it, but it was never as sturdy as it had been when in stood on its own legs. I’m 6'1" tall, and as I said I can’t stand to work bent over. Also, my vision is not the best, so I like to get close to joinery and other fine work. So, my everyday workbench is 39" high. I have another bench that I use for planing (surfacing) with hand planes and some hand sawing. It is 38" high but I plan to lower it to about 34" high. Also, your work surface should be deep enough to hold most of your assembly work, but shallow enough that you can reach the wall behind it without straining. For me, that’s about 32-33". The length of the bench is the most flexible dimension. I think a bench should be at least 6' long, but mainly it should fit into the space available. Sturdiness is another relative criterion. I want a bench that I cannot move with my full strength, for example, while planing or sawing, but it does not need to be made out of a half-ton of railroad ties. I really mean resistance to racking, which is primarily resolved by design and joinery, and only secondarily by mass. The resistance of the bench to sliding is a separate issue. I address it by weighting it down (stuff on the shelf) or bolting it in place. This bench does not slide. The top should be dead flat, so that I can use it as an assembly table (reference surface). Some solid wood tops can warp slightly with changes in ambient humidity, and need periodic flattening. The top should be stiff, thick enough to hold bench dogs and the like, and durable enough to stand up to pounding and the other insults that a woodworker’s bench will receive. Some soft or brittle woods like larch are not really strong enough. Hardwood plywood (with lumber or veneer core) usually is. Economy Finally, there is economy. I admit that I greatly admire $1000 - $2000 ready-made solid beech Swedish benches, and the English or French reproductions that take six months to make. It might be fun to make one. I seriously considered it. At the end of the day, however, I decided that they are economically impractical in terms of time and money. I could not wait the months required to make a fancy bench, especially because I would need a bench to make one. Besides, once you had a 51-1 Table of Contents Sturdiness 351 51-2 Workbench bench like that, would you really be willing to endanger it by using sharp tools anywhere near it? Materials to build the bench described below cost less than $200. The bench itself should take about a weekend to make. Description 1. Dimensions overall should be 32¾ x 81½ x 39"H. 2. The base is frame construction, using KD Douglas fir. The top is a 2" thick exterior flush door, laminated to a sheet of ¾ birch plywood.a This makes a 2¾" thick top, which is stiff, and thick enough to hold bench dogs. You could optionally add another lamination: another sheet of birch plywood and/or a sheet of ¼" Masonite on top. Procedure Decide on modifications of the design to accommodate vises. I decided to use a steel face vise as an end vise. It has worked out very well. Your dimensions will be different in some respects, e.g., mounting positions for vises. FRAME 1. Cut the legs to length. Mark the top end of each leg FR (front right), FL, BR, BL. Mark a . This assumes that you can find an inexpensive (perhaps used or cosmetically damaged) solid core flush door. You want the top to be dead flat, which is why a flush door is a good choice. If one is not available, you can use a torsion box instead. If you make a torsion box, you will still need a dead flat surface to make it upon (could be a hollow core door). Also, plan the webbing to be located where you will want the dog holes to be. Table of Contents them for ½" W x 2" L x 2" deep mortises in the legs starting ¾" and 22¾" from the top. Front legs get 3 mortises each, two at ¾" down on adjacent inside faces and 1 at 22¾” down on the rear face. Rear legs get four mortises each The midline for the mortises for the back long rails are at 1¼" from the outside edge. Center the mortises for short rails (aprons). 2.Cut the five long rails to length. Save the cutoffs as stock for the short rails. Mark and cut a ½" wide x 1½" deep x 2" long mortise centered on the top edge of the bottom rear rail and the bottom edge of the top rear rail. 3. Cut the 4 side aprons (short rails) to length. On the two top aprons, mark and cut a mortise ½"W x 1½"D x 2"L centered on each bottom edge beginning at 9" from the front shoulder (11" from front end of rail). (Mortise ends are at 9 and 11" from the front shoulder of the apron. (Mark these aprons) Mark and cut a ½"W x 1½"D x 2"L mortise centered on each top edge at 9" from the front end. (These mortises correspond to mortises in top rails) Mark and cut a ½"W x 1" D" x 2"L vertical mortise centered on each inside face of each short rail at 7½" from the front shoulder. Mark which is the top and front end of each. 4. Cut a rabbet ¾" W x ¾" D (as thick as the shelf will be) in the top inside arrises of the bottom long rails and about 14 ½" along the side spreaders. 5. Lay out tenons (nominally ½"W x 2"L x 2"deep) on the ends of the aprons and rails to match the mortises. Mark each mated board. Cut tenons. 352 Notes and Reflections While Shaving Wood 7. Lay out tenons (nominally ½" thick x 2"W x ¾" L on the other rails and spreaders 8. Assemble the ends. Make sure that they are flat TOP 1. Prepare four gluing cauls (2 x 4 or 1 x 2) about 44" long with four 1/16' thick pads. Have eight 3½ - 4" “C” clamps ready with 816 clamp pads. Have six blocks of 2 x 4" waste. You will be able to lay the flush door on the workbench frame to provide support while you apply adhesive, but you should plan in advance how you will support the second and subsequent layers, if any. The flush door will not warp when you apply adhesive but plywood may warp. Therefore, you should be able to clamp the laminations to the flush door (etc.) until the adhesive has set. Possibly the most straightforward approach would be to let each lamination dry before you continue to apply laminations. PVA will start to grab in two minutes or less. For longer working time, consider liquid hide glue. 2. Cut the plywood (and Masonite) to size. Apply adhesive to the top of the flush Table of Contents 51-3 door with a paint roller. Apply adhesive to one side of the ¾" plywood in the same way. Put the two adhesive surfaces together. Quickly align the sheets using a stop block. (You have less than 2 minutes before PVA starts to grab). Using the cauls and the “C” clamps, clamp the stack to the work surface. Clamp about 15" on center with cauls. Let dry overnight. Repeat for Masonite and any additional layers of birch plywood. 3. Joint the edges of the top with a hand plane or router to that they are even 4. Cut 1 x 4 oak trim (fascia) to fit the plywood top. Allow an extra 1½" to cover the ends of the side trim The trim should account for the front and end vises. Joint the top edges of the trim so that they conform to the work surface. Use flat work surface as reference. 5. Trim each proud edge with a bearing bit in a router. 6. Prepare the vise mountings. Front mounting pad is 10 x 17¾ x 1¼. End pad is 7 x 13 x 1 7. Dry assemble the trim pieces around the plywood and Masonite with clamps. Lay the end vise pad in the corner. 353 51-4 Workbench Figure 1. Workbench Exploded View. Cut List for Workbench Part # No. Needed Name Dimensions Milling 1 4 Legs 4 x 4 x 36 #1, 2 2 4 Side aprons and short rails 2 x 4 x 27" ovl 3 3 5 Front/Back Long Rails 2 x 4 x 64" ovl 5,6 3 Vertical Rails 2 x 4 x 21½” 1 Top Skin ¼ x 32 x 80" Masonite Top ¾ x 32 x 80" birch plywood 4. 5 1 Top Base Layer 32 x 80 x 2-3" thick exterior flush Door 6. 1 Long top trim Oak 1 x 4 (nom) x 81½ Table of Contents 354 Notes and Reflections While Shaving Wood 7. 8. 2 1 Short top trim Oak 1 x 4 (nom) x 32 Shelf ¾" plywood about 19"W x 60" cut to fit End vise mtg pad plywood 7 x 13 x 1" ?? 9. 1 Front vise mtg pad Materials (Shopping List): 51-5 plywood 10 x 17 x 1¼" ?? 32 x 80" solid core flush door 5, 64" 2x4 long rails, includes Douglas Fir or So. Yellow Pine ...4, 30" 2x4 short rails ... and 3, 21½" 2 x 4 vertical rails (sides and back) 8 gluing cauls, 7, 2 x 4 x10 4, 36" 4x4 legs 2, 4x4x8 KD Douglas Fir or So. Yellow Pine 2 , ¾" x 4 x 8 2 birch plywood 1 , ¼” x 4 x 8 Masonite 2, 32" 1 x 4 x ¾" oak 1 pc 32" long Oak or maple 1, 81½" 1 x 4 x ¾ oak 1, 1 x 4 x 12' Oak or maple pipe clamps 4 steel pipe at least 8" long 4 36" bar clamps 2 24" bar clamps 3 6" “C” clamps 8 Adhesive for top Accessories Of course, you will want at least one vise. This is a subject of its own. I have been satisfied with two steel face vises, one at the left end of the bench and one as an end vise. Table of Contents gallon of titebond or liquid hide glue You will at least need dog holes to work in partnership with the dogs on the vise jaws and to hold planing stops. By all means, plan on round ¾" dogs, which you can buy in brass (safer for sharp blades) or make out of oak dowels. A bench hook can sometimes be used 355 51-6 as a substitute for a vise. The cleat along the bottom is held against the edge of the workbench, while the workpiece is held against the top cleat. There’s a lot more to say about workbenches. See pp. 6-7 in Notes and Reflections While Shaving Wood. An excellent reference for further reading is Christopher Schwarz Workbenches (Revised Edition): From Design & Theory to Construction & Use Appendix: Mortise Jig There are 14 large mortises to be cut in the legs. If you have a plunge router, a simple, purpose-made mortise jig will help cut them. It is a modification of a Frid jig, named for Tage Frid. It is easy to make, simply a trough that is a bit more than 3½" wide (assuming the actual dimension of the leg is 3½" by approximately 3¾" high in inside cross (Figure 2). The top rails must be equally high (equidistant from the bottom) and coplanar with the bottom. If you have guide bushings, you can make a top with a slot in it that will establish the position and length of the mortise. (That’s the modification to the original jig.) It would be most convenient to cut the slot so that the mortises will be centered on the legs but if you want them to be offset, just pay attention to the reference surface of each leg when you put it in the jig. The correct width of the slot and thickness of the top will depend on the guide bushing that you use. If you do not have guide bushings, you would not use the top. Instead, you could use a square base plate that runs between rails on the tops of the sides, with stop blocks tacked onto the sides to limit the length of the mortise. You position the leg in the jig, clamp it in place, and cut the mortise with the router. I strongly recommend removing no more than ¼" per cut and using an upcut spiral bit. References If you are not familiar with cutting mortises with a router, please refer to Chapter 20, Safe and Accurate Routing.. Figure 2. Mortise Jig with Top Tage Frid's Mortising Jig is described at http://www.highlandwoodworking.com/tage fridsmortisingjig.aspx section and long enough to support the leg. 356 Table of Contents Chapter 52 – Reflections on Imperfection Why, then, 'tis none to you; for there is nothing either good or bad, but thinking makes it so. Wm Shakespeare, Hamlet M me. Frankly, I would have preferred to junk it and start over in my pursuit of perfection. But more lumber would have come out of my pocket and the pain of making it again would have been greater than the pleasure of getting it right the second time. It survived my dissatisfaction through final finishing. I was still not quite satisfied with the appearance of the finished case. In fact, I was sick of it. My best friend and harshest critic (a.k.a. wife) thought it looked fine, even after I pointed out the flaws. We decided that delivering it would be the least worst conclusion. So, I did. I made up my mind not to let my disappointment show. As my clients looked at it for the first time, however, I could not stop myself from saying that it was challenging to get the piece to turn out as I wanted. (I must have felt that I wanted to give them an opening to express their disappointment.) They did not seem even to hear me. They were delighted. They are normally a bit reserved, but they were beaming and enthusiastic. They started to express pleasure about how it fit into their kitchen, how well the color matched their cabinets, and how nice their treasured ceramics would look displayed on it. It’s easy enough to explain this with, “beauty is in the eye of the beholder.” That’s a bit glib, since they had already accepted more than 99 % of the appearance of the piece from a rendered drawing and a color sample. The quality of craftsmanship seems to lie in details. But the details are different for the craftsman and the client, and that’s the point of this reflection. ost woodworkers strive for perfection. It’s a matter of pride. We feel that we owe it to our client, whether family, friend, or customer. We don’t want to spend a lot of time to make an imperfect piece. Nobody would pay a lot of money to put an imperfect piece of furniture in his home. Of course, perfection is an idea, but never a reality. When you consider the number of operations required to make a piece of furniture, it is obvious why we can’t achieve perfection. The closest we come is when a client says that a piece is perfect, even despite minor flaws that are obvious to us. (There is, of course the opposite experience, of producing something that you think is quite well done, only to have a client pick on it or dismissively say that it’s “nice”.) A piece might have minor flaws, even when things have gone very well. When things did not go well, a few problems can quickly add up to an imperfect result. This is a woodworker’s equivalent of having a crazy aunt living in his attic. Everybody knows she’s there, but nobody knows what to do about it, so we don’t talk about it much. Short of perfection, what constitutes excellence? My short answer is that an excellent piece must be attractive, durable and useful. But we have to recognize at least two perspectives on each of these criteria: the craftsman’s and the client’s. I recall a display case that I had planned carefully, but not carefully enough. As a consequence, the execution did not go well. I considered it on “life support” since glue-up. Even though it was square and soundly joined, its imperfections grated on Table of Contents 52-1 357 52-2 Imperfection For the craftsman, the important details may be craftsmanship itself, such as how closely a joint fits, whether the drawer reveals are equal all around, etc. For the client, the important details may be the impressions given by rounded vs. sharp edges, minute shades of color, or the appearance and feel of the surface. For some clients, the very fact that they possess a one-of-a-kind piece, hand made just for them, may blind them to small flaws. A piece of furniture has more detail than the viewer can take in all at once. A woodworker spends a lot of time looking at a piece that he is making, striving to carry out each operation as well as possible. Most clients look closely at furniture a few times and then put it to use. This isn’t to say that they don’t continue to admire it, but they no longer examine it minutely. In fact, when we want to say that something is not very striking, we say that it is just part of the furniture. So, the principle that first impressions are important surely applies as much to furniture as to relationships. Some woodworkers openly criticize their work to visitors or even clients, and only reluctantly express pride in the good points of their work. This may feel like modesty or even honesty, but it’s a mistake. If a piece of furniture can sell itself on first examination it can become a valued possession immediately. Its details rapidly recede into “the woodwork.” If not, its flaws may irritate the owner every time he sees it until it goes to a back bedroom or Goodwill. There is so much to see, even in a simple piece of furniture. The client may never notice the flaws that have been irritating the craftsman for days and weeks. Perfection may be in the details, but they are the details important to the client rather than the craftsman. Table of Contents This can lead us to a new understanding of producing, marketing and selling custom furniture. We are all influenced to some extent by the way that factory furniture is mass produced. Those products are typically designed according to the needs and wants of the typical customer, mass produced, marketed to a “target consumer” and sold by convincing the customer that the piece fits his needs. This is not the sequence for a custom furniture maker. With custom furniture, marketing comes first and selling comes at the very end, if at all. True, when we are making a proposal, we may need to sell ourselves, but we should not be selling furniture. Leave that to stores selling mass-produced goods. Marketing in this sense is not packaging, distribution and advertising. In this sense, marketing is the creation of a product that meets the unique wants and needs of the specific client who is commissioning it. A custom craftsman does not need focus groups or surveys, but he does need to talk to the client to understand her (his, their) wants and needs in detail. What does attractiveness, durability and usefulness mean to the client? Of course, we should suggest design ideas. But then we should present drawings, finish samples, and even mock-ups to make sure that the client loves the piece in advance. Some clients, it is true, are sophisticated (and assertive) enough to expect this, but even when the client does not expect it, we should offer it as a part of our package. Complete satisfaction (and even sign-off) before the piece is built should be a normal part of custom furniture making. The selling of a piece of custom furniture should come at delivery time, when we show the client how beautifully 358 Notes and Reflections While Shaving Wood our work actually satisfies what he asked for. That’s as close as we will come to perfection. If it fits every other expectation, Table of Contents 52-3 even a piece with minor flaws may be accepted as “perfect.” 359 Table of Contents 360 Notes and Reflections While Shaving Wood Chapter 53 – Large Bedside Tablea T The front of the top bows out about 1" and the edges of the top are molded. The door has a raised panel, the legs taper slightly at the bottom, and the door and drawer have glass knobs. The back of the piece is finished, to extend its usefulness. It could someday become an end table in a living room. My second version, shown in Figure 2, is the one described in this article. It is 24" x 18," maybe about the maximum size. The top has the same ridged roundover edge treatment as the previous mahogany version, but the front corners are radiused about ½". The outside edges of the legs are also rounded over with a ½" radius. The solid wood top added significantly to the cost of materials. Edge-banded veneered plywood could be substituted. It has double doors. his project is a good exercise in frameand-panel construction. I consider it intermediate in difficulty. It uses a variety of construction methods, such as M&T joints, grooves and rabbets, dovetails, finger joints, drawer construction and fitting, raised panel doors. When I wrote on p. 13 about choosing a project that was achievable but challenging, this is the one I had in mind for myself. The table shown in Figure 1 was my first serious “fine woodworking” project. I made a pair of them. They took me a long time to complete. I made some mistakes, learned a lot and gained confidence. The purpose of a bedside table is to hold things such as a lamp, an alarm clock, books, and a glass of water, so the top should be large enough to accommodate such items. The original table plan had a top 15" x 18", which is about the minimum size. Also, the height of the table should approximately match the height of the bed, or be a bit lower. M y modifications, in mahogany, of Tunnard’s original design (in pine) were to match the furniture style in the room that the table Figure 3. Smaller Mahogany would be in. Table Construction 1. Cut off a piece of 8/4 stock wide and long enough for the legs (allowing for saw kerf and the need to trim the legs square). Rip the legs a bit oversize and plane to a final thick-ness of 1e". Trim them to length and squareness Figure 4. Larger Version a Adapted from plans by Ken Tunnard http://www.canadianhomeworkshop.com/proj/night_stand.shtml Table of Contents 53-1 361 53-2 Large Bedside Table Mark the top of each leg for its position on the table (Left, right, front, back) and mark the top with arrows to indicate the inside surfaces. This will avoid errors later when cutting mortises and grooves, rounding edges, tapering bottoms, etc. 2. Lay out the pieces on the 4/4 stock with a piece of chalk, marking oversized locations for rails, etc. for the carcase. (Although the final size of door and drawer parts should be determined after the carcase has been assembled, some sizes can be planned in advance, e.g., width of rails and stiles, approximate width of door sides and back, etc. Rough stock should be surfaced now to reduce set-up time and so that it can be milled with the legs and rails.) (I made a layout that would eventually break down into 4" widths so I could use my 4" jointer for initial surfacing.) Separate ¾ thick pieces (back and side rail stock) from front rails, if they will be thicker. (Plans call for f" finished thickness for front rails.) 3. Cut the finished pieces to size. Rip all pieces to width (oversize if they will need final smoothing) & then cut to length. (Cut oversize for drawer and door parts). Number every piece in chalk according to the cut list. 4. Cut mortises in legs. a. Mark locations for the front rails on the front legs, and cut mortises into the left and right inside faces. I make a story stick to ensure that the mortises are precisely the same distance from the top. Note that the top of the top front rail should be flush with the tops of the legs. b. A plunge router with a ¼" up-cutting fluted spiral bit works best for me, as described in Chapter 20, Routing. In brief, I prefer to use a ½" O.D. guide collar running in a ½" slot cut into a piece of ½" Table of Contents plywood, screwed to a nice square piece of wood or MDF, Make sure that the leg is snug up against the template so that the mortise will be square to the surface of the leg. c. I don’t square router-cut mortises with a chisel unless they are going to show through on the opposite side, and none of these will. I prefer to round over the corners of the tenons with a fine wood rasp. Sand paper on a flat stick also works well. 5. Cut tenons on rail ends. a. Mark out each different size tenon on at least one workpiece. Marking each one is better, but I did not do that for this project. See Chapters 19 and 21, Cutting Tenons. b. I cut these tenons with a commercial tenoning jig on a table saw. It’s important to use a rip or dado saw blade with a square top. 6. Cut ¼" x 5/16"-deep grooves in the back rails, side rails and legs to receive the stub tenons and the plywood panels. Nominal quarter-inch plywood is often 15/64" or even 3/16" thick. It will be loose in the quarter inch grooves. Consider cutting small wedges or shims for glue-up time. You could cut undersize grooves but sometimes these are too tight and make the panels difficult to assemble. a. Cut the grooves with a router, preferably mounted in a table. Use a stop block to stop the dado at 2½ - 3" from the bottom of each leg. (You could also use a dado stack on the table saw, and finish the grooves with a chisel or a router plane.) The horizontal position of the grooves on the legs is an appearance issue. Just be sure that they are all precisely the same offset from the front. The grooves on the legs and rails must line up well to receive the plywood sides and back. Also, note that you will have to reposition the fence for half of the cuts, unless the legs are exactly square and the grooves 362 Notes and Reflections While Shaving Wood 53-3 happen to be exactly centered. b. Dry-fit the side and back rails and check the location of the grooves in the rails. 7. Rout c" wide slots in the inside of the top front and top rear rails. Cut them about 2-3" long, at about 3" from each end, and e" from the top. These will receive the drawer kickers. (The slots can be wider than 3" but should line up front rail to back rail.) Use a slot cutter, or a router plane. 8. Rout ¼" x d" deep grooves centered in door stile and rail stock. 9. Rout ¼" x ¼ “ grooves in stock for drawer sides, ¼" from what will be the bottom edge. 10. This is the right time to decorate the legs if desired, e.g., taper the bottoms, round the outside edge, etc. Shallow beading would add a note of formality to the design. 11. Dry-assemble legs and rails (i.e., without glue) . a. Measure for the plywood side panels. Add distance for the grooves to the distance from bottom of top rail to top of bottom rail, etc. b. Measure for drawer guides a. Distance from top of leg to top of middle front rail. b. Distance between back and front legs (from front of back leg to back of front leg.) c. Measure for bottom: distance between rails in both directions (front to back and side to side) 12. Cut plywood back and side panels. Dryassemble and test-fit for square, etc. Figure 5. Simple Drawer Guide Assembly 13. The drawer guides run between the front and back legs. The simplest version is made from secondary hardwood -- oak is ideal. It is “L”-shaped in cross section, consisting of a side guide and a bottom guide glued together. a. Cut the lower drawer guide from a piece of hardwood about ½" thick by 1¼" W x 14½" L. The actual length should fit from just behind the middle front rail almost to the groove for the back panel. Cut notches at each end so that the lower guide can fit around the front and back legs. b. Cut another piece ¾" square by 13e"” L. The actual length should fit closely between the front and back legs. The actual width should be the distance from the plywood side panel to the inside of the leg. Glue the bottom guide to the side guide. These will be fitted into the case and be attached (glued, screwed) to the inside of the legs when the sides are assembled. c. Alternatively, make a grooved drawer guide out of 1-1/2 x 3/4" oak and mortise it to the front and back legs. i. Cut tenons in each end (by hand) ii. Plow a ½" x ½" groove, ½" from the bottom edge iii. Trim 1/4" off the top, leaving a 1/4" deep runner and a ½" deep bottom support 14. Bottom. Cut the bottom of the cabinet and Table of Contents 363 53-4 Large Bedside Table cut notches for the legs by hand. Cut it slightly (1/8") undersized from front to back. Plywood is preferable, because solid wood may expand too much in humid weather. Cut long ledger blocks and attach them to the front, back and side rails. Allow the bottom to float, or glue it to the blocks. If the bottom is made of solid hardwood, glue it only on two adjacent sides, e.g., front and right side rail, to allow for seasonal expansion and contraction. the router will copy any dips or ridges across the whole edge.. I use a ½" roundover bit with a d" bearing, precisely centered. On a ¾" top, this leaves a c" wide ridge at the top and bottom. This looks even better if the top is f" thick. I cut the edge treatment on a router table, using the guide pin and bearing to control the work. Doors The doors are rail & stile construction, with a solid wood frame surrounding a panel. You can use plywood (¼" to ¾") or solid wood. A solid raised panel looks a lot more refined than a flat plywood center panel. If you glue up a panel, remember that the bevels on a raised panel may cut across the joint, so avoid dowels, etc. Use a butt or rabbet joint. Figure 6 Dimensions for Curve in Front of Top 15. Prepare the top now, Make up a solid wood panel, an edged plywood panel, etc. I like a curved front that protrudes an inch or two from the top rail. A nice curve for the front has an arc with 40" radius that intersects the straight part of the front 3" from each side. So the chord of the arc is 18" and the rise is about 1". Cut out any curved lines on the bandsaw and fair them with spokeshaves and sandpaper. The edge must be “perfectly” smooth and regular before you cut an edge treatment with a bearing-guided router bit, or Table of Contents 1. Cut the rails, stiles and panel to fit the opening, a bit oversize to permit trimming. Cut the panel to the size of the inside of the frame, plus enough for the tenons to fit inside the rails and stiles. In this case the opening for the two doors would be 18 f x 12", each door would be 9 f x 12"(nominally) , the rails and stiles are 1½" wide, so the opening in each door would be 6 7/16 x 9". Since you cut ¼" deep grooves in the rails and stiles, you need to allow a bit less than ½" Cut the panels ½" wider in each dimension, i.e., 6 15/16 x 9 ½, and then plan to plane off a bit at assembly time to allow for expansion.. 2. Cut ¼" thick by ¼" deep grooves in the rails and stiles, centered. If the tenons (tongues) on the edges of the panels are centered and the grooves are centered, the panel will be flush with the rails and stiles. For a raised panel, see Chapter 36, Cutting Raised Panels. Cut the panel to size and then cut a centered ¼" wide x d-¼" deep tenon along all four edges of the panels. Cut 364 Notes and Reflections While Shaving Wood the bevels using a panel raising jig. The sawblade should just barely touch the corner of the tongue. Cutting the bevels is a bit tricky. The workpiece will tend to slide downward and so must be held securely against the jig. A zero-clearance throat insert is necessary so that the cut edge can’t catch in the throat plate. Once the rip fence, jig, saw etc are set, do not move them or subsequent bevels will be cut asymmetrically. Don’t glue a solid wood panel into the frame. Cut or trim the panels slightly small to leave room for expansion/contraction within the grooves. If necessary, use a spot of glue only at the grain ends near the center of the panel. Assembly 1. Dry-fit carcase for a final test, including clamping. Check everything for square, diagonals equal, etc. Remember drawers and doors are inset so openings must be exactly square and straight. 2. Top drawer guides The top drawer guides (drawer kickers) will strengthen the carcase and serve as an anchor point to secure the top. If they are approximately the same thickness as the top front rail, they will also help prevent the drawer from tipping forward when fully opened, when using the simpler, “L” shaped guides described above.. a. Measure the distance between the front and back top rails. Cut two pieces of secondary wood or plywood to fit that opening (about 7/8" x 3" x 14-7/8") b. Cut c" x ½" deep slots in each end. The slots should line up with the slots cut earlier in the top rails, so that the tops will be flush with the tops of the top rails (about 3/8" from the tops). c. Cut hardwood splines, 1/8" thick, x Table of Contents 53-5 2-3" L x 1" W, two for each end of each top drawer guide. d. If you are using a solid hardwood top, cut a groove ½" from the top along one or both edges of each top drawer guides, to accept metal or wood clips for attaching the top. 3. Fit the doors. Mark exact locations for knife hinges on top and middle rails or for butt hinges on the front legs, and on the doors themselves. (Note that a single door should swing away from the bed, so the doors on a pair of bedside tables should swing in opposite directions.) 4. Take it apart. a. Cut mortises for hinges. It’s much easier to cut hinge mortises while the piece is disassembled., especially knife hinges. b. Finish sand interior surfaces as needed. 5. Glue-up. This is usually the most anxious part of construction. Decide on a procedure before you start, and have everything at hand. a. It is best to assemble the sides first, so that there is room to measure and attach the drawer guides. b. Attach the drawer guides. c. Assemble back rails & panel, and then front rails. d. Glue splines into the c x ½" slots cut in the insides of the top rails. When they are in position, apply glue to the splines and slide the upper drawer guides into position. If these were cut accurately, they will help to square up the carcase horizontally. e. Check everything for square. Don’t forget to check the front legs for parallel and the tops of the legs for evenness. The top won’t be glued and it can wait. 365 53-6 Large Bedside Table Drawer Cut the required parts to final length, including the solid wood drawer front. The front should be cut a bit (e.g., 1/16") oversized so that it can be trimmed to fit with a consistent reveal all around. (Consider a raised panel or grooved effect for the drawer front to match the raised panel on the door.) The layout of the drawer sides may be critical if you chose a grooved shape (step 13c) instead of a “L” shape for the guides. The drawer guide needs a ¼" groove cut into the outside of the drawer side, ½" above the bottom edge. Either use ¾" thick stock for the sides, or fit the bottom into a rabbett along the bottom edge instead of a groove. (I forgot about this and had to glue a 1/4" strip along the drawer side to catch the drawer guide. This meant that the dovetails had to be recessed that extra 1/4", and that I had to cut plugs to fit the recesses.) Groove (rabbet) inside drawer front, drawer sides and back for drawer bottom. Groove sides for drawer guides if necessary. The drawer front joins to the sides with 1-3 half-blind dovetails on each side. Cutting the big dovetails with a handsaw and chisel turned out to be the neatest for me. For the drawers with three dovetails, I used a dovetail bit in my router to do the bulk of material removal, followed by some handwork with a chisel to square up the corners. Use box joints, through dovetails or drawer lock joints for the sides and back. Note that “drawer lock” joints change the length of drawer sides from nominal) Finish Figure 8 Three-Tail Layout for Drawer Figure 7. Single Dovetail on Drawer All dried glue squeeze-out has to be removed Table of Contents 366 Notes and Reflections While Shaving Wood applied a thinned final coat of varnish. from wherever it is visible from the outside. After that was done, I sanded to 120 grit and used a card scraper to prepare the surfaces for finishing. I normally don’t color mahogany unless there are distinct color differences among boards. Materials I made this table from maple, and the previous, smaller version from mahogany. Tunnard’s original was made in pine. Your choice. The “secondary hardwood”, however, should be hardwood. You need at least 1½" thick surfaced stock (8/4 rough stock) for the legs, and 4/4 for everything else. Good 4/4 stock that can be surfaced to f" or so will look good for the top, but you could use hardwood plywood with edge banding. For a procedure for surfacing rough-cut lumber, see Preparing Lumber for Use in Chapter 4. When I made a pair of tables out of maple, I used the following mixture: Boiled Linseed Oil Solvent (mineral spirits & naphtha) Brown Oil Stain Minwax Antique Maple Oil Stain 53-7 6 oz 3 oz 4 oz 2 tsp This made just enough to stain two tables. This mixture popped the grain well and colored the wood moderately. There was some blotching, so I feel that this was a bit more pigment than I should have used for this first coat on maple. An alternative would be to reduce the amount of brown oil stain or apply a washcoat of varnish or sanding sealer cut 1:3 or 1:4 with solvent. After the BLO dried overnight I sealed it with a coat of dewaxed shellac. Then I toned some lighter-colored pieces by wiping on a thin coat of Polyshades Pecan, as it happens to be a close color match to the oil mixture. The outside then received two coats of varnish. The inside received two coats of shellac. I gave the top another coat of varnish, levelled it with 220 sandpaper (dry) and This project requires about 11 bf of primary hardwood • 1, 1" x 8" x 8' long = 5.3 bf • 1, 2" X 8" X 24" = 2.7 bf legs • 2, 1" x 3" x 6' of 1 1/2" primary hardwood ( 3 bf) for top or hardwood plywood • Secondary Hardwood roughly 1 x 4 x 48" • Hardwood Veneer Plywood: roughly 18 x 24" for top or solid hardwood • Quarter sheet of ¼" or ½" hardwood plywood about 30" x 15" • Hardware - hinges, ball catch, knobs Cut List Large Bedside Table (24"W x 18" d x 24" H Number Name Material Rgh Stock Size Qty. primary 8/4 1 5/8" x 1 5/8" x 22-1/2" 4 Case 1 Legs Table of Contents 367 53-8 2 Large Bedside Table Top primary 6/4 or ply* 7/8" x 18" x 24" 1 *Top may look more customized if f" thick. Make up panel with 6/4 rough hardwood stock (e.g., 144" of 3" wide stock) or ¾" veneer plywood with f" wide hardwood edging. Need 84" of 1 x 2 for edging. 3 Front rails primary 4/4 3/4" x 1- 3/16" x 20" 3 4 Side rails primary 4/4 3/4" x 2" x 14-1/2 4 5 Back rails primary 4/4 3/4" x 2" x 19-1/2" 2 6 Side panels 1E plywood ¼" ply 1/4" x 16-5/16" x 14-1/2" nominal 2 7 Back panel 1E plywood ¼" ply 1/4" x 16-5/16" x 19-1/2" nominal 1 8 Bottom 2E plywood 3/4 3/4" x 19-7/8" x 14 5/8" nominal 1 9 Drawer guides 2Ehardwood 4/4 1/2" x 3/4" x 15" measure, cut & apply after sides are assembled, but before sides are attached to back and front 2 10 Drawer guide sides 2Ehardwood 1/4" x 1" x 13-3/4" 2 17 Top drawer guides 2Ehardwood or plywood f" x 3¼" x 11c" 2 4/4 Doors – each table needs two doors - remeasure & cut to fit case 11 Door stiles primary 4/4 3/4" x 1-1/2" x 12-1/8" 4 12 Door rails primary 4/4 3/4" x 1-1/2" x 7-1/8" 4 13 Door panel 1E hardwood if raised panel 1/4 ply or 3/4" solid 1/4 ply or 3/4" solid x 97/8 x 6-15/16 2 4/4 3/4" x 4-1/8" x 18-3/4" nominal 1 Drawer - remeasure & cut to fit case 14 Drawer front Table of Contents primary 368 Notes and Reflections While Shaving Wood 53-9 15 Drawer sides 2E hardwood or 8-ply birch plywood 4/4 1/2" x 4" x 15" nominal – measure actual space after assembly legth depends on lock vs finger joint 2 16 Drawer back 2E 4/4 3/4" x 4" x 18-3/4" nominal 1 18 Drawer bottom 2E plywood ¼" ply 1/4" x 14-1/2 x 18-3/16" allows for 1/4" deep groove in front, back and sides. 1 Knobs 3 Hinges 2 pr. Door Catches Milling Summary Cut mortises 1/4" thick for front rails centered, with bottoms at 31/8", 16-3/16" and 21-7/16" measured from bottom of leg x 7/8 long x 5/8 deep Tenons 1/4" thick front rails x 5/8" long x 7/8 W. Side and back rails x 3/8 long x 1-5/8" W Door rails (later) x 3/8 long x 1-1/2"W (full width of rail minus groove) back of front legs, front of back legs, insides of back legs, centered 3/8" deep door rails & stiles , centered 3/8" deep drawer front, sides and back, The bottom of the groove is 1/4" from bottom edge of drawer front, side, back 1/4" deep. Drawer guides groove ½" from bottom ½ x ½" Box joints ½" dado Grooves 1/4" wide Table of Contents 369 53-10 Groove for drawer guide Table of Contents Large Bedside Table outside of drawer side, ½" from bottom edge 370 Notes and Reflections While Shaving Wood 53-11 Figure 9. Sketch of Bedside Table (From Tunnard, footnote a) Table of Contents 371 53-12 Figure 10. Detailed Drawings (From Tunnard, Footnote a) Table of Contents 372 Notes and Reflections While Shaving Wood Chapter 54 -- Sabbathday Lake But, whistling, make it smooth again. A maple desk beside a wall On graceful gently tapered legs Reflects the windowed morning light, Dappling leaves from swaying trees. This table, poised on dancers’ feet Invites your gaze and draws you near. Some may hate the bold conceit That kills a tree to make a desk, May fear the danger, noise, and smell, Deride my pride in tools honed well. But wood worked well, will show its grain, My skill can make it live again. It welcomes you to sit, reflect. Perhaps to pen a peaceful poem, Or maybe just a shopping list, Or a tardy letter to a friend. And while you sit, perhaps you’ll see The lovely grain that flows from end to end. I struggled many years until I was the master of my tools -And bent their chaos to my will. Each project moved a hundred steps Along my path to hard-won skill, Some lessons served me well, some ill. Once this was a maple tree. In spring, it surged with sugary sap. Pitiless saws sent it crashing down. A sawmill sliced it into planks That sat neglected, except for birds, ‘Til all the sap had dried away. My first instructors, stern and hard, Were busy with their money jobs, With little love for awkward trolls. Instead, it’s clear, they thought they should Cut down our spirits and pound our souls, As if we boys were made of wood. Then more machines, with growl and whine, Shaped those planks and sliced again, Indifferent if they cut a board Or the cautious hands that fed the wood. Hot smells of sawdust filled the air With creative agitation. Once trained with pain, we blamed the wood If it should split or poorly fit, We blamed our tools as cheap and dull. No pounded, cut, and injured soul No broken spirit, not quite whole Could make a table such as this. It was I who measured every cut, Trimmed every piece to fit its place, Caressed the wounded surfaces, Smiled with pleasure, read the grain So my plane not tear the wood. 54-1 Table of Contents 373 54-2 Sabbathday Lake I could never make a lovely thing With grudging, hasty laboring, If I would measure out my time Like copper coins from a miser’s purse. The clock will run. Time takes its fee I won’t trade time for mediocrity. But then I found my rhythm I found my song and sang it out Craftsmanship is grown-up play! Craftsmanship must be a dance. The joy of work must lead the way You cannot force it to advance I know the scorn of hasty men Who hammer joints until they fit, They claim that craftsmanship is dead, Killed by commerce, or so they said. They are the sons of those joyless souls Who had no time to train us boys. An injured spirit can be smoothed. If deeper scars may still remain, They lend a certain character If you blend them well into your grain. Pursuit of craftsmanship can turn An awkward boy to a easy man. But cutting words ne’er changed my shape, Nor scornful pounding make me fit. My only wish was to escape. Bigger tools did not make better work. Had I not found another way, This table would not exist today. I turned myself some Shaker legs I sang and danced a tune so bright Recalling how the Shaker folk Could always stay within the light. I think I’ll keep on dancing ‘Till things “turn out just right.” 374 Table of Contents Notes and Reflections While Shaving Wood Chapter 55 – Mahogany Drop Leaf End Table T tenon joinery also could be used if only the top 2-4" of the tenon were glued and the mortise were cut a little long as with dovetails, above. . Some people recommend two tenons. The bottom of the tenon (or lower tenon) could be pinned with a dowel if the hole through the tenon (but not the leg) were a bit elongated to allow for wood movement. his is a copy of a drop leaf table of unknown age, probably hand made. This project involved a number of interesting and educational issues, namely, sliding dovetails, turned legs, double tenons, closely fitted drawer kickers and rule joints for the drop leaves. Wood Movement Turned Legs One concern was wood movement of the 9" wide table aprons (sides and back) against the legs. The aprons might expand or contract across the grain by as much as ¼" with changes in ambient humidity, while the legs would not change along the direction of the grain. (See Becksvoort) This seemed like a good application for a sliding dovetail joint. The mortise should be cut a bit (½") longer than the tenon to allow for expansion, Only the top 2-4" should be glued. I chose this because it would allow transverse (seasonal) wood movement while strongly resisting lateral movement, without the need for a dowel pin. I did, however, consider some alternatives. The aprons could be made of veneered plywood because the ends and top sides of the plywood would be hidden by the legs and the top. Then normal M/T joinery could be used. In this case, I wanted to be faithful to the original. Also, I had some 5/8" stock that I could use for the aprons. With solid wood, normal mortise and The turned legs should match each other. I made a “negative” template and marked the diameter of each cove bottom and bead top, etc. on it. The ball feet are a subtle challenge because any trimming to even the legs would show up as asymmetries in the ball feet. The legs have to be cut a little long to leave room for the spur center on the lathe. The ball Figure 1 Reproduction Drop Leaf Table 55-1 Table of Contents 375 55-2 Drop Leaf End Table should be turned first, and then the template for turning the legs should be registered to it, particularly the bottom. After turning, the legs will have to be trimmed to length at the top before mortising. (In this case, I made the template to match the original turnings.) - ¼" tenon - ¼" space -¼" tenon - ¼" shoulder. Sliding Dovetail Rails Sliding dovetails should be cut with a slight taper. Otherwise, they are quite difficult to slide together. In their usual application, in bookcases and the like, they are cut with a hand-held router or dovetail saw. The groove is tapered by 1/32-1/16" over its length. In this application, I wanted to cut the groove on a router table. On a table, the groove cannot easily be tapered, so I decided to taper the tail. Cut each groove down the center of the leg. First, hog out a ¼" wide by ½" deep groove with a straight bit and then cut the dovetail groove with a ½" 1:6 dovetail bit. Cut it 8" long. Leaving the height of the cutter unchanged, set the fence slightly less than 3/16" back from the front of the dovetail The top rail is thick enough to have a single tenon. The middle and bottom front rails are ¾" thick and 1-¼" wide. Frid says that side grain to side grain is the only strong glue surface, and the “end grain is not a glue surface.” He wrote that from the perspective of the hide glue era. Polyurethane glue is claimed to be strong for end grain, but I took Frid’s advice and used double tenons. I cut the tenons on my bandsaw. The result was OK but I had to re-cut one of the double tenons. Note that the double tenons are 1-¼" wide and symmetrical, which basically makes them two side by side ¼" tenons : each has ¼" shoulder Figure 2. Apron on shooting board with dovetail plane, leg in background Mortising the Legs Cut the mortises in the legs (to receive the aprons) with a ¼" Forstner bit on a drill press and clean out the waste with a chisel. Alternatively, cut the mortises with a router and jig. (I tried using a ¼" spiral upcut bit in the drill press to remove some of the remaining waste between the holes. This did not work well. The spiral bit is difficult to control and one of the mortises came out a little large.) 376 Table of Contents Notes and Reflections While Shaving Wood cutter. Cut the tails across the width of each apron and then trim off about ¼" for a shoulder, to hide the bottom of the groove. Because the apron stock was e" thick, the fence setting caused the tails to be a bit thick, so I could plane them to fit the grooves. When I planed them with a dovetail plane, I took a bit more off of the bottom end than the top. Soon, the tails slid easily into the grooves, except for the last two inches, which required some force. This is exactly the objective given in textbooks for sliding dovetails with tapered grooves. (See appended notes) Drawer Guides and Kickers Dry-assemble the side aprons to the legs. Each apron should slide in easily with moderate force, and the leg should wobble very little if at all. If you have to use a lot of force to assemble it, the dovetail is too tight and you should carefully fit it by shaving off a few thousandths of an inch wherever it begins to bind hard. Be mindful that you will need to disassemble it to add glue in the next step. If the fit is sloppy and the leg wobbles a lot, glue the thinnest possible strip of wood to the dovetail and re-fit the joint. When you have a snug, square fit, remove the leg from the apron, apply glue to the top 3" or so of the dovetail tenon and reassemble it. Do not apply glue to the mortise, or to the bottom of the dovetail tenon. Apply glue and assemble each joint in turn. Work fast after you have spread the glue. PVA adhesive will act as a 55-3 lubricant for about 15-30 seconds and then will begin to grab as the joint faces slide by each other. At this point, use whatever force you need to bring the leg flush with the top of the apron. The drawer guides are made of two pieces of ¾" hardwood (oak or maple preferred) glued together to form a side and a bottom guide. (See Figure 5 in Chapter 52.) Attach the drawer guides and kickers. The drawer kickers are installed with only small clearance above the top of the drawer sides. This makes a drawer that works as nicely as if it had the guides in a groove. Attach the drawer guides and kickers to the side assemblies with screws but do not glue them yet. They may need some adjustment while you are fitting the drawers. Otherwise, you may need to plane, sand and shim them to make the drawers fit correctly. Figure 3. Dry- assembled Table Stands Without Clamps 377 Table of Contents 55-4 Drop Leaf End Table Top and Drop Leaves Make up the 18" x 22" top and two 9 x 22" drop leaves either from solid wood planks (see Chapter 35) or edge banded plywood (See edge banding in Chapter 37). I like a somewhat thicker top, say f" finished, but ¾" thick is fine. If you use plywood, keep in mind that you will be cutting rule joints and adding hinges, so size the edge banding accordingly. Also, if the hinges will be in the edge banding, consider using a splined or tongue and groove joint to attach the edge banding to the plywood along the rule joint. The drop leaves use a rule joint – a ½" radius round-over cut in the table top and a matching ½" radius cove cut in the leaves. Matching cutter pairs are available but I used cutters purchased separately. A half-inch round-over leaves a d" quirk in f" stock. (See Figure 4.) Perhaps I should have used e" radius round-over & cove but they were not easily available. I made a model rule joint, complete with mortised hinges, out of scrap before cutting the edges of the actual top and leaves. Figure 4. Diagram of Rule Joint Geometry (adapted from Campbell’s Fig. 1) See the article by Mac Campbell in References, and Making a Rule Joint in Notes. 1. Mark each side of the edge and leaves, top and bottom to avoid confusion while cutting. Then joint the top and the leaves just as if you were making a butt joint. 2. Lay out the profile of the joints on the ends. Use the underside of each board as the reference for marking out the joint. (The tops may be trimmed to final thickness later.) Mark the center of the round-over on the underside of the top. 3. Rout the top on the router table, with the fence set even with the front of the guide bearing. Make multiple passes, e.g., four, raising the cutter about c" on each pass, up to the layout mark. Make the last pass very thin, 1/32" or so. Two pitfalls to avoid are letting the bearing get above the top of the board and letting the board get into the dust recess in the 378 Table of Contents Notes and Reflections While Shaving Wood fence at the beginning of a pass (before the board contacts the bearing.) Consider a secondary fence that closes the dust recess somewhat. After cutting, if the quirk is not even in cross-section, plane the top to make the quirk flat. 4. After cutting the round-overs on the top, change to a cove cutter bit and cut the coves on the leaves in about 3-4 passes. Don’t cut all the way to the layout line. Since the top of the leaf is up when cutting on a router table, the cut is referenced to the bottom of the table. After about 3 passes, lay out the joint on a flat surface, top up, to check the fit (in the open position). The final pass, to the layout line, should make the joint fit well in the open position. Trim the edges a bit, and ease the bottom of the round-over by 1/32" or so. 5. With the top assembled on a flat surface I checked the thickness of the boards and planed the tops to be even. 6. Hinge placement is critical. The center of the hinge pin should be set 1/32" closer to the edge of the top than the center of the round-over. The quirk should be above the center of the round-over, so this distance should be the same as 1/32" from a line dropped from the quirk. Campbell gives more explicit directions. I used the underside of the top and leaves as the center of the round-over. If the center of the roundover coincides with the underside of the table, then T=R+Q, and you should mortise the hinge to depth “A”, which 55-5 is the distance from the face of the hinge to the center of the hinge pin. in addition, rout a deeper groove for the hinge barrel or knuckle. See Figure 4. 7. After mortising the hinges, drill one pilot hole for each leaf and test-assemble the joint using steel screws. (Expect many assembly/disassembly cycles. You will wear out brass screw heads, and you certainly won’t want to break off a screw in the hole.) The fitting procedure above should make the joint fit well while open (flat). Some sanding may be necessary for the joint to work smoothly. Use pieces from the test joint as sanding blocks for these curved surfaces, use appropriately sized round and hollow planes (Chapter 27), or use a scraper of appropriate dimensions. Locate the high spots by putting carbon paper in the joint and working it. Remember, any wood scraping at this point will be worsened by the thickness of the finish. The amount of wood you remove is not critical, but the round-over on the leaf does show when the leaves are down, so be conservative. Final Glue-Up The sides have already been assembled, so final glue-up requires assembly of the rails and inserting the rear apron. Assemble the rails first, clamp the two sides lightly to support them, and slide the rear apron into place. In final assembly, be very careful to 379 Table of Contents 55-6 Drop Leaf End Table square up everything, especially the drawer and door openings. The drawer fit will be inset, and the reveal should be even. Do not install the top, yet. Drawers Cut the drawer fronts to fit the openings after the frame has been assembled, about 1/16" oversize, and then trim them to fit. See Chapter 40, Fitting Drawers and Doors. Cut a ¼" x ¼" groove into the bottom inside of the drawer sides to receive the drawer bottom. The drawer sides attach to the front with two half-blind dovetails; the back is attached to the sides with finger joints. It’s a great idea to plan dovetail pins to suit the width of an available chisel. I cut the dovetail tails by hand. I cut the pins on my bandsaw, using a 1:6 taper jig run along the fence. (It might be easier to make the pins symmetrical across the width of the drawer side so you could just flip the board over.) Remove the waste from the pinboard with a coping saw and finish up with a chisel. When fitting drawers with close-fitting lower guides and kickers, remember that the drawer is three-dimensional. One corner of the front may be high because the opposite corner of the back is low, etc. You may need to change the position of the drawer guides a bit to make the drawers work well. Spacer blocks attached to the rear of the drawer guides, after final assembly, hold Figure 5. Diagram of Dovetail and Box Joint Layout the drawer front in position when closed (¼" back from the front edge of the rail). Drop Leaf Supports There are many ways to support a raised drop leaf. The simplest are commercially available metal supports, but they would have been entirely inappropriate for this table. The original table has spinners attached to the tops of the side aprons. This seems to put a lot of torque on the aprons. I fashioned sliding drop leaf supports that attached to the top and the drawer kickers. Like a spinner, they fit into a square notch cut into the top of the side apron. 380 Table of Contents Notes and Reflections While Shaving Wood Another alternative would be a vertical bracket that is hinged to the aprons. Finish I dyed the wood to brown with NGR dye, 55-7 aiming at the color of mahogany after a few coats of shellac and varnish. I had to go a bit darker than that, to match different boards of various natural colors. It is difficult to get an even color with wood dye. Consider washcoating and mixing in some NGR dye retarder to slow drying. Brush it on liberally but carefully, avoiding drips over the edges, and rub out streaks vigorously with a rag or synthetic wool. Finish as desired. (I used gloss alkyd varnish. (See Chapter 32.) Figure 6. Drop Leaf Support 381 Table of Contents 55-8 Drop Leaf End Table Figure 7. Sketch of Drop-Leaf End Table. Drop leaves and supports are not shown. 382 Notes and Reflections While Shaving Wood 55-9 Cut List 1 Top 18 x 22 x f" Primary hardwood (mahogany) 1 Drop Leaves 9 x 22" x f Primary hardwood (mahogany) 2 2 Legs 23-¼ x 1-¾ x 1-¾ cut 1" too long & trim to length after turning 4 3 Side Aprons 17-½ x 8-f x e" Primary hardwood (mahogany) 2 4. Back Apron 13-½ x 8-f x e" Primary hardwood (mahogany) 1 5 Top Front Rail 14 x 1-½ x ¾" Primary hardwood (mahogany) 1 6. Middle and bottom front Rails 14 x 1-½ x ¾" Primary hardwood (mahogany) 2 7. Upper & lower drawer guides 17-¼ x 1-½ x ¾ Secondary hardwood 6 8. Side drawer guides 16-½ x 1-½ x ¾ Secondary hardwood 4 9. Knobs 1-½" D x ¾" L Primary hardwood (mahogany) 2 Buy about 8 BF of primary hardwood. Note that veneered plywood could be used for side and rear aprons. Drawers (2) 2-f x 12-½ x 18-1/2 10. Drawer Front 2-11/16 x 12-½ x ¾ Primary hardwood (mahogany) cut 1/16 oversize 2 11. Drawer Back 2-e x 12-½ x ½ secondary hardwood (soft maple, birch) or Baltic birch plywood 2 12 Drawer sides 2-e x 18-¼ x ¾ 4 383 Table of Contents 55-10 Drop Leaf End Table 13. Drawer Bottom 19 x 13 x ¼" mahogany plywood 2 Milling List Legs turned to match template Rear Legs If solid wood will be used for aprons, dovetail mortises ½" x ½" x 8" long, on two adjoining inner sides Front legs REAR: If solid wood will be used for aprons, dovetail mortise ½" x ½" x 8" long, centered, on side INSIDE: mortise 3/4" deep x 3/4" long x 1/4" wide, ½" from front of leg and 1/4" from top of leg, for top front rail INSIDE: Double mortises 3/4" deep x 3/4 long x 1/4" thick, at 5/8" and 1-1/8" from front of leg for middle and bottom rail -tops of middle rail mortises are 4-3/16 from top of leg; -tops of bottom rail mortises are 7-5/8" from top of leg Side & Back aprons If solid wood is used, cut sliding dovetail tenons (pins) ½" x ½" x full width (8-7/16") and then trim e" off bottom end of tenon to make a shoulder. 1/8" wide x ~1" long x ~1/2" deep grooves, running 3/8-1/2" from tops of sides and back for top clips Top rail tenon 3/4" deep x 1" long x 1/4" wide. 1/4" shoulder all around Middle rail & Bottom rail double tenon 3/4" deep x 3/4 long x 1/4" thick at 3/8" and 7/8" from front of rail. Top roundover and cove for rule joint 384 Table of Contents Notes and Reflections While Shaving Wood 55-11 Appendix. Sliding Dovetail M & T Router bit: ½" W x ½" long 1:6 dovetail router bit. Also, a “U” shaped setup gauge is very useful. TO CUT THE MORTISE 1. Using a straight bit, cut a c" - ¼" wide, ½" deep groove along the centerline of the intended mortise, e.g., center of table leg. Use a stop block to limit the length of the groove. A router dovetail bit is thin at the neck and has to remove a lot of material, which is trapped inside the cut. Also, the cutter is trapped inside the cut. It is better to remove some wood in advance by cutting a groove with a straight bit. 2. Install the DT router bit & set height to ½". Set the fence on the router table so that center of DT router bit lines up with the groove. Reset the stop block. Cut the mortise (tail). 3. Check the mortise. It should be ½" deep & ¼" wide at the narrowest end. TO CUT THE TENON 1. Leave the height of the DT bit as it was when you cut the mortise. 2. Set the fence according to the formula: W/2 - c where W is the width of the workpiece on which you are cutting the tenon. Measure from the front of the cutter back to the fence. For example: Figure 8. Router Table Setup Gauge ¾" wide board . . . . . . set fence to expose ¼" of cutter bit (measured at widest point) e"wide board . . . . . . set fence to expose 3/16" ½" wide board . . . . . . set fence to expose c" These measurements are nominal. In practice, set the fence about 1/32 - 1/16" (.03 - .06") closer, i.e., to remove less wood, thereby making the tenon thicker. 3. Cut the tenon (both sides) If it’s too tight, move the fence another 1/32" (.03") or trim it with a dovetail plane. If its’s too loose, glue a thin strip to one side and cut it again (more carefully) after the glue has cured. References Becksvoort Chris. Understanding wood movement, Fine Woodworking Sep/Oct 2003 ______________. Stop guessing at wood movement, Fine Woodworking Nov/Dec 385 Table of Contents 55-12 2006. Drop Leaf End Table Campbell Mac. Routing a Rule Joint Fine Woodworking Jan/Feb 1990. 386 Table of Contents Notes and Reflections While Shaving Wood Chapter 56 – Workshop Esthetics "We don't have to work pretty to make pretty." Anonymous F tend to trip over it instead. So, I made a rolling tool cabinet out of Sandeply. I made oak drawer fronts for the sake of practice. Nice varnish finish with some woodgrain effects, also for practice. Brass and porcelain knobs salvaged from our kitchen rehab. In my shop, this tool cabinet was a rose among thorns. Then, I made some wall cabinets for the garage, with oak trimmed doors, again to practice making panel doors. I painted the plywood panels white. Pretty sharp. Then another tool cabinet to fit under my grubby workbench. I actually bought knobs to match the other two cabinets. (I think this was the beginning of my slide into workshop esthetics.) The drawer fronts were made out of some left over maple ply. I made 1" edge banding, again for the practice. This made the workbench and the rest of it look even grubbier. Then, my biggest mistake. Now, in my defense before I tell you, my workbench was rough enough to give me splinters, the top was not flat, and gritty debris fell between the boards onto the treasures stored below. So, I re-covered the workbench with ¼" Masonite and varnished the skin with 3 coats of poly. Wow. The shop looked as nice as the house! Well, except for the glue and varnish drips on the floor, the painted CD/X shelves or years, my shop was grubby but happy. My 45 year-old workbench looked like my father's -- dark, chipped, beat up boards. When I first got some power tools, in the 1960's, I made tool tables out of crusty old oak pipe pallets that I picked up along the road in Iowa City when they were laying pipe for a new factory, left over 2 x 4's, boards salvaged from a farm building, and so forth. Shop shelves were sheet metal from Woolworths, K-Mart, etc. I made some shelving out of 5/8" CD/X left over from hurricane shutters. I stored hardware in coffee jars and Edgeworth pipe tobacco cans. (Actually, I still do, although I stopped smoking 35 years ago.) You get the idea. It was not likely to be featured in Fine Woodworking magazine. Life was good. A guy on a woodworking forum expressed my attitude well – "We don't have to work pretty to make pretty." (If he didn't say it that way, he should have.) When we moved into a new house and I got half of the garage for my shop, I promised my wife, Barb, that I would clean up every night before I put everything back onto my side. On my quest to become a better craftsman, I naturally began to acquire tools. As I acquired more tools, I naturally needed more storage space. Also, as I have aged I have become less tolerant of mess. I’m not as good at stepping around it as I once was. I 56-1 Table of Contents 387 56-2 and the remaining pipe pallet tables. But it looked pretty sharp. Even Barb commented favorably. I should have enjoyed it while I could, maybe taken a snapshot. Because then I actually did some work on my pretty workbench and scratched the nice varnish. Now what? I recognized this as a Defining Moment in my woodworking career -- would I recoat the workbench and stay pretty? Hell, no. Back to grubby. We don't have to work pretty to make pretty. But it was nice as long as it lasted. 388 Table of Contents Notes and Reflections While Shaving Wood Chapter 57 – Large Square Coffee Table T his table measures 36 x 36 x 16¾". It is made of solid oak (legs, rails, drawer fronts, and edge banding), oak plywood (top, side panels) and Baltic birch (drawer sides, backs and bottoms). It’s heavy, about 50 pounds with the drawers. The front and rear rails are turned “on edge” for strength, in case somebody should sit or stand on the table. The ¾" furniture grade oak plywood I used for the top had some very shallow ripples that did not show until the finish coats were on. My client and I felt that the top was completely acceptable for its intended use in a family room. Also, I could have added a few more coats of varnish and flattened it, but in this case the spec was that the pores and oak grain show through a black stain. I now have reservations about using plywood for the top of a fine piece. I could use veneered MDF, but that would have made this table even heavier. Mortise and tenon joinery was used throughout the base. The side panels are set in grooves in the top and bottom rails, and the legs. (Wood movement would be a problem if the side panels were made with solid wood.) The legs are tapered from 1 ½" square to 1" square along the bottom 5". The table has four drawers. Drawer kickers attach to the top front and back rails with M&T joints. They are installed flush with the tops of the legs and rails, and also act as stiffeners for the top. Figure 1. Sketch of Table. Top is 36 x 36", 16 ¾" high The center drawer guide support is attached to the bottom front and back rails with M&T joints. The center drawer guides are built up on this support. Base Build the base first. Cut out parts 1-9. (See Cut List) 1. Cut the legs from 6/4 stock, preferably rift-sawn. If you can’t get at least 1½" thickness and width, you may want to cut the mortises a bit shallower than ¾", or you will need to miter or notch the tenons. 2. Cut rails, stiles, center drawer support, drawer guides from ¾" stock. Cut side panels a bit over-size for later trimming to fit. 57-1 Table of Contents 389 Large Square Coffee Table Legs 1. 2. Mark the inside faces of each leg. (See Fig 2.) Taper the bottoms to 1" square. Start the taper cut at 11" from top of leg. If you taper the legs symmetrically, the taper is a rise of ¼" over a run of 5", (an angle of about 2.8E). If you taper just the insides of the legs, the rise is ½" (5.7E). A band saw is best. Lay out each line and then cut the tapers freehand, outside the layout lines. Then smooth the tapered faces with a plane and spokeshave. If you use a table saw to cut the tapers, you will need a taper jig. Be sure not to let the tapers extend beyond where the rails will fit against the legs. (For more on cutting tapered legs, see Popular Woodworking Feb, 2009) 3. 4. MORTISING Cut all mortises first, so that you can lay out tenons to fit them. 1. On faces 1, 2,5, and 6, cut 2, ¼" W mortises, 1½" L x ¾" D. The top mortise starts ½" from the top of the leg, and the bottom mortise starts 8" from the top of the leg. 2. On faces 3, 4, 7, and 8 cut two, ¼" W x ¾ "L x ¾ "D mortises. The top mortise starts ¼" from the top of the leg, and the bottom mortise starts at 9" 5. 57-2 from the top of the leg. Cut a ¼" w i d e groove, ¼" d e e p , between Figure 2. the top and bottom mortises to receive the side panels. For a ¼" reveal on the side rails, the outside edge of the groove should be e" from the outside of the leg. Cut one ¾"L x ¼"W x ¾"D mortise at the middle of each front/back (wide) rail to receive the center stile. The mortises into the rails are problematic because side grain contacts side grain only at the top and bottom of the joint. The other two sides of the joint have long grain on the kickers and drawer guide support against end grain on the rails. You could cut double tenons, one above the other, to increase the long grain glue surface. I just decided to cut wider mortises, even though that reduced shoulder width at the sides, because these joints do not need to add great strength to the structure of the base. They are mainly to hold the kickers and drawer guides in place. Cut two 1½" x ½" x d” D mortises on the top front/back rails at 8" and 24" from the left end to receive the 390 Table of Contents Notes and Reflections While Shaving Wood 6. kickers. Cut one 1" x ½" x d" D mortise in the exact center of the bottom front/back rails to receive the center drawer guide support. TENONS Mark both sides of each joint uniquely, e.g., “LB top”. The mortises and tenons are not interchangeable, at least not when I cut them. 1. 2. 3. 4. 5. 6. 7. Lay out 1½ L x ¼ W x ¾ D tenons on the front/back rails to match the mortises in the legs. The tops of the rails should be flush with the tops of the legs. Lay out ¾" W x ¼ x ¾" D tenons on the ends of the side rails to match the mortises on the legs. Lay out tenons on the ends of the kickers and center drawer guide support. Cut the tenons. Dry assemble legs, rails stiles and center drawer guide support. Mark the position of grooves on top of bottom side rails and bottom of top side rail, lined up with grooves on legs. (Will receive side panels side rails.) Cut ¼ x ¼" grooves in side rails. Check measurements for side panels & trim them to fit. Reassemble with side panels in place. Mark out positions for drawer guides. 57-3 Drawer Guides and Kickers The table has four drawer guides, each made of a ¾ x ¾" horizontal piece glued to a e” (nominally) x 1½"-2" vertical piece. All drawer guide dimensions -especially the thickness of the vertical drawer guides -- should be fitted to the dry-assembled Figure 3. Outside Drawer Guide, Rear Perspective table. The drawer View, shown resting on top guides are of a bottom side rail. screwed in place so that they can be adjusted if necessary. With the base assembled and well clamped, Measure for the drawer guides. Lengths should allow about 1/16" clearance at each end so that the drawer guides cannot impinge on the rails or legs during final glueup. 1. 2. The horizontal drawer guides fit between the front and back rails (31¼"). The outside horizontal guides will be glued to the vertical drawer guides and screwed to the legs. The inside horizontal drawer guides will be glued to the vertical drawer guides. The vertical drawer guides are nominally ¼" thick, but their actual thickness should make them flush with 391 Table of Contents Large Square Coffee Table 3. the drawer opening, or set back no more than 1/16" so that the drawers will not rack while being operated. They are nominally 30½" long, but must fit between the front and back legs. They should protrude at least ½" above the horizontal drawer guides. The inside vertical drawer guides are screwed to the center drawer guide support (Figure 3). The outside vertical drawer. guides fit against the side panel and rest on top of the side rail., or (Figure 4) They are glued to the horizontal drawer guides. Cut the drawer guides and attach the horizontal & vertical parts. Mark each end and each location. Top The top is ¾" oak plywood surrounded with ¾" x ¾" oak edge banding. 1. Cut each piece of edge banding about 37" (i.e., an inch too long) if you intend to miter the corners. 2. Joint the plywood top and the edge banding to make an invisible glue line. 3. The banding can be butt jointed to the edges of the plywood. I have made many drawers with edge banding that was butt jointed to the plywood. Even if the edge catches sometimes when the drawer closes, the joint is strong enough. I decided nonetheless to reinforce the joint with splines. A. Cut a c" wide by ¼" deep groove in each edge band on the router table. 4. 5. 57-4 B. Use a thickness planer to make spline stock that is exactly the right thickness for the grooves, and then cut them about 7/16" wide on the table saw. (If you cut them exactly twice the depth of the grooves, e.g., ½" wide, the glue may add enough width to hold the edge banding away from the plywood.) These were quite difficult to assemble, because the edge banding was not perfectly straight. One went on just fine. One would not go together no matter what I said to it. I eventually ruined the spline and had to quickly plane it off, amidst the gluey mess. For the last two, I used splines at the ends and had to register the banding manually in the center, with a block of wood clamped to the edge, with folded paper spacer to make the edge banding just proud of the top. I pinned some of the edge banding to the top with ¼" dowels to reinforce the butt joint. You can use butt or mitered joints at the ends of the banding I much prefer to miter the ends, but it is much fussier because the miters need to be cut perfectly and, possibly, planed with a shooting board to fit with no gap. Splines make it a bit difficult to slide them together while fitting and measuring. (See Chapter 37, Edge Banding.) Starting at one corner, cut and plane each half of the first mitered joint to 392 Table of Contents Notes and Reflections While Shaving Wood 6. match the other half. When you are satisfied with the fit, mark the opposite end one of the edge bands and cut it to the outside of the mark (at 45E). Then cut the other side of that joint & plane both sides to fit. Work your way around the top in this manner. Do not expect a bunch of 45 E cuts to come out. Round the corners so that they will not hurt someone who bumps into them. If desired, cut a molding along the bottom arris of the edge banding after it has been attached to the top. Assembly • • • Assemble the sides first. Square up perfectly. Apply two coats of shellac to the inside of the side rails, side panels and the underside of the top. Avoid getting shellac on the legs or other surfaces that you will stain. Dry-assemble the base. Install (fit) the guides to the base. Use a straightedge to make sure that the guides line up exactly with the drawer openings. Shave or shim the vertical guides as necessary. Drill screw pilot holes through the outside guide assemblies into the legs. (Do not screw the outside guides to the thin side panels.) Screw one or two ¾ x ¾ x 1" support blocks to the inside of the bottom side rail to reinforce the outside drawer 57-5 Figure 4. Drawer Detail guides. (One is barely visible as a white rectangle under the drawer guide in figure 4.) This large table will take up space in the shop after it is assembled. (It is also heavy.) It is possible to fit the drawers to the dry-assembled base, or the base can be glued up. Measure for the drawers. Each may be different, so mark each one clearly. Drawers The drawer sides attach to the front with two half-blind dovetails and to the back with finger joints. 1. You will want the fingers on the box joints to protrude 1/32"initially, so that they can be planed or sanded flush with the sides. But you may also want to recess the dovetails into the front by 1/32" in case the drawer needs to be trimmed. These cancel each other out, so you can cut the back the same length as the front. Note that the back is narrower than the front or 393 Table of Contents Large Square Coffee Table sides so that the bottom can slide underneath it. 2. Cut the drawer parts to fit the opening after the frame has been assembled. If each opening is perfectly square (as it should be), you can cut the front exactly the size of the opening & then trim it all around to get the desired reveal. If the opening is not perfectly square, cut the drawer front oversize so that you can trim it to fit. 3. Cut ¼" deep grooves along the inside bottom edge of the front and sides to accept the drawer bottom. Note that a dovetail (on the side) should ideally cover the groove. A ¼" wide half pin seems strong enough in oak. If you want a wider half pin at the bottom, the groove should either stop ¼" from each end of the front or should be cut far enough from the bottom edge to allow space for a thicker half pin, e.g., d" from the bottom edge. See Chapter 39, Boxes and Drawers for construction details. The drawer back is attached to the sides with finger joints. It is ½" shallower than the front and sides. Starting at the bottom edge, lay out the finger joints so that the drawer sides start with a finger at the bottom, and the back also starts with a finger at the bottom. The ¼" thick drawer bottom slides under the back and into a ¼" groove in the sides and a ¼" dado in the back of the drawer front. If it is plywood or Masonite it should be glued in. 4. 57-6 Assemble the drawers on a dead flat surface so that it s squared up in three dimensions, i.e,. the bottom will sit flat on the guides with no rocking. When fitting a drawers that is a little out of plane, with close-fitting lower guides and kickers, remember that the drawer is three-dimensional. One corner of the front may be high because the opposite corner of the back is low, etc. Plane, sand and shim the bottoms of the sides toward the back to make the drawers fit correctly while preserving the desired reveal at the front. Spacer blocks attached to the rear of the drawer guides, after final assembly, hold the drawer front in position when closed (¼" back from the front edge of the rail). Finish Sand plain sawn oak with a block. Otherwise, you will wear away the early wood, causing ripples on the surface. Oak, especially plain sawn, is difficult to stain evenly with oil (pigment) stain because the early wood darkens more and the dense late wood darkens much less. This gives it an uneven look, especially unattractive on large panels of plain sawn oak veneer plywood such as the top of this table. It is better to use a wash coat of varnish or sanding sealer diluted 1:3 in thinner and then to apply wood dye or oil stain. It is wise to finish the underside of a 394 Table of Contents Notes and Reflections While Shaving Wood 57-7 of the drawers with shellac. After the finish hardens, rub it with 0000 steel wool & apply paste wax to even out the gloss. The specification for this table was a jet black finish that showed the grain and pores. • Figure 5. The Completed Table large table top so that it will be less likely to warp with changes in humidity. If you have not already done so before assembly (see above) you should apply two coats of shellac the underside of the top and the insides of the side panels. Use shellac on the interior because varnish will continue to smell for weeks after the piece is placed into use. For the same reason, finish the interior • • I used NGR black dye directly on the wood (no washcoat), followed by Minwax ebony oil stain. Dye spatter was a problem when I was applying dye to the drawer fronts. I should have sealed the drawer sides with sanding sealer or dilute varnish before I dyed the fronts, or dyed the fronts before I assembled the drawers. (I don’t like to use varnish on cabinet interiors, but the dye spatters would have penetrated shellac.) I carefully “painted” the dovetail pins on the drawer fronts with thinned black oil-based enamel paint & finished the rest of the drawer fronts as above. I sealed the Minwax with one coat of dewaxed shellac and finished with two coats of polyurethane 395 Table of Contents Large Square Coffee Table 57-8 Cut List 1 Legs 16 x 1½ x 1½" Taper to 1 x 1 starting 11" from the top (wide) end. Cut four mortises on inside faces (see narrative). Cut ¼" W x ¼" D groove between mortises 4 2 Front/Back rails 32 x 2 ½ x ¾ OA Cut tenons on ends (to match with mortises on legs.) Cut centered mortise on the bottom of the top rail and the top of the bottom rail. Cut two mortises on inside of in top rail (for kickers). 4 396 Table of Contents Notes and Reflections While Shaving Wood 57-9 3. Side Rails 32 x 1½ x ¾" OA Cut ¼" thick tenons on ends, ¾W x ¾ L (to match with mortises on legs.) Cut ¼" W groove for side panels. 4 4 Front & Back center stile 6 1/2 x 1-1/2 x ¾" OA Cut ¼" tenons on ends, ¾W x ¾ L 2 5. Drawer guide vertical 31¼ x 1½ x ½ 4 6. Drawer guide horizontal 31¼ x ¾" x ¾" 4 7. Kickers 32 x 2½ x ¾" Cut tenons on ends 1½ x ¼ x d 2 8. Center Drawer support 32 x 2½ x ¾" Cut tenons on ends 1½ x ¼ x d 1 9. Side Panels 31 x 7 d x ¼" OA 2 Drawers (4) Nominally 5 c H x 14 ½ W x 16" D 10 . Drawer Front 5 c x 14 e x ¾ Cut to fit actual opening 11 . Drawer Back 4 e x 14½ x ½ secondary wood 12 Drawer sides 5 c x 15 e x ½ Length depends on thickness of front. Overall length of drawer = 16" 13 . Drawer Bottom 15 ½ L x 14 W x ¼" oak plywood 14 Top 34 ½" x 34 ½" ¾" Oak Plywood (See discussion) 1 15 Edge Banding 36 x ¾ x ¾" Oak (select for grain) Miter ends 4 Top 397 Table of Contents 57-10 398 Table of Contents Notes and Reflections While Shaving Wood Chapter 58 – Getting Realistic (Life is Too Short to Cut Corners) F table saw using the miter gauge. I checked the drill press for runout. And so forth for almost every tool. I was surprised to discover that my tools were not actually set up as well as I had assumed. The jointer tables were not quite coplanar, and every pass distorted the board a tiny bit more. The miter gauge was not cutting at 90E. The drill press was cutting concentric circles because of runout. No wonder some holes I drilled were too big. My frustration had often been caused not by poor workmanship, as I had thought, but by my failure to set up my tools correctly. I had to learn how to set up my tools, then set them up, then check the setups. I admit, this took considerable time at first. But now I just have to maintain them. Appreciation of the need for precise adjustment has changed how I do some operations. For example, it takes many tries to get the miter gauge on my table saw to exactly 90E. Once it is adjusted, I hardly ever move it. (See Chapters 10, 16, and 18.) Maybe you always knew this, learned it in shop class or someplace. I didn’t. I wondered why this was. Part of the reason was a cowboy attitude, part was feeling that I did not have time, and part was my role models. “Norm,” i.e., the typical presenter on the typical TV woodworking program, shows us only the seconds required for the cut, or perhaps only the results. No setup, sharpening or many years, I lived in an imaginary world where tools came ready to use unless they looked wrong or the owner’s manual told me otherwise. I believed that detailed plans were a waste of time. Just sketch it and create it. Let the work tell me what the next step should be. I allowed a weekend for a project that actually should need a week. I was inspired by the pros on TV who could work so fast. Design plans, sharpening, or adjusting, tools take time. Most often, however, when I skipped these steps, it cost me more time, effort or whatever it was that I thought I would save. Sometimes, rework was not possible and I lived with an unsatisfactory result. I gradually figured out that this was a problem because, at the end of many projects, I wished I could make them again, now that I knew how I should have done it. Furthermore, being in a hurry lengthened the time I needed to improve my craft. I had to change my attitude in order to improve my woodworking. I now believe that I don’t have time to cut corners or to be in a hurry. Now I do take time to sharpen and adjust my tools, to plan my work, and to do it right the first time (well, I try to do it right the first time). I realize that this sounds a bit like a sermon, but I wish somebody had convinced me of this sooner. I was not a complete klutz. My tools were sorta adjusted, “close enough” I thought. When I eventually got frustrated enough, I laid a straightedge along my jointer table and really sighted under it. I did a five-cut on my 58-1 Table of Contents 399 58-2 Getting Realistic or test cuts.a I thought I understood that TV shows like this left out the “boring” parts, but I did not appreciate that they omitted the great majority of the work required to produce a cut. “Norm’s” cheerful, casual competence truly did mislead me for a while. His show concealed so much that I was deceived despite myself. So when I tried to emulate what “Norm” did, my board did not come out as well as his did. I should have known that “edu-tainment” is 3 parts edu and 8 parts tainment. Of course, stock preparation and tool setups make boring TV. Most people would be dismayed by the setup time required for a simple weekend project. Perhaps they would change the channel and not buy the tool that he is demonstrating. Some beginning woodworkers (like me) might think that they lack some skill that “Norm” has. That may be true, but for most tools, the essential skill is proper preparation. A personally meaningful example for me is a shop-made box joint jig. The jig is very simple, basically just a square peg glued into a wooden extension to the miter gauge on a table saw or router. I made such a jig but the joints I made with it did not fit together. My son, Drew, made one with the same result. I couldn't see why my jig did not produce the intended result. After a fair amount of research and thinking I recognized that tolerances of this jig have to be within thousandths of an inch. This necessity is obvious once it's pointed out: each socket is measured from the one you just cut, and any error adds progressively along the joint. The jig and its position relative to the cutter has to be made far more precisely than I had ever attempted. This is not a secret, but the enthusiastic stories in woodworking magazines seemed to omit the necessity for high precision and the ability to tweak the jig. (Or maybe I just wasn't listening.) So, now I make time to prepare my tools. I take the time to get (or make) detailed plans. It's true that I sometimes feel like the world's slowest woodworker, but I make up for it because I don't have as much rework as before. I can do more difficult operations now, so the furniture I make is a bit more challenging. I get pleasure when the tool cuts where I intended and the measurement of the part is right on. The final payoff comes at assembly time when (if) each piece fits and the assembled piece goes together cleanly, tight, and straight. a I’m not referring specifically to Norm Abrams or The New Yankee Workshop. I’m using his first name in a generic sense because he was the archetype and because his show is still so very popular in re-runs. My point is about the impression that these shows may create in naive beginners like I was. 400 Table of Contents Notes and Reflections While Shaving Wood Chapter 59 – Dining Table With Drawer T his is an informal dining table with turned legs, solid wood aprons and a drawer. (Figure 1) I have made more than one version of this table. The inclusion of a drawer in a kitchen table is a nice convenience, but requires wide aprons. They could be made out of plywood, which would be more stable dimensionally. I prefer solid wood because of the drawer opening. If the aprons are made out of solid wood, their width can expand and contract with changes in moisture content. (Aprons of 4" wide plain-sawn oak theoretically could expand or contract by c".) I made this table in a humid environment. If the aprons were attached to the legs with 3½" tenons, glued along the width of the apron, wood shrinkage in a dry environment might split the aprons. Or so they say. (See Chapter 8) Typically, wide solid wood aprons are attached to the legs with forked tenons, with the top tenon glued and bottom tenon pinned rather than glued. (See the article by Will Neptune.) Another solution to this problem is to attach the legs to the aprons with sliding dovetails, glued only at the top 2" or so. I used this idea in the drop-leaf end table (Chapter 54.) The dovetails must fit very well, just barely loose enough to slide together; they also can be wedged slightly with shims to make the joint mechanically tight. I wanted the table to be strong enough to take the weight of a person sitting on the long edge, so I placed the drawer on the end of the table. Also, a drawer at the end allows the addition of one or more drop leaves. Figure 1. Dropleaf Dining Table Here are the specifics for this particular table. • Height should be 30", com• • • • 59-1 Table of Contents prising legs 29" long and a f" - 1" thick top. Aprons (rails) 4" wide. This leaves leg room of 25" (the minimum in my opinion). Drawer 14 W x 18 D x 2½" H. A 2½" deep drawer is the maximum possible with a ¾" top and bottom rails in a 4" apron. This should hold napkins & placemats or a shallow flatware tray. Place the drawer in the center of the short side. Add two 1c x ¾" top & bottom doubler rails. The sliding dovetails are 3¾" long, and the sockets are 3f" long. Slightly taper the dovetails to slide home easily. Dovetails are glued at top 2" (no more) to allow for wood expansion/contraction. Drop leaves 12" wide, with rule joints 401 59-2 Dining Table Legs Mark the tops of the legs so that you can keep track of their relative position. The legs for this table are 29" long; 2½" square x 6" long at the top. The turned portion is 23" long, and tapers from about 2" to 1½". To turn the tapered legs, calculate the required diameter at 1-3 points in Figure 2. Mark the between the top and Tops of the Table bottom of the Legs tapered portion. Begin by cutting those points to the desired depth with a parting tool on the lathe. Use calipers to check. Then use a gouge and skew to turn the leg to those depths. Hold a straight piece of wood along the leg as you turn it so you can easily find bulges or hollows in the taper. Then, sand the turned legs smooth. I had trouble getting the tapers straight enough. I turned the legs to nearly the desired taper, and then turned off the lathe and used a concave spokeshave to straighten out and smooth the legs. That produced nice straight, smooth sides. Then I turned the lathe back on and finish sanded the legs to remove any ridges left by the spokeshave. Make a template for any beads, etc. and turn those. Cut the dovetail (DT) mortises on a table-mounted router with a fence. Before cutting the 3 f" long DT mortises in the two inside faces, use a ¼" straight bit to cut a straight groove in one or two passes. This will lessen the load on the DT bit. Finish the legs with a slight (¼") roundover along the outside corner of the square portion. Aprons DRAWER OPENING Before you cut out a piece for the front apron, decide how you want to cut the opening for the drawer. For the most attractive figure, the drawer opening can be cut into the front apron or you can make up the apron by gluing up two rails and two side pieces cut from the same board. To use the first method, lay out the opening on the front (show) side of the apron. Drill two d - ½" starter holes at opposite corners. Drill from front to back so that any tearout will be on the back of the apron. Each starter hole should touch two sides of the corner (i.e., the two sides of a corner should be tangent to the starter hole). Cut out the opening with a keyhole saw or straight router bit. Set the router fence so that a straight ¼"-d" router bit will cut along the length of the layout line. It is best to use a fingerboard to hold the apron snug against the fence. Set a stop block to stop the cutting at the other end of the opening. 1. 2. Lower the cutter to about ¼" or less above the table. Hold the work down firmly with a push shoe or small parts holder and 402 Table of Contents Notes and Reflections While Shaving Wood turn on the router. Make the first pass. A push shoe is necessary to keep your Figure 3. Cutting Drawer Opening on a Router Table fingers away from the workpiece. 3. Raise the bit ¼" and make the second pass. 4. Raise the bit to full depth and make the final cut. 5. Flip the board end for end and repeat for the opposite long side. Note that this is potentially a dangerous operation. In a router table, the router bit turns counter-clockwise, i.e., the cutter closest to the fence is moving from right to left, in the same direction as you are feeding the stock. This would tend to grab the workpiece and throw it to the left, if it were not for the cutter on the opposite side of the bit, which is also cutting wood and tending to push the work in the opposite direction, against the feed direction. (See Figure 3.) If anything reduces this opposing force, e.g., punk wood or holes, or increases the climbing force, e.g., a knot, you may lose control of the workpiece. If the board flies away, you may be unable to keep your fingers away from the cutter. Use fingerboards and a push shoe. I use a push shoe with a hold down that keeps the work from lifting off the table. By the same token, you cannot just go 59-3 back and trim the opening after completing the cut. There would be no countervailing force and you would be climb cutting. You might be able to feed the work from left to right, or take a very light cut. Otherwise, you risk injury. The work is likely to be ruined and may even fly off the table. Finally, cutting the short sides of the drawer opening with a table router probably Figure 4. Grain Matched Apron and Drawer Front cannot be done safely because the short end of the apron does not give enough stability against the fence (even if the fence can be moved back far enough). Cut the short sides with a coping saw and finish up with a chisel and wood rasp. Remember to work from the show side so that any tearout will be on the back side. The second method, shown in Figure 4, will yield a drawer front that is cut from the same board as the rest of the apron, and fits into almost the same position. This will allow the best grain matching. The center piece, which would be used for the flush drawer front, might be a bit too small for the opening because of the wood lost to saw kerf. You can compensate for this by cutting the stock a bit oversize. Then you can trim off the side pieces equivalent to the width of saw kerf. 403 Table of Contents 59-4 Dining Table DOVETAILS Figure 5. Routing Dovetail Tenons on a Router Table. Note that the workpiece is not trapped between the cutter and the fence. (Modified from a drawing by Melanie Powell in Fine Woodworking) The height of the sliding dovetails (tenons) should be slightly (e.g., 1/128") less than the depth of the sockets (ways) to allow room for glue. The tenons can be cut with a hand-held router using a guide, but I prefer the tablemounted approach. I use an L-shaped push shoe behind and outside of the work, and a clamp riding along the top of the router fence to hold the apron vertical and square while I am feeding it across the cutter. (See Figure 5.) The fit of the DT tenons has to be “perfect” (accurate within less than 1/128") or the mechanical rigidity of the joint will be compromised. Any tiny slop in the joint will allow noticeable wobble in the leg. This fussiness is the major disadvantage of this design. On the other hand, this joint is mechanically very sound. With glue at the top 2" or so the joint should still be stronger and longer lasting than a mortise and tenon. If the glue fails someday, the top can be removed, and the joint can be shimmed without the need for disassembly. In any case, the joint should still hold the leg in place. Unfortunately, a perfectly fitting DT tenon may not slide into the ways after glue has been applied. Therefore, either the ways or the tenon should be slightly tapered. When cutting these joints with a router, it is easier to taper the tenon. This requires very cautious wood removal with a dovetail plane, wood rasp, or sanding block. It is all too easy to take too much off of the wide end of the tenon accidentally with a flat sanding block or wood rasp. The DT plane or specially constructed sanding block is easier and safer to use. To make the sanding block, take a piece of oak about 1" x 2" x 6" and run it by the DT bit, cutting the DT profile into the bottom half. Then stick a piece of sandpaper to the top of the cutout. (Figure 6). While the DT mortises in the legs are cut precisely to the dimensions of the DT router bit, the tenons have to be fitted to a precise thickness. This is inherently subject to error, so precise layout, router setup and repeated fitting is necessary. Cut the tenons a bit wide and plane or sand them until they fit perfectly. 404 Table of Contents Notes and Reflections While Shaving Wood 59-5 FINAL MILLING If you will attach the top with sliding metal clips or sliding wood blocks, cut short grooves in the inside of the top of the aprons to receive the clips. Finish the aprons with a bead, roundover or chamfer along the bottom outside edge. Drawer Guides Figure 6. Sliding Dovetail Sanding Block Router Setup It is important that the workpiece (apron) run on the side of the cutter away from the fence. If it is trapped between the cutter and the fence and pushed from right to left this will be a climb cut and the work may be ruined as the cutter grabs it out of your control. Also, you may be injured if (when!) this happens. Suppose the apron stock is ¾" thick, and the DT mortise is ½" wide. Then the widest part of the tenon should be c" back from the plane of the apron surface. A 14E cutter would be 9/32" wide at the narrow end, so the fence should be set about 15/64" from the widest part of the cutter: c" + ½ (½ - 9/32) = c" + 7/64 = 15/64" In my experience router calculations are often approximations to the actual result, so test cuts are essential. Also, it may be necessary to cut each side in more than one pass. Do not force the DT cutter: it can break off. If you get tearout, lightly score the line across the apron where the tenon shoulder will be. The top doubler reinforces the apron across the span of the drawer. It is glued and screwed to the apron. For additional strength it could fit into a notch cut in the tops of the legs Make sure that it does not protrude into the drawer opening or above the top of the apron. The bottom doubler is just glued and screwed to the apron. The drawer guides are made up of two pieces of wood glued together to form the horizontal and vertical parts of the guide. The guides and kickers “hang” on the doublers at the front with half-lap joints and on two cross supports at the back. The top cross support is mortised into the top of the apron with a sliding mortise. The bottom is joined to the apron with a half lap joint. The parts for the drawer guide assembly must be fitted to the piece, in particular the drawer opening. Also, it is best to dry assemble the legs and aprons to mark the position of the mortise and cutout for the rear supports on the long table aprons. 405 Table of Contents 59-6 Dining Table Drawer To see what the drawer opening might look like with an inset drawer, try a squared piece of wood inside the opening. If the sides of the opening are not perfectly straight, square, and smooth, a flush inset drawer may be too difficult to fit well. In that case it may be better to use a lipped drawer front. The lip should be 3/16 - ¼" wide, on the top and the sides, or all the way around. It should be about ¼" thick if it is made of hardwood. You will need at least ½" for dovetails. Therefore, the minimum thickness for a lipped drawer front is ¾". I used a 1" thick piece with a d" lip and f" for the joint. I wanted to join the drawer front to the sides with half-blind dovetails, but the lip would interfere with cutting these by hand. I decided to cut the pins (mortises) with a DT router bit on the router table. It is better to rabbet for the lip first, because cutting the rabbet for the lip will tear out the sides of the pins, especially if you cut the rabbet with the router. Then I cut the tails by hand (actually, laid them out, cut them with a bandsaw, and finished them with a chisel.) after trimming, allow at least ¼" extra width per board for jointing. You will trim the top to width and length after assembly. 1. Cut the boards to length plus an inch or two. 2. Glue up the panel as described in Chapter 35. Remember to mark the grain direction so that you will know which way to plane after the top has been glued up. 3. Trim the top and cut the edge treatment. I used a ¼" radius roundover for the ends and ½" radius for the sides as part of the rule joint. You can also make a drop leaf with square edges. 4. Smooth the top. Examine it carefully for imperfections and correct them. DROP LEAVES Drop leaves are optional, of course. If you add drop leaves, see Chapter 54 for a description of rule joints and hinges. Of course, you can make drop leaves with square edges, but you still should consider using drop leaf hinges because they have unequal leaves. Since the underside of the table is taken up with the drawer assembly, use drop Top The width of the top should include an overhang along the sides. The amount of overhang should allow for hinged drop leaf supports, e.g., about 3". When planning the glue-up of the top, please see Chapter 35, Glued Up Panels. Examine the boards carefully. Decide what, if any, parts need to be trimmed away. Even Figure 7. Hinged Drop Leaf Support 406 Table of Contents Notes and Reflections While Shaving Wood leaf supports that are fastened to the outside of the aprons with heavy hinges. (See Figure 7.) Drop leaves of this length (48") should have two supports per leaf, especially if the supports are hinged to the aprons. To adjust the height of the raised drop leaf, glue a thin wedge and a stop to the drop leaf corresponding to the full 90E swing of the supports. Regarding the width of the drop leaves, consider whether you want to store chairs under the sides of the table when the drop leaves are down. A chair seat is typically 16-18" off the floor. The table top is 30" off the floor, so you may be able to accommodate chairs under the drop leaves by making a small adjustment in the width of the drop leaf. Assembly Assembly is fairly straightforward. Apply adhesive (slow-set epoxy or Titebond III to the top 2" of the sliding dovetail tenons and slide them into the groove. Shim the angled part of the dovetail tenons if necessary to draw them tight (Don’t try to tighten the joint by shimming the flat shoulder of the tenon because this will push the joint apart as it tightens it.) Assemble the drawer guide frame in place and fit it to the drawer. If you have not yet done so, cut 59-7 grooves in the end aprons and the rear kicker support to receive the clips for mounting the top. A top of this size must be able to expand and contract across its width. However you attach the top to the side aprons, allow at least c" for contraction/expansion. (I just used the same clips on the side aprons as I did on the ends.) Attach the top. Finish Finishing is, of course, a matter of taste. I prefer smooth, protective finishes, especially on a dining table, desk, etc that may hold wet glasses and spills. Consider using a “grain filler,” such as Bartley’s Paste Wood Filler, on the top if it is oak. Test on a piece of smoothed scrap. Colored paste wood filler may show up some tearouts as ugly dark blotches. Consider washcoating cherry or maple with 1# cut dewaxed shellac. Getting the right color and matching the top to the base, may require some toning. Tone the parts as needed with diluted Polyshades or your own tinted varnish. A Finisher’s Color Wheel is very helpful for selecting which UTC pigments are needed. Use very little UTC (less than ¼ tsp, in 6 oz of varnish). The mixture will look cloudy in the can but will brush on transparent. 407 Table of Contents 59-8 Dining Table Cut List for 48 x 33 Top With Drop leaf 1. Rear Leg Front Leg 2. Rear Apron 3. Front Apron 4. Side Aprons 5. Primary (1E) hardwood (oak or maple) 2½" square x 29" L 1E, 24" ovl x 4" W x ¾" turn bottom 23" from 2" to 1 ½". Cut 3f" long 14E (¾" x ¾") dovetails, centered in top portion, on two faces As above, but notch top of front legs about 1" x ½" x ¾" deep to accept top doubler 2 2 dovetails 3 ¾" long centered 1 cut 14" x 2 ½" rectangular opening centered (4" from each end & ¾" from top & bottom edge) 1 1E, 41¼" ovl x 4" W x ¾" dovetails 3 ¾" long centered. Cut ¼" x 1" x ½" D mortise 17 1/16" from front end of apron (incl DT) Top Doubler Secondary (2E) hardwood, 2" x ¾ x 22¼ L OVL distance between legs is 22½" Cut doubler longer if you want to fit it into a notch in the top of the leg 6. Bottom Doubler 2" x ¾ x 22 ½ L OVL 7. Vert. Drawer guide 2E hardwood 1¾ x ¾ x 20 1 c x ¾" notch cut in each end 2 8 Horiz Drawer Guide 2E hardwood 1 ½ x ¾ x 18e OVL ¼ W x 1"L x 1/2"D tenons cut in ends 2 9 Rear Drawer Support 2E hardwood 1c” x ¾" x 25¾ ¼ W x 1"L x 1/2"D tenons cut in ends 1 1E hardwood 48 x 33 Make up from planks. Size is apron length + leg width - f + overhang, e.g., 24 +2½ - f + 1½ 1 10 Top Drop leaf 1 1 1E hardwood 48 x 12 408 Table of Contents Notes and Reflections While Shaving Wood 59-9 2 11 Drawer Kicker 18 c x 1 c x 3/4 12 Rear Kicker Support 25 ¾ x 1 c x 3/4 13 nominal dimensions. Half-blind DT front, box joints at rear. Groove for bottom. 1 Drawer Front 1E hardwood 2 ½ x 14 x ¾ 14 Drawer sides 2 ½ x 18e x ½" BB 2 15 Drawer back 2 ½ x 14 x ½" BB 1 Drawer Bottom 13½ x 18½ x ¼" plywood ot masonite 1 16 1 409 Table of Contents 59-10 Dining Table 410 Table of Contents Notes and Reflections While Shaving Wood 59-11 References Taunton Press p. 150-157. Lonnie Bird, Drawer Construction, Fine Woodworking, Nov/Dec 2003 (Tips on making sliding dovetails) Gary Rogowski, Gluing Up Tabletops, Fine Woodworking Nov/Dec 2003. Will Neptune, Engineering a Table With Drawers, in Practical Furniture Design Pat Warner, Sliding Dovetails http://www.dewalt.com/us/articles/printable. asp?ArticleID=511 411 Table of Contents 412 Table of Contents Notes and Reflections While Shaving Wood Chapter 60 – Shop Rules YOU CAN LEARN A LOT IN THIS SHOP, COMMON SENSE AND COMMON UNLESS OF COURSE YOU ALREADY KNOW EVERYTHING. KNOWLEDGE ARE TWO OF THE LEAST COMMON THINGS IN THE WORLD. OUR WORK IS QUICK, CHEAP AND TOP QUALITY. CHOOSE ANY TWO. “WHY?” IS OFTEN A SNEAKY WAY TO SAY “NO.” ALWAYS RECOGNIZE THE DIFFERENCE DON’T ASK “WHY?” MANY PEOPLE DON’T KNOW WHY, BUT WILL MAKE SOMETHING UP. BETWEEN BENDING OVER BACKWARD AND BENDING OVER. (A.R. Gingras) THE FACT THAT ALL YOUR COMPETITORS WE ARE NOT SO MUCH WHAT WE HAVE LEARNED, AS WHAT WE HAVE LEARNED TO LOVE. ARE DOING IT DOES NOT MAKE IT ETHICAL. IF YOU CAN BE INTIMIDATED, SOMEBODY WILL EVENTUALLY FIND OUT. IT’S WHAT YOU DO THEN THAT MATTERS. CHECKS ARE OFTEN LOST IN THE MAIL BUT BILLS ALWAYS ARRIVE ON TIME. MOST PEOPLE WOULD RATHER LIVE WITH IF YOU ARE NOT SURE, FIND OUT. IF IT MAKES YOU NERVOUS, TURN OFF THE TOOL. A PROBLEM THEY CANNOT SOLVE THAN ACCEPT A SOLUTION FROM SOMEONE THEY DISLIKE. NOTHING WORKS EVERY TIME. THE HARDEST QUESTIONS ARE THE ONES YOU ASK BEFORE YOU ARE READY FOR THE ANSWER. IF A THING IS NOT WORTH DOING, IT’S NOT WORTH DOING WELL. (Sidney Harris) PROBLEMS ARE EASIER TO SOLVE THAN CRISES. DO NOT PUT OFF UNTIL TOMORROW THAT WHICH YOU CAN PUT OFF INDEFINITELY. IF PEOPLE HAD BEHAVED THEMSELVES, KNOWLEDGE IS NOT A SUBSTITUTE FOR THINKING. COMMUNISM WOULD NOT HAVE BEEN NECESSARY. (Jenny Holtzer) IN WOODWORKING, WHAT YOU DON’T 60-1 Table of Contents 413 60-2 Shop Rules KNOW CAN HURT YOU. WE ARE EACH ENTITLED TO OUR OWN OPINIONS, BUT WE ARE NOT ENTITLED TO OUR OWN FACTS. (Daniel Patrick Moynihan) FACTS WITHOUT THEORY ARE TRIVIA. THEORY WITHOUT FACTS IS BULLSHIT. Beiser’s Brass Tack: FURNITURE IS THE MARRIAGE OF FORM AND FUNCTION. FORM WITHOUT FUNCTION IS ART. DON’T ASK THE CLIENT WHAT HE THINKS OF THE PIECE. HE MAY TELL YOU. FLATTERY IS THE CHINESE FOOD OF SELF ESTEEM: HALF AN HOUR LATER, YOU ARE HUNGRY AGAIN. INTERVAL OF BLISSFUL IGNORANCE (IBI): THE TIME BETWEEN THROWING SOMETHING AWAY AND NEEDING IT. NORMALLY, THIS IS TWO WEEKS (BELL’S RULE) BUT CAN LAST A YEAR OR MORE (IN WHICH CASE YOU WILL SPEND A DAY LOOKING FOR THE THING YOU FORGOT YOU THREW AWAY.) “BEFORE YOU CRITICIZE SOMEONE, YOU SHOULD WALK A MILE IN THEIR SHOES. THAT WAY WHEN YOU CRITICIZE THEM, YOU ARE A MILE AWAY AND YOU HAVE THEIR SHOES.” Jack Handey 414 Table of Contents Notes and Reflections While Shaving Wood Chapter 61 – Small Writing Desk T his project is a bit more advanced than the bedside table in Chapter 52, and the directions are not quite as detailed. This a writing desk for a pre-teen child or small adult. It has the usual table height of 30" but, with the drawers, the leg space is 23¾" high, 2-3" less than recommended for an adult. A man sitting on a seat 16" above the floor might not be able to get his legs under this desk comfortably. This is easily modified by making the legs longer. Here are the specifics for this particular desk. • The floor area required for this desk is 23" x 45" Height should be 30", comprising legs 29" long and a ¾"- f" thick top. Top 23 x 40 x ¾" thick, leaving an overhang: 1" front and sides Three drawers Raised panel or plywood side and rear aprons. Gallery or bookshelf • • • • • Figure 1. Small Writing Desk Legs the inside surfaces of the legs, ¾" long x ¾" wide x d" thick The side and rear panels fit into grooves in the legs, between the mortises for the rails. Cut ¼" x ¼"grooves for the side and rear panels The legs for this desk are 29" long; 1¾" square x 8" long at the top, tapering on two sides along the bottom 20" to 1¼". 1. Cut and smooth leg blanks from straight-grained 6/4 or 8/4 stock (rift cut would be ideal. ) 2. Mortise the legs to receive tenons on the ends of the rails. (See Figure 5.) These mortises are centered across the 3. 61-1 Table of Contents between the mortises. (See Figure 5.) Starting about 8" from the top of the leg, taper each leg using a taper jig on a table saw and finish up with jointer or handplane. (Plane downhill to avoid 415 61-2 4. Small Writing Desk tearout.) Add decorative c" deep groove around the leg with a “signmaking” router bit support for the drawer guides and kickers. A secondary wood like poplar is strong enough.. 1. 2. Rails The front of the desk is made up of two horizontal rails, two doublers screwed and glued to the rails, and two center dividers. This makes openings for three drawers. The rails, dividers, and drawer fronts can be cut from a single piece of stock. (Figure 2) Then, except for the width of saw kerfs and any wood removed in the final fitting process, all of the components fit into the same position as in the original board. This will allow the best grain matching. The width of the stock for the front must be equal to or greater than the width of the two rails plus the width of the drawer front plus twice the width of the saw kerf, plus an allowance for final fitting. The length of the board must be equal to the length of the rails (including tenons) plus four saw kerfs, e.g., ½" or so. The doublers are intended to strengthen the front rails and to provide 3. 4. 5. 6. 7. Cut the side and back rails. Calculate the horizontal dimension for the dividers and drawer fronts. From the board chosen for the front components, cut the top rail, then the drawer front/divider blank, then the bottom rail. Cut the drawer front/divider blank about 1/16" too wide so that it can be trimmed later. Mark each piece as it is cut. Cut the front rails to final length. Cut the tenons on all rails ¾ x ¾ x d thick, (See step 2 under Legs, above.) Fit the front rails into the front legs and trim the center dividers to fit snugly. (Remember that the grain runs horizontally). Clamp them exactly in place and drill d" diameter holes through the top and bottom rails and into the dividers. The holes are to accept dowels to reinforce the butt joints between the rails and dividers.. Cut doublers for front and rear rails (top and bottom). Cut them a hair thinner than the rail so that they won’t interfere with the top or the drawers. Lay out and cut notches for drawer kickers on front and rear top doublers Side and Rear Panels 1. Figure 2. Built-up “Apron” Assemble (dry-fit) side rails and rear rails and measure for panels (nominally 4½" wide x 19½" long). If 416 Table of Contents Notes and Reflections While Shaving Wood 2. 61-3 you use hardwood panels, leave c" space for expansion of the panel width. Cut and mill panels to fit into grooves in rails. (I cut c" wide rabbets on both sides of the ½" thick plywood panels, leaving a ¼ x ¼" tongue to fit into the grooves in the rails and legs.) Shelf (Gallery) Join the shelf supports (sides) of the shelf to the shelf with keyed miter joints, also called mock finger joints. 1. Cut the sides of the shelf and the shelf itself from the same board. 2. Cut out a decorative opening in the sides to make “legs” 3. Cut 45E miter joints using a jig on the table saw. (A jig is more accurate than tilting the arbor. If you do tilt the saw arbor to 45 E check it well by cutting scrap or the joint may not assemble to a 90 E joint.) 4. After the glue has cured on the miter joints, use a cradle to hold the assembled shelf at 45 E to the saw table. Cut evenly spaced saw kerfs about f" deep across the width of the miter joint (e.g., 8 cuts each 1" apart). (This assumes that you use ¾" thick stock for the shelf and sides. The face of the miter across ¾" stock is a bit more than 1" long. If you are using stock of a different thickness, adjust the depth of cut accordingly.) 5. Cut maple keys to fit the kerf (1" W x Figure 3. Dovetail Layout for Drawer 1½” L x kerf thickness) & glue them in. When the glue has cured, plane them flush with the shelf and the sides. Drawers Flush drawers would take full advantage of the matched grain on the front. Fit the drawer fronts to the openings, leaving about 1/32" reveal all around. (This may be too tight and you may wish to trim this when you finally fit the finished drawers.) Then make the drawers to fit the openings. I chose to join the drawer fronts to the sides with half-blind dovetails (DT). I cut the pins (mortises) with a ½" 14E dovetail router bit on the router table. The pins were ½" wide, laid out symmetrically with the first pin ¾" from the edge and ½" between pins. Therefore, each pin was 1" from the adjacent one(s). (Figure 3.) This gave a balanced look to the drawer sides and allowed me to use 1" wide spacers along the fence of the router table. Before cutting the pinboard and tails, cut the grooves for the drawer bottom in the 417 Table of Contents 61-4 Small Writing Desk sides, front and back (¼" x ¼" groove). Space them ¼" up from the bottom edge. This should be done to mark the bottom inside edge of each piece and to simplify aligning the pieces while marking out the joints. 1. Set a stop block so that the DT bit will cut the mortise ½" deep. This matched the thickness of the drawer sides, plus a 1/32" inset in case I needed to trim the drawer front for final fitting. 2. Set the fence at 2¾" for the “third” mortise, cut it, remove one 1" spacer, cut the middle mortise, remove the second spacer and cut the “first” mortise. 3. Finish the mortises with a chisel. In effect, the DT cutter serves the role of a dovetail saw. You still have to chisel away some of the mortise until the sides are square Mark the location of the mortises on the inside of the drawer front with a mortise gauge set to ½", and remove the remaining wood that was left by the neck of the DT cutter. Then, from the side, chisel the mortises until the sides are square. Be careful. You may split the drawer front if you strike the chisel too forcefully. Just tap the chisel with the mallet until you get the feel of the chisel cutting and then meeting more resistance at the bottom of the mortise. 4. Mark the tails and lay them out on both sides of the tailboard. You will get a better fit if you clamp the side in place against the front before you mark position for the tails. 5. Cut the tails by hand or with a bandsaw. A bandsaw with a taper jig against the fence can be much quicker and more accurate than hand cutting. (The taper jig is simply a piece of wood with the same taper as the DT.) Cut to the outside of the layout line. (It is too easy to cut the tails a bit too small.) Cut away the waste between the tails with a coping saw and chisel. To avoid tearing out wood, It is important to score both sides of the tailboard before chopping out the waste, especially if the tailboard is plywood. The sides and back are joined with ½" box joints. (See Chapter 39, Boxes and Drawers) The bottom is ¼" plywood. Drawer Guides The drawer kickers are attached to the top doublers at the front and back with half-lap joints. Figure 4. Two Piece Drawer Guide Make up the drawer guides from two pieces of wood glued together to form the horizontal and vertical parts of the guide. The horizontal guides will fit flush between the front and back lower rails. The longer vertical guides 418 Table of Contents Notes and Reflections While Shaving Wood will hang on the doublers and be screwed in place. Cut the “notches” in the vertical piece even with the position of the horizontal guide, so that when the drawer guide is hanging on the doubler, the horizontal guide will be exactly even (or 1/32" beneath) the lower rail. The assembled drawer guides should be fitted beside the drawers, or they can be fitted to the drawer openings. It’s best to wait until the drawers are done to attach the guides. Fit them beside the drawers using about a 1/16" shim on both sides to allow a bit of space for drawer movement after the drawers have finish applied, and screw them to the doublers. Top The top is ¾" maple plywood with a mitered 13/16" x ¾" maple edge band glued all around & planed flush with the top. (See Chapter 37.) (I pinned the edge banding at the sides with three ¼" dowels.) Relieve the corners and round over the top edges (arrises) slightly with a hand plane or ¼" radius roundover bit in a router. It is very important to cut with the grain when rounding over the edges. 61-5 4. Assemble the back panel and rails to the back legs. Check legs for square. 5. Attach the top and bottom doublers to the back rails. 6. Assemble the side panels and rails to the front and back legs. Check legs for square. 7. Make sure the desk is squared up horizontally. The drawers and guides fill the space, and may not fit unless the frame is square.. 8. Screw and glue the kickers in place. 9. Starting at one side, dry-fit the assembled drawer guides beside the drawers using about a 1/16" shim on both sides to allow a bit of space for drawer movement after the drawers have finish applied. Make sure that all four guides and three drawers fit well inside the base. (It may be necessary to plane down some vertical guides.) Screw the drawer guides to the doublers. 10. This is also a good time to fit the drawer fronts to the openings. 11. After the drawers have been fitted and the drawer guides installed, add stop blocks to the rear end of the drawer guides so that the drawers will stop in the desired position . Finish Assembly Some of the following steps can be carried out whenever you are ready. 1. Assemble the front rails to the front legs. 2. Attach the top and bottom doublers 3. Insert the vertical dividers and lock in with dowels. Check legs for square. Maple is notorious for blotching when stained. Washcoating will reduce stain penetration and blotching. Another approach is to use lightly tinted linseed oil to pop the grain. This is not technically washcoating but it will reduce the penetration of subsequent applications of oil stain or wood dye. 419 Table of Contents 61-6 Small Writing Desk I think the so-called classic maple color, e.g., Minwax Colonial Maple, is too orange. I prefer a lighter color, more toward brown tones. 1. Pop the grain (and washcoat to prevent blotching) with the following mixture: Boiled Linseed Oil 6 oz Mineral Spirits 3 oz Minwax Colonial Maple Oil Stain 1 oz Minwax Polyshades Pecan 1 oz Brush on, let penetrate for 15 minutes or so, and wipe off well. Allow to dry overnight, sand very lightly 2. Seal with dewaxed shellac. If desired, the 3. 4. 5. 6. 6a. shellac can be tinted slightly to change the color tone. Apply 3-4 coats of alkyd varnish to outside surfaces. Shellac interior surfaces, including drawers and drawer guides. Don’t neglect the underside of the top. Carefully wet sand top as needed with 320-400-600 grit paper, lubricated with water (with 1 drop of dish detergent). Smooth remaining surfaces of drawer, including bottoms of sides (runners). Rub runners lightly with wax candle. Optionally, allow to cure a week or so, then rub with pumice in light mineral oil on a felt block. Then wax. If you don’t choose to carry out step 6, rub top and drawer fronts with 0000 steel wool and wax. 420 Table of Contents Notes and Reflections While Shaving Wood 61-7 Figure 5. Exploded Sketch 421 Table of Contents 61-8 Small Writing Desk Cut List 1. Rear Leg Front Leg 1½ x 1½ x 29 8/4 soft maple, tapered, mortices and grooves. See description. 2 2 2. Rear Top/ Bottom Rail 1¼ x ¾ x 36 (Overall (ovl) 4/4 soft maple ¾" tenon both ends (TBE) 2 4/4 soft maple ¾" TBE. Cut with drawer fronts 2 3. Front Top/ Bottom Rail 4. Side Top/ Bottom Rail 1¼ x ¾ x 20 (ovl.) Front & Back Top/ Bottom Doubler 1¼ x ¾ x 34½ 6. Drawer Kicker 1¼ x ¾ x 19¼ (Ovl) 7. Vert. Drawer guide (outside) 1¾ x e x 18 d 8 Vert. Drawer guide (inside) 1 ½ x ¾ x 20 2E wood half-lap mortise (notch) both ends 9. Horiz Drawer Guide 1 ½ x ¾ x 18 d 2E wood Fit to case 10 Top 40 x 24 x 7/8 5. 4/4 maple 4 2E wood (poplar) 4 2E wood half-lap mortise both ends 4 2E wood. Fit to case 2 1 A 38½ x 22½ x 3/4 ¾" Maple plywood B 40 x 13/16 x ¾ C 24 x 13/16 x ¾ 4/4 soft maple, mitered both ends, fit to case 11 Shelf 38 x 8 x ¾ 12 Shelf supports 6x8x¾ 4 4/4 soft maple 422 Table of Contents Notes and Reflections While Shaving Wood 61-9 13 Shelf Back 5¼ x 33 ½ x ½ ½" Baltic Birch (BB) Ply 14 Side Panel 19 x 4 1/2 2 15 Back Panel 35 x 4 ½ ovl ½" BB Ply cut ¼ x ¼ “ tongue all around Drawer Front 10 x 4 x 34 roughly (rgh) 4/4 maple, cut out with rails, fit to openings 3 Front Divider 1 ½ x 4 x ¾" rgh cut out with rails, fit to openings 16 x 3 f x ½" ½" BB Ply dovetail front, box joint back 6 16 17 18 Drawer sides 1 19 Drawer back 10 x 3 f x ½" rgh ½" BB Ply , fit to openings 3 20 Drawer Bottom 15 ½ x 10 x ¼" rgh ¼" Ply, fit to drawers 3 References Drawers, in Practical Furniture Design Taunton Press p. 150-157. Will Neptune, Engineering a Table With 423 Table of Contents 424 Table of Contents Notes and Reflections While Shaving Wood Chapter 62 – Official Translations Many common phrases used in woodworking are actually “terms of art” that mean something quite different than the plain language would indicate. Here are a few tips to promote understanding among woodworkers and clients. ASKING FOR ADVICE He says . . . He means . . . I don’t remember how I have no idea How would you cut this with a router? I have no idea, but I have a router What’s the best way to do this? I can only think of one way but I’m not sure I can pull it off Do I need an edge guide? I’ve never used this tool before Would you show me what you mean? It would be easier for both of us if you just do it for me How do I get this to fit? I’ve cut it twice and now it’s too short RECEIVING ADVICE He says . . . He means . . . Everybody knows . . . I believe It’s obvious . . . I can’t explain why I think this Anybody with half a brain can see I can’t explain why I think this, and I’m afraid you’ll ask me to Common sense method How my first boss did it. If I ever have to do it, this is what I’ll try Standard method I read it in a magazine. It’s the only way I know Normally, I do it this way I tried it once and it worked Accepted procedure I have done it this way twice 62-1 Table of Contents 425 Official Translations 62-2 Classic approach The guy who showed me did it this way. I tried it and it worked. How Norm did it on TV It will probably work, but I don’t know all the steps required and I have no idea how long it will take. I always do that with hand tools Stop bothering me/ Go ask somebody else Try it on a piece of scrap first You’ll probably screw it up the first time like I did Client Relations You say . . . Translation . . . Well, the one in the store is mass produced out of MDF in a factory I want this job. My machines are small and cute. I want this job. This won’t take us very long to make Please give me the deposit now We’re finishing it now It broke during glue-up It’ll be done next week We forgot to order the lumber It’s the next project in line We are trying to find the work order I promise I’m gonna hang up the phone now We still have one or two small projects ahead of yours We’re sorry that we took this job, you cheap SOB I don’t know if we can get any more of the wood you wanted We’re sorry that we took this job, you cheap SOB The truck is broke, I don’t know when we can get it out to ya Come and pick it up yourself, [you cheap SOB]. Hand made furniture always has a few tool marks – that’s how you know its hand made! We are not going to do it over. You have to pay for it anyway 426 Table of Contents Notes and Reflections While Shaving Wood 62-3 Tool Terminology The ad says . . . Translation . . . Automatic You can’t adjust it Cordless Get an extra battery Digital You can’t fix it Exclusive House Brand. Get extra parts now. Economical Start saving for a replacement. Japanese You can’t re-sharpen it. Lightweight Plastic Portable Get a bigger toolbox Starter set More pieces than you will ever need. Also, see economical Versatile Lots of setups. Keep the manual handy. American Made The label, that is 427 Table of Contents 428 Table of Contents Notes and Reflections While Shaving Wood Chapter 63 – Jake's Chair, Amendedb Introduction J ake's Chair is an Adirondack chair. The back is curved and the seat is contoured. Tom Gauldin, who designed the original version of this chair, wrote that he called it "Jake's Chair" in honor of Judge Jake Robertson, of Marshall, MO. This is a large chair. As modified it requires about 36 x 38" of floor space, plus the footstool. It is comfortable without cushions. It would not be difficult to design a narrower chair if desired, but this size is luxurious for persons of ample proportions. Gauldin’sa design is stable and reasonably easy to build. It includes level arms at a good height for holding a book. The chair arms extend forward to assist one in rising from the chair; they are wide enough to easily hold a glass or dish. I have modified the design somewhat. I made the back longer, to support the head of a tall person. The longer back required that I add a stretcher near the top of the back. The upper back support (UBS) is attached to the arms. (See Figure 2.) In the original design, the UBS was screwed to the endgrain of the arm runners. This was a design weakness that Gauldin recognized, because the UBS could sometimes split. Attaching it to the arms Figure 1. Jake’s Chairs, Amended, and 27" Footstools, Varnished Cypress makes a stronger joint because long grain is glued to long grain. I also increased the angle of the back to the seat, or at least I revised the construction details so that the angle would be more reproducible. One could follow the original directions carefully, and still produce a chair with a seat angle of less than 90E. An angle of 90E is the minimum comfortable angle for most people, especially if they use the footstool. An angle of about 100E provides lumbar support and makes the chair more comfortable for a larger person, although it might require a pillow behind the head. Finally, the front leg looked disproportionately wide, and I doubted that ¾" cedar or cypress would withstand lateral stress such b Jakes Chair, by Tom Gauldin, as revised April 08, 1998. The plans, photographs and text file for Gauldin’s version of Jake's chair and footstool are available for free in .ZIP format from many sites on the internet (just search for Jakes chair), including http://www.jakeschair.com/download.php. In essence, Gauldin put them into the public domain, with the caveat that they may not be sold. His original description was written to walk a novice woodworker through the construction process. In addition to design modifications, I have heavily edited the original text, refined the estimate of wood required, and added some detail to the rough cut list. Gauldin is not responsible for any errors and omissions that I might introduce in my version. 63-1 Table of Contents 429 63-2 Jake’s Chair, Amended as being dragged across a floor. I narrowed the width of the front legs, thickened them to 1- 1¼" and added a secondary leg as a seat frame support to the inside of the front leg. This yields a narrower, stronger front leg structure. Figure 2. Side View of Jake’s Chair, Amended 430 Table of Contents Notes and Reflections While Shaving Wood Plans Download and print out each of Gauldin’s .bmp drawings of the parts (P1.bmp P12.bmp, see footnote a). Almost any graphics display program should handle the bmp format. They will print out on 8 ½ x 11" paper. Also, CAD drawings, prepared by Scott Brownell, are available on the web.c Prepare full-size drawings. You can draw them easily enough, or you can print Brownell’s CAD drawings to full scale. Theoretically, you can enlarge the BMP drawings, but in reality it is much quicker to draw them than to get each one to accurate full scale. (Each bmp drawing has a different scale.) With the exceptions noted below, the parts for the amended chair are identical to the parts for the original chair. The following parts are different in my amended version: 12. The seat frame is 7-7¼" wide, so it can be cut from a nominal 8" wide 5/4 board. The original dimensions of the seat frame (aka lounging leg) called for 8" actual width of ¾ thick stock. I think a 7" wide seat frame is plenty strong enough in 1" thick cypress (5/4" stock) if the stock is dense and sound. Or, you could buy a 5/4 x 10 for this part. You will also use 5/4 stock to make the front legs (See Fig 9, Layout). 13. The legs are about 3½" wide, so that two can be cut from a nominal 8" board. They require 5/4 stock. 14. The chair arms (See Figure 4.) should be c Try http://www.xmission.com/~sherwin/downlo ad1.htm 63-3 about 29" long (2" longer than in the original design shown in p2.bmp). It is simplest to add the length to the part of the arm with straight sides (the wide part). If you want to make them shorter, you can move them back, leaving less overhang in front of the legs. 15. The seat back blanks range from 35" long to 31¾" long. (I just cut them all 35" long.) 16. The upper back support (aka upper back brace – part #4) should be cut to 32" (rough) and trimmed to fit during assembly time. 17. The back stretcher is an additional part. It should be about 1¼- 1½" wide, cut on the bandsaw with a inside radius of 38". Cut it from stock about 3¾" wide by 27" long. 18. The secondary legs are also added parts. The tops are cut on a bias to fit the seat frame: they are 3 ½" wide (same as the legs) and about 9 f" long (the short edge is about 8 e" long). Ideally, they should be trimmed to fit snugly under the seat frame, glued to the front leg. Patterns and Templates Before beginning, make sure that the stock is smooth and uniform in thickness. (Ripple marks from the sawmill planer will be very visible after a glossy finish has been applied.) Draw the parts on paper and trace them to the surfaced lumber, or draw the outline directly on the surfaced lumber. If you're building more than one chair, templates are surely worth making for curved or odd-shaped cuts. They are not necessary for rectangular pieces, such as seat slats, legs, back etc. Instead, just rip the appropriate length of 431 Table of Contents 63-4 Jake’s Chair, Amended material & cut it to length. Draw the patterns on ¼" Masonite or plywood pattern stock. If you will use a router to fair the cut edges, use pattern stock ¼" or thicker to give the bearing a good surface to ride on. See Figure 9 for a suggested cutting layout. Study the drawings. Many parts of the chair may be cut differently than the specified dimensions. For example, many edges on the seat frame, arms, and braces are for appearance and do not have to be exactly as drawn. Some dimensions, however, should be respected. DETAILS The following points need emphasis. They will make more sense if you refer to the drawings of the parts. 1. The two ends of the lower back support (LBS, part # 5) rest on the seat frame in a notch (“ear”). See Figure 3. The depth of the notch in the seat frame should match the thickness of the LBS; therefore, the notch should be cut deeper if you use thicker material for the LBS. 2. The angle between the seat back and the seat is influenced by the width of the LBS, in particular the length of the fingers that rest in the notch on the seat frame, and by the position of the UBS at the end of the chair arms. Changing these dimensions, in particular the length of the fingers on the LBS or the position (length) of the arms will change the angle of the seat back. See Figure 3. If the “fingers” on the LBS are too long the seat back angle will be too Figure 3. Lower Back Support. The inside arc should be 4¼ - 4½" in front of the notch in the seat frame. great. 3. The width of the chair (nominally 27") is determined by the length of the LBS. The length of the chair slats and the UBS should fit that width. The seat slats should be exactly long enough. If they protrude past the edge of the seat frame they will interfere with the glue joint between the seat frame and the front leg. 4. The left and right arms (p2.bmp) are mirror images of one another. Lay out the arms wide end to wide end. Since wood may warp slightly when damp, it is best to orient the left and right arms with the annular rings "up" (bark Figure 4. Modified Chair Arm 432 Table of Contents Notes and Reflections While Shaving Wood side up) so that swelling will cup the sides or the arm downward and center upward, thus permitting water to run off more easily. When laying out parts on the lumber, allow for saw kerf (about c"). Also, leave about 1/16" to allow you to cut outside the layout line on the bandsaw, so you can fair the edges with a router, spokeshaves or sandpaper. Figure 5. Rear View You can partially overlap some pieces, e.g., the seat frames, and nest some others, but plan how you will cut down longer pieces of stock so that they will fit in the bandsaw. CUT PARTS After laying out all of the pieces for both the chair and footstool (34 pieces for the chair and 21 for the footstool) first cut out the irregularly shaped pieces, then cut the straight pieces. Some boards can be cut up to make them easier to handle or to fit in a bandsaw. Bandsaw cuts will leave saw marks and 63-5 will need to be smoothed. Cut curves about 1/32" outside the line to leave wood for smoothing/fairing. 1. Cut the front of the seat frame with a flat, slanted 1¾" edge to receive the front slat. (This is not clear from the diagram, but see Figure 2.) 2. Cut the notch in the seat frame carefully. It will receive the LBS and should be cut at 90E. 3. The angle at the bottom rear of the seat frame should be 18 E 4. The inside edge of the LBS should be cut or chamfered according to the seat back angle, e.g., 100E. 5. The inside edge of the UBS should be cut at the correct angle, e.g., 30E. (The seat back on the constructed chair is about 118E from horizontal. 6. The length of the lower back support and the seat slats should be the same (nominally 27"). Cut the seat slats a bit long and trim them after you have attached the LBS to the seat frame. Likewise, do not cut the UBS to final length until you are ready to attach it to the arms. After cutting, smooth all edges with a jointer, hand plane, spokeshave or sand-paper. If you will round over edges with a router and a bearing-guided trim cutter, remember that the cutter will follow and magnify any imperfections along the edge. LAYOUT BACK The back comprises seven boards. The boards are cut square on the bottom, but the tops are 433 Table of Contents 63-6 Jake’s Chair, Amended cut so that when assembled, they form an arc. 1. Cut seven boards for the back to a width of 3-1/4" and a uniform length of 35". 2. Place the boards together, face down, on a flat surface. The corners of the bottoms of the back slats should touch. The top pieces should be flared to a width of 28¾".The gap between the tops of all seven boards should be uniform (a bit less than 1"). 3. Mark the center of the center slat 17 ¼" from the top. Draw a 17" radius from this point across the top of the seven slats, using a and arm supports, I prefer about ¼" radius roundover, just to soften the edges. I used a ½" radius roundover router bit for the seat and footstool slats, but rounded over only the slats on the curved parts. I did not round over slats that would be attached to the straight portions of the seat frame or footstool, e.g., seat slats # 4-10 (counting from the front). Edges that will be joined with other pieces should be left square, for the most bearing surface and to improve looks. Sand the rounded edges and other surfaces. Start with 100 grit if hand sanding (or 120 if using a ROS). This is the easiest time to sandpaper the parts. Assembly These directions are detailed, in the hope of being explicit. This much detail may be counterproductive for some people. It should be possible to read them through, take note of the order of assembly and special notes, and then do it your way. Figure 6. Front View. Wide slat position on Footstool is highlighted beam compass or a pencil tied with string. These mark the top of each slat. (See Figures 5 and 6) EDGE TREATMENT Unattached edges should be rounded over. This would be fastest with a bearing-guided roundover bit on a router table. I prefer to use hand tools except for the seat and footstool slats. On the rear of the seat frame, legs, arms GLUE, SCREWS, AND PILOT HOLES I prefer PVA Type II glue (Titebond III) for gluing this chair. It is waterproof, inexpensive and cleans up with water. Polyurethane glue (Gorilla Glue) ia an alternative. It is waterproof, not quite as strong as PVA on wood, and cleanup is a pain. Stainless steel (SS) screws are greatly preferable for outdoor use, especially if you will varnish the chair. Number 8 screws are a good choice for screw joints in softwoods, and square drive will save a liot of time and annoyance. (See the appendix on materials.) It is definitely worthwhile to drill pilot 434 Table of Contents Notes and Reflections While Shaving Wood Figure 7. Bottom View (Photo by T Gauldin) holes for all screws. Each chair and footstool set requires about 160 screws. Most of them show, and many need to be aligned to look right. Softwoods like cedar and cypress often split if you insert a screw without drilling a pilot hole, especially in narrow pieces like corner braces and seat slats. Furthermore, a larger pilot hole in the “top” part will prevent bridging and let the top part be drawn tightly against the lower piece. Therefore, you need to drill some parts twice, once for the shank and once for the threads. Adjust a power driver to set the screws snug but not to bury the head into the wood. Over-tightening with a power driver in softwood is a common mistake when you are used to driving screws into hardwood. This is pointless, a bit unsightly, and might split the wood or strip out the threads. You can check tightness with a hand screwdriver. It will save time to set up the coun-tersink and 5/32" bit on a drill press with a fence and 63-7 stop blocks to get the screw holes at the ends of the seat slats in the right locations, centered on the slat and seat frame so that they will line up after assembly. The 1-½" seat slats take one screw at each end to join them to the seat frame. The front (2") seat slat takes two screws at each end for more strength. The location of some holes shown on the part drawings may not be exactly correct. Don’t pre-drill any other holes until assembly time, after you have dry-fit the part. Drill through both parts with a 3/32" bit (for #8 screws in softwood). Then, countersink the holes in the top part (the part being attached) using a #8 (5/32") adjustable pilot/countersink drill, set for the thickness of the top part. (If the depth of your pilot/countersink bit isn’t adjustable, you will need to remove the top part before drilling) CHAIR BASE ASSEMBLY 1, Place the seat frames upright, parallel and square 27" (outside measurement) apart. Place the lower back support (LBS) on the top of the seat frames with the "ears" firmly against the stops. If necessary, trim the ends of the LBS to be flush with the sides of the seat frames. Use a story pole to make sure that the two seat frames are parallel to each other. (See Figures) 2. Using the approximate hole locations from the pattern of the LBS, drill through the LBS into the seat frame, apply exterior glue, and screw the LBS to the seat frame with 2" stainless steel screws. 3. Before installing the 2" by 27" front seat slat (Figures 2 and 6), place the front edge of the seat frame on scraps of 1/4" material to raise it slightly above the workbench. The 435 Table of Contents 63-8 Jake’s Chair, Amended rounded edge of the first slat should slightly "overlay" the front of the seat frame and cover its lower front edge. Use a story pole or measure the distances between the insides of the seat frame at the front and middle. They should be the same. Position the wide (2" by 27") seat slat against the front of the seat frame. Its ends should be flush with the sides of the seat frame and its bottom edge should be flat against the workbench, ¼" lower than the seat frame. (Figure 2) . (The end of the seat frame should have a flat spot at the end to mate with the slat.) Attach the slat to the seat frame firmly with glue and four 1-5/8" ss screws. 4. Turn the seat over and install the two "L"-shaped corner brackets under the lower back support. (Figures 5 and 7) The grain orientation of the corner bracket matters somewhat. Long grain will make a stronger glue joint with the seat frame than end grain. The screws run from the lower back support down into the corner brackets and from the seat frame into the corner brackets. Pilot holes in the corner brackets are important to avoid splitting them. The corner braces should force the two seat frames into square, but take care that the seat frames also remain parallel with each other and are not racked into a parallelogram shape. 5, Now, using two small scraps of ¼ - 5/16" material as spacers between each seat slat, begin at the front slat and glue/screw each seat slat to the seat frame. (If the spacing is too narrow, you won’t be able to get a painting pad between the slats.) The only wide seat slat is the one used as the starter at the front. All others are the narrow ones. As you approach the back support in the rear, adjust your spacing to give an even spacing between the final 2-3 slats. It may be necessary to cut away part of the last slat to fit it under the LBS. The narrow seat slats are important to the comfort of the chair. They may flex a bit when someone sits on the seat. Check the third slat on the seat. This will take all the weight of a person sitting on the edge of the chair. If it is too flexible, consider gluing and screwing another slat underneath it as a reinforcement. Likewise, if the slats at the back are too narrow (trimmed to fit) they may need to be reinforced. CHAIR ARMS AND BACK ASSEMBLY 1. Install the arm braces to the outside top of the front legs, centered and flush with the top of the leg. (Figures 2 & 6) The screws run from inside the front legs into the arm braces. 2. Cut two scraps of wood 10-½" long and use them as temporary legs to support the front seat. The front slat should be 10-½" above the workbench. This should give an 18E angle with the horizontal. The taper cut at the rear of the seat frame should be flat on the workbench. Clamp the front legs to the outside of the seat frame 3" back from the lower front tip of the seat frame (not the front slat), being sure that the legs are plumb (Figure 2). Mark the position, apply glue, and screw the legs to the seat frame. Add the secondary legs. 3. Pre-drill the four screw holes in the front of the arm runners, with the counter-sink on the inside. Install the arm runner to the inside of the front legs, flush at the top and front of the front legs, and level (assuming the workbench is level). The four screws pass 436 Table of Contents Notes and Reflections While Shaving Wood from inside the arm runners into the front legs. The arm runners should be parallel. Check the distance between them at the front and the back.) 4. Clamp the arms in position on the arm runners. The insides of the arms should be about 1½" from the inside of the arm runners. The fronts of the arms should protrude about 2-2¼" from the front of the legs. (Figures 2 and 6) The arms should be "bark side up". 5, Measure the distance between the outsides of the arms at the rear. Clamp the upper back support (UBS) under the arms, against the runners. Draw a guideline on the UBS that is flush with the arms at the outside and the rear. 6. The angle of the seat to the seat back should be between 90E and 100E. Figure 8 Figure 8. Proportions to yield 100E Seat Back Angle shows the approximate geometry required to get a back angle of 100 E. Point “A” is the intersection of the front of the leg and the seat, “B” is the intersection of the UBS and the seat back, and “C” is the front of the LBS. d d If distance AB is 23 1/2", the seat angle will be 95E. The law of cosines is c2 = a2 + b2 – 2ab cos C, where the lower case lettrs designate the leg opposite the corresponding angle. 63-9 (These dimensions are theoretical. You can make an adjustable triangle out of scrap with these dimensions and use a protractor to get the correct final dimensions.) Move the arms back or forward to adjust the seat angle and mark the underside of the arms where they meet the front of the front legs. 7. Install the arms to the front legs and arm runners, trim the UBS to fit flush with the outside and rear of the arms, and install the UBS to the arms with glue and three screws in a triangular pattern. (The triangular pattern is to avoid lining up the screws along the grain pattern of the arm.) 8. Pre-drill two screw holes in the center back slat, d" from the bottom and ¾" from each edge. Install the center seat back slat, with the bottom of the slat flush with the bottom of the LBS, centered in the LBS, and plumb from side to side. (To make it easier, you can clamp a support to the bottom of the LBS to hold the slat.) Then drill two holes in the correct position at the UBS (¾" in from each edge of the back slat). Apply a thin coat of glue and screw the center back slat to the UBS. The bevel angle on the UBS should approximately match the slope of the back slat. 9. Install the rest of the seat back slats (Figures 5&6). Work from the middle slat outward, keeping the inner bottom corner of each new slat touching the corner of the existing slat. There is no gap between slats at the bottom and equal gaps (a bit less than 1" at the top or ½" at the top of the UBS). Since the slats fan outward toward the top by about ¾"-1" each, each slat will intersect an inner slat corner slightly "lower down" on the lower back support in an arc. You may pre-drill the inner screw hole at the bottom of each slat but 437 Table of Contents 63-10 Jake’s Chair, Amended not the others, since it’s too hard to predict where the holes should be until the slat is in place. You can lay a flexible steel ruler on the back slats to line up the screws neatly. Use two screws at both the upper and lower back support contact points. The outermost seat back slats are intended to clear the inside of the arms by about ¼". Space the slats all at once and adjust them for uniform gaps at the top. The top of the chair back should measure about 26" at its widest point. Check the rear of the backslats for glue runs. 9. Install the back stretcher 11-12" above the UBS. 10. Finally, turn the finished chair over and install nylon glides to each leg where it contacts the ground. FOOTSTOOL ASSEMBLY 1. Install the corner braces to the center of one end of each 12-1/2" tall riser. (See Figures 6 and 7.) Be sure that the edge of the corner brace is flush with the end of the riser. Screw through the riser into the brace. 2. Glue, clamp and screw the upper and lower footstool ends to the risers. (Figure 7) The footstool ends are installed on the outside of the risers. The screwheads are on the inside. 3. Stand the risers up with the corner braces to the inside on the top. Screw one of the narrow slats, centered on the corner braces, across the top of the risers and ends. Use one screw at each end. Flip the footstool upside down and add two additional screws at each end up from the corner brace into the bottom of the center slat. 4. Screw each of the two wide slats to the lower frame on both sides of the riser. This will help to keep the two risers parallel and plumb for the remainder of the assembly. Finally, install the remaining slats to the upper and lower footstool ends, spacing them equally (about ¼" apart). 5. Turn the finished footstool over and nail the four remaining nylon glides to the lower footstool ends. Finish Please see Chapter 6, Outdoor Wood. Clean up any remaining glue squeezeout, sand off all pencil marks, smooth and fair rounded arrises and fill open knots, etc. I would not fill screw holes on this chair, which is why I use stainless steel screws despite their expense. A final sanding with 120-150 grit should be adequate if you plan to apply varnish or paint. If you will apply oil stain, finish sand with 180 grit, with the grain, by hand. I recommend thinning the first coat of varnish 50-50 to increase penetration. Sand lightly between coats with 180 grit paper. On the chairs shown in Figure 1, I used four coats of McCloskey’s Man O War Marine Spar Varnish for upper surfaces. (Polyurethane spar varnish for the undersides) The McCloskey’s toned the cypress to a pleasing honey color. I thinned the first coat and used a brush to apply it. I used a small paint pad to apply subsequent coats. One quart of spar varnish will cover the upper surfaces of one chair and footstool with 4 coats if you brush it out well. References Begnall. Tom, Torture Test for Outdoor 438 Table of Contents Notes and Reflections While Shaving Wood Finishes, FWW May/June 2009 Flexner, Bob Protecting Exterior Wood, American Painting Contractor, Douglas Publications, April 1999. http://www.painting-pros.com/articles/Prote ctingExteriorWood.pdf Forest Products Laboratory. 1999. Wood handbook—Wood as an engineering material. Gen. Tech. Rep. FPL–GTR–113. Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Laboratory. http://www.fpl.fs.fed.us/documnts/fplgtr/fpl gtr113/fplgtr113.pdf Some of the dotted lines used to indicate edge rounding make the drawing harder to understand, so a small drawing of the part with solid lines is included. ___ A solid line indicates the edge should be sharp on both sides (you might want to soften the edges a bit) • All dimensions are in inches. • All drawings are made to scale (exceptions are noted), but each drawing is to a different scale. A 10" line is drawn on each drawing to show the scale • • Jake’s Chair http://www.jakeschair.com/download.php Gauldin’s Notes for Jake's Chair Drawings (Files P1.bmp - P12.bmp) 63-11 • - - - A dashed line indicates that the edge should be rounded over on both sides. The recommended rounding bit has a 3/8" radius. The actual shape of non-dimensioned surfaces is left to discretion, Just draw a smooth curve to fill in the shape NTS means "not to scale". TYP means the other similar features have the same dimension. R 32.5 means the indicated dimension is a 32.5" radius The screw holes are drawn over-scale. They should be clearance holes for the screws you use to assemble the chair. Note that it is probably better to drill the holes while you are assembling the chair, rather than drilling them ahead of time Rough Cut List (Boards are ¾" finished thickness except for #1 and #3 No. Quantity Name Rough Dimensions 1 2 Seat frame 7 x 38" x 1"(actual thickness) 2 2 Arm 7 x 29 x ¾ 439 Table of Contents 63-12 Jake’s Chair, Amended Rough Cut List (Boards are ¾" finished thickness except for #1 and #3 No. Quantity Name Rough Dimensions 3 3 Front Leg 3.5 x 25 x 1" (actual thickness) 4 1 Upper Back Support (Brace) 6 x 32" (rough dimension of blank) 5 1 Lower Back Support (Brace) 6 x 27" (rough dimension of blank) 6 4 Footstool End 2 e x 12 7 2 Arm Support 3 ¼ x 16 8 1 Back Slat Assembly 7 Back Slat 3.25 x 35 9 2 Arm Runner 3.25 x 23.5 10 11 Seat slat on chair 1 9/16 x 27 (nominal length) 10 13 Seat slat on footstool 1 9/16 x 27 11 4 Lower Brace 6x6 12 1 Seat front slat on chair 2 x 27 12 2 Seat front slat (footstool) 2 x 27 13 2 Foot Stool Riser 6 x 12½ 14 1 Back Stretcher 27 x 3¾ rgh 15 2 Secondary Leg (seat frame support) 3 ½ x 9 f (long side) end beveled to 8 e on short side. Fit during assembly Cut to length after assembly of frame 440 Table of Contents Notes and Reflections While Shaving Wood 63-13 Figure 9. Layout Diagram (Parts not to exact scale) Materials WOOD Based on the layout diagram (Figure 8) the amended Jakes chair plus a 27' wide footstool requires about 41 bf. Board “A” should be 5/4 (1" finished thickness). Board “B” can be 5/4 or 4/4 (¾" finished) depending on the quality (density, strength, and surface) of the lumber. The amount of lumber to buy depends on lumber yield/waste. I’d consider buying 20-50% over depending on the lumber grade. 4/4 softwood boards, especially cedar and cypress in my experience, often contain longitudinal cracks, which lower the yield of furniture parts. Careful selection is very important. Choose heavier, darker boards that are mainly heartwood. The chair requires 75%-80% of the total, the footstool 20-25%. The chair is designed to be made out of softwood. Redwood, cedar, and cypress are decay resistant. Oldgrowth cypress and redwood are significantly more decay resistant than newgrowth, and heartwood is more resistant than sapwood. I would re-design this chair if I wanted to make one from a hardwood, because the chair would be heavy and very expensive. (See the FPL Wood Handbook) Pine is not very decay resistant. It can be used, but the endgrain should be sealed and the rest should be painted with top-quality marine paint. All joints should be sealed 441 Table of Contents 63-14 Jake’s Chair, Amended with caulk. Also, the edges of the back slats should be sealed before they are installed because it is impossible to get paint into the narrow gaps between slats at the bottom of the back. Pressure-treated pine lumber (treated with ACQ or copper azole) might be OK if it were well dried before use. I would paint it, also. A chair made of softwood would weigh about 50 lb. to join ¾" material side-to-side, and for attaching seat and footstool slats. The 1e " screws are used to fasten the arms, braces, and back slats. The 2" screws are used to attach the seat and footstool brackets and the LBS. Length Chair Footstool Total SCREWS 1-1/4" 40 40 80 (about ½ lb.) Number 7 or 8 stainless steel wood screws will last the longest and will not discolor wood, but are also the most expensive. No. 6 screws, galvanized or coated screws also may be used. Square drive will be easier to drive without damaging the heads and will save time. The necessary lengths are 1-1/4", 1-5/8" and 2". The shortest screws are used 1-5/8" 60 2" 12 60 (about ½ lb) 8 3/4" nail-on/screw-on nylon button-type glides 20 10 442 Table of Contents Notes and Reflections While Shaving Wood Chapter 64 - Truth in Woodworking (My Diabolical Dictionary) CHISEL (WOOD CHISEL): In many shops, a variation of bladed screwdriver, except that chisels are used more for poking than for prying. BAND SAW: A stationary power saw used by people who can’t follow a line with a hand saw. (Some can’t cut straight with a band saw, either. See Crap, Drifter) Also used to cut curves. Band saws require frequent blade adjustment, blade replacement, guide adjustment, and accessory purchases, making them an ideal woodworking tool. Cutting straight requires acquisition of a “decent fence,” i.e., one that is not crap. (See Crap.) Cutting curves with a band saw justifies buying at least two spokeshaves or a spindle sander to remove saw marks. (Serial tool acquisition significantly advances Woodworking) See also Cool Blocks. May be used to create firewood by cutting to the inside of the line instead of the outside (or to remove fingertips if operator concentrates too much on cutting outside the line). CLAMP: Clamps come in many forms, such as “C” clamps, bar clamps, pipe clamps. All clamps are used to mark wood by placing dimples or dents wherever they touch the wood. Pipe clamps are more efficient than other clamps because they also mark the wood with metallic oxides along the entire length of the clamp. Clamps are also used to deform mis-measured parts until they fit during assembly (glue-up). CLAMPING PAD: Device used to frustrate the primary function of clamps (i.e., denting wood). Fortunately, clamp pads remain in place only until actually touched by a clamp, when they fall to the floor and scurry under the workbench. BELT SANDER: An electric tool that removes wood rapidly and uncontrollably. Commonly used instead of a hand plane to redimension wood that you mis-measured. When used to flatten glued-up panels, replaces uneven butt joints with a visually more interesting random pattern of gouges and ridges. Also used to convert minor touch-up jobs into major re-surfacing jobs. Excellent source of sawdust, even more productive for this than a miter saw. COOL BLOCKS: A bandsaw accessory, but the name says it all. Acquisition of cool blocks may not automatically make the woodworker cool, but at least they prove that he won’t tolerate crap. COPING SAW: This saw is used to teach woodworkers that wood has three dimensions. Even if you follow the line on one side of the board, the line on the other side may look like a Salvador Dali sculpture. See Hacksaw. BENCH GRINDER: Despite its name, this tool is not used to grind benches. Its main use is to change silver-colored chisels to a more interesting shade of blue-black. CRAP: Term used to describe a tool or accessory whose actual capacity or 64-1 Table of Contents 443 64-2 Truth In Woodworking performance is less than the operator’s expectation, e.g., a circular saw fitted with a dull plywood blade being used to cut 8/4 oak: “This saw is crap.” Often used to describe the standard fences and bearings shipped with a bandsaw. Occasionally the operator will use a crap tool as a Dammit tool rather than adjust his expectation. DAMMIT TOOL: This tool should be kept available at all times for you to grab and throw across the shop while reciting an incantation such as “DAMMIT”. This is usually the next tool that you need, so you get to repeat the incantation more loudly when you find that you just threw it across the room. This tool is also used to demonstrate unrealized human potential, because you can usually throw the most vulnerable end of the tool against the most vulnerable piece in your shop. DRIFTER: A person who can’t cut a straight line with a bandsaw. This is always considered a defect in the saw. (See Crap.) DRILL PRESS: A stationary power tool useful for hurling a chuck key into your chest, the darkest corner of the shop, or onto a freshly-painted project. Also used to smack yourself in the chest with an unsecured workpiece. Other uses include burning holes in wood with Forstner bits or hole saws. DUST COLLECTOR: (1) A tool used to raise the noise level and amperage requirement in a woodworking shop. Minimum size of dust collector is usually selected by Ahm’s Law: DCA + TA > C (The amps needed to start the dust collector plus the amps needed to start the tool always exceed the capacity of the circuit breaker.) Also used to start pointless arguments about design efficiency, minimum CFM, cyclonic action, and static pressure (suckiness). (2) A tool, usually expensive, that you bought but never get around to actually use. Indicates the interval since the last shop cleanup. ELECTRIC HAND DRILL: This is an extremely versatile tool. Its main use is to drill cocked holes in wood and to stretch wrist joints when drilling sheet metal. (Can also quickly convert sheet metal into a cutting tool.) Also used (a) with Phillips screwdriver bit to accelerate the demolition of the screw head or to twist it off altogether, (b) with a screw extractor to permanently mark the area around a broken screw, and ( c) for spinning pop rivets in their holes until you get bored watching. Occasionally used to perforate the operator’s left hand if he holds the workpiece in the most natural position. EXTENSION CORD: Used to decorate shop floor and, with sawdust and cutoffs, to test the agility and balance of the woodworker. If equipped with enough plugs, also may be used as a temporary heat source. GLUE-UP: The phase of custom furniture making in which design and cutting errors are revealed, usually for the first time. Glueups help woodworkers to gain a perspective on lesser annoyances such as divorce and accidental loss of body parts. Also used by the woodworkers family and friends to 444 Table of Contents Notes and Reflections While Shaving Wood gauge the woodworker’s mental health, stress tolerance, and level of addiction. HACKSAW: A saw used for cutting a crooked, random-walk line. This tool, like its relative the coping saw, is used to teach Zen principles about the unpredictability of life. The more you try to control it, the larger the error becomes. Equipped with a wingnut useful for work hardening the operator’s thumb. HAMMER: A striking tool that is used to awaken primal urges in craftsmen of all ages. Its main uses are (a) to place dents around nail holes so that they can be easily located even after the actual hole has been filled and painted over and (b) to assemble mortise and tenon joints after the glue has almost dried. (In the latter application the dents will permanently mark the location of those joints.) Also used in conjunction with the Dammit tool to shorten the workday by crushing thumbs and pinky fingers. LATHE: A tool for rotating wood, used to make spindles or bowls decorated with random nicks and divots. Also used to test the structural integrity and strength of bowl blanks and of chisels, by slamming them against the tool rest. MITER SAW: A crosscut device used to cut rails and other pieces that are exactly 3/32" too short. Can be used to place a little zigzag in the cut if the workpiece is wider than 6". Also used to fill the shop with sawdust. A good miter saw will defeat any dust collector. 64-3 PLANE (HANDPLANE): A hand tool that was developed to demonstrate that woodworkers will actually pay more than $100 a pound for iron. Also used (a) to create interesting textures in expensive figured wood by tearing out random chunks and (b) for occupational psychotherapy. After a nearly complete curly maple tabletop is ruined by a handplane, the woodworker is advised to sharpen the plane iron for hours on end to calm himself. NB A handplane is seldom a good choice for a dammit tool. PLANER: A power tool used to produce an attractive ripple effect on the surface of expensive lumber. PLIERS: Hand tool used to round off bolt heads. A vise grip pliers is generally used after pliers to finish rounding off the bolt head. Also used to ruin threads and to create blood-blisters. PRY BAR: A tool used to take apart a joint you just hammered together (see Hammer), after you recognize that you put it together backwards. Also useful for marking the spot with a dent for future reference. RE-PURPOSING: Using a tool in a manner or for a purpose that was not intended by its designer. The mark of an efficient woodworker, who does not waste time reaching for the right tool. For example, any tool can be used as a hammer (once). Using a workbench as a work surface instead of a storage shelf is semi-re-purposing because it might originally have been designed as a work surface. 445 Table of Contents 64-4 Truth In Woodworking ROUTER: A hand tool used to cut random walk designs called mortises or dados. Also used to raise chunks of wood for novel edge treatments. In so-called climb cutting mode it can propel wood great distances. SCREWDRIVER (BLADED): Used to open paint cans or when a pry bar is out of reach. Also used to widen and taper screw slots so that the screw can neither be driven home nor removed. This procedure with this tool is the only known way to produce a blister in the center of your palm. SCREWDRIVER (PHILLIPS): Used to stab things like the seal under the cap of a Mineral Spirits can or for opening canned goods. Also used as a cam action countersink to prevent Phillips head screws from being rotated in either direction. For faster results, may be used with electric drill SCREWDRIVER (SQUARE DRIVE): This tool is often selected by mistake instead of a Phillips screwdriver. Its cam action on a Phillips screw head is faster and it produces a smoother, square dimple that has even less chance of ever being rotated in or out. SQUEEZE-OUT: The glue that flows out of a joint when clamp pressure is applied. Used to indicate that you applied too much glue (again). When squeeze-out can no longer be removed with sharp instruments, it indicates that the glue has fully set. Especially after the application of dark oil stain, squeeze-out may also be used to show the client that enough glue was used to make a strong joint. TABLE SAW: A large stationary power tool commonly used to place semi-circular brown marks along the edge of a board. Also used to launch wood projectiles for strength testing of belt buckles, drywall, and internal organs. THICKNESS PLANER: A tool used to change a warped board into a thinner warped board. UTILITY KNIFE: Used to scribe or score wood. The thick handle obscures the straightedge, usually insuring that the score mark will be crooked or offset from where you wanted it. Also used to cut the contents of cardboard cartons before the carton is even open. Can be left open and placed in a back pocket for slicing car seats, your pants, or your ass. WIRE WHEEL: Used (a) to clean paint and rust off metal parts by heating them to 200E and (b) then to throw them under the workbench when you loosen your grip. Also used by people in the witness protection program to remove fingerprints from their fingers and to test the speed of the blink reflex. WOODWORKER: A woodworker is a kind of tool collector with twin fixations on tools (ergophilia) and expensive wood (xylophilia). The woodworker typically defaces and otherwise abuses the wood with a variety of sharp objects (“furniture making”) but in a common variant the woodworker simply accumulates tools, lumber, or both (this is called “setting up the shop”). An extreme form, called ergonoia 446 Table of Contents Notes and Reflections While Shaving Wood (tool brain, the woodworking form of dieselhead), can be recognized by the feeling, “I am only as good as my tools”. Diagnostic question for ergonoia: “You must have ten thousand dollars tied up in here.” Diagnostic response: “Yes. Thank you.” WOODWORKING: An enterprise intended to justify acquisition of tools or furnituregrade hardwood lumber. Also useful to 64-5 produce kindling wood. May be less expensive than alcoholism and safer than glider flying. ____________________ * Inspired by, and based on, earlier anonymous versions found floating around the internet. I could not have thought of the Dammit tool by myself. 447 Table of Contents 64-6 448 Table of Contents Notes and Reflections While Shaving Wood Chapter 65 -- Angels in the Workshop The Zen of Woodworking A on measuring and cutting accurately. I set up a stop block on the saw. I cut the two longer pieces and checked their length. OK. As I picked up the third piece, a thought entered my mind, something about the length not being 29 ½". Immediately another thought appeared. Of course the stiles I had just cut were 29½"! I had a mildly annoyed feeling. I had triple checked them, dammit! I went back to the cut list and checked. That’s right – 29 ½". Then, I cut the third piece, to 29½". It was only after I went to the next item on the cut list that I remembered that the center stile was supposed to be 26½". Well, it was easy enough to cut a little more off. No big deal. The story is not about the mistake. The little “voice” was reminding me that the piece I was about to cut should not be 29½“ The second voice changed the subject away from the board I was about to cut, back to the ones I had already cut. The second voice was defensive – I had not made a mistake, dammit! It silenced the voice that was trying to tell me that I was about to make a mistake. The little voice did not return when I went back to the saw. There was no internal “argument.” On another occasion, I had set up a cut on my band saw and was ready to switch it on. A thought occurred to me about the fact that I like to relieve the tension on the blade each night before I put the saw away. I was thinking that some people don’t do that, but that I think it’s wise. I was wondering whether it was worth the trouble, etc., etc. Just idly ruminating as I prepared to make my cut. Then, I remembered that I had not yet re- s you can tell from Ch. 13, I’m interested in workshop safety. I also read comments, anecdotes, etc from other woodworkers on forums like Sawmill Creek and Woodnet. Many of the postings are about accidents or near misses. Many of these mention a sense of danger or a feeling of “don’t do that” just before an accident. Sometimes the person stopped what he or she was doing and felt that an accident had been avoided, barely. Sometimes, he writes that he wished he had stopped. I do not recall ever experiencing foreboding before an accident in my workshop. On the contrary, I have experienced foreboding when nothing whatever was wrong. Such forebodings are rare for me, in any case. My experience is a bit different. I occasionally notice a quiet thought in my mind. It’s easiest to call it a voice, but it is not an auditory experience or hallucination. Sometimes it is formed verbally, but sometimes not. I experience it as a clear idea that rises to consciousness. These experiences are subtle, not at all dramatic. They do not compare to the sense that an actual injury is about to happen. But they more than make up for their lack of drama by occurring frequently. I wonder if others have similar experiences, and if so what they think of them. Here’s an example. I had prepared 30" long stock pieces. I wanted to cut three pieces, two 29½" long and one 26½" long. I wanted them to be exactly square and two of them to be exactly the same length. I was thinking that I had to check the other end of each piece for square, and I was concentrating 65-1 Table of Contents 449 65-2 Angels in the Workshop tensioned the blade. If I had started the cut, the blade might have come off the wheels or at least the cut would have been crooked and the work spoiled. Again, not a life-shattering experience either way. But, how strange that I would be ruminating about the blade tension and only then recognize that I was about to make a mistake. Sometimes the “voice” appears as the idea that sometimes we don’t get the result we expect. If it were an actual voice from somebody in the shop with me, I would say that he was very quiet and remarkably tactful and indirect. The “voice” almost never says, Watch out!, or Wait a minute! It seems to be gently opening up a topic for discussion, hoping that I will notice something. Like a person quietly coughing to get your attention. Sometimes, it appears as a memory of a time when I made a mistake. I hardly ever turn on my drill press without remembering the time that I left the key in the chuck and it hit me in the chest. Now I recognize that thought as a gentle reminder to check on the whereabouts of the key. Often, however, I fail to see any relevance to what I am doing. I think of it as just something that happened to come into my head. It’s easily displaced by attention to what I am doing. Then, later, I may see why that particular thought had occurred to me at that particular moment. The quiet “voice” is almost always trying to help and guide me. In the distant past, theologians called this experience hearing the voice of an invisible guardian angel. Religious artists drew cherubs to represent them. According to popular psychology, these ideas come from the right hemisphere of my brain. The left side of the brain is supposed to be better at analytical, logical, sequential, and verbal processes. The right brain is supposed to be better at holistic, intuitive, simultaneous (non-linear), and spatial thinking. Here’s an interesting contrast: the left side of the brain can calculate the trajectory of a baseball based on mass, force, acceleration and velocity. The right brain doesn’t do that. It can, however, move your body to the exact position in the outfield to catch the ball. When you think about it, this commonplace occurrence is amazing. This view of the brain has some scientific support, although it gets distorted in the popular media. It’s largely a metaphor describing something that we do not understand -- a bit more realistic, but a lot less beautiful than the metaphor of angels. Guess which side of my brain got overdeveloped by 20 years of education, followed by 40 years in academia? According to Betty Edwards (Drawing on the Right Side of the Brain) a lot of people like me have developed bossy, overbearing left brains, because mathematics and language are emphasized in school and are required for success in many fields. One point of Edwards’ hugely successful book is to learn how to shut off these left brain functions. I suspect that the experience of going to sleep while thinking about a problem and waking up with an original solution might be related. Maybe the right brain can play while the left brain is sleeping. So, maybe my “voices” are like the child in The Emperor’s New Clothes who sees the world as it really is. He tells the deluded grownups about reality: the emperor is not wearing new clothes. In fact, the Emperor is not wearing any clothes. His voice is small. 450 Table of Contents Notes and Reflections While Shaving Wood Some adults will not listen, but he does not bother to argue with the bossy grown-ups. So, What Now? Whether I call it an angel or my right hemisphere, I’d like to learn how to pay more attention when it speaks to me. I could at least avoid mistakes, maybe avoid injuries. It would be like having a complementary point of view, but without the hassle of a committee meeting. Sounds like a good deal to me. Many of the classic techniques for doing this, meditation for example, are really not practical in the workshop. Edwards teaches exercises that are intended to quiet the overbearing left brain, so that the right brain can express itself in a drawing. I have tried them and they work amazingly well – for drawing. In woodworking, however, I need my left brain to stay alert while I am working. I just need a way to calm it down or make it less dominant. Richard Pirsig touches on this from another perspective in his book, Zen and the Art of Motorcyle Maintenance. Pirsig says, So the thing to do when working on a motorcycle, as in any other task, is to cultivate the peace of mind which does not separate one’s self from one’s surroundings. When that is done successfully then everything else follows naturally. Peace of mind produces . . . work which will be a material reflection for others to see of the serenity at the center of it all. So, I have to at least stay calm and unhurried. Happy and contented, delighted to be in my shop is even better. When I feel annoyed, anxious, or hurried, it is usually my left brain. I think my right brain just goes and hides 65-3 somewhere. When I am delighted, perhaps my right brain comes out to play. Second, I suspect that it is possible to concentrate too much on what I am doing, e.g., on a new procedure. That’s a novel idea to me, given that intense focus on the problem at hand has been a major tool in my professional life. But it does seem as if concentration can block out context. Look at it like this: I want to do the right thing well. Doing the right thing requires an understanding of context. Doing it well requires concentration. So, when context is important, concentrating too much could result in an inappropriate action done correctly, like cutting a board exactly to the wrong length. Or winning the wrong argument, or coming up with a beautiful solution to the wrong problem. If I can get a procedure to where it’s familiar, perhaps my left brain can calm down a bit, so to speak. Perhaps, if I need to concentrate that much I’m not ready to do it “for keeps”, whatever it is. So, this is another reason to develop basic skills and standard routines. If I have too many things to think about, the little “voice” often cannot get through at all. Even if it does, I’m likely to ignore it. In the anecdote above, after I marked the cut and placed the stop block, but before I cut the stiles, I did a fivecut check on the squareness of the saw setup. This is a whole ‘nother sequence of operations. (I told you I was concentrating on being accurate.) That broke my rhythm, so to speak. It interrupted my main work flow and added one more complication. I interrupt my work fairly often. The reasons are not all obvious (short attention span?) but one reason is that I am still setting 451 Table of Contents 65-4 Angels in the Workshop up shop after all these years. I still discover, in the middle of a process, that I need a jig or something. Perhaps, eventually, I will have done enough basic operations enough times that I will usually be ready to carry them all the way through without interruption or distraction. Finally, I should find a way to make friends with my right brain, so to speak, to respect it, to listen to my angel. Maybe this is the strangest part of this strange essay. But I think I could make time in my rhythm to pause and give it a chance to “say” whatever it has. Think of it as crew management. Just take a moment on the intercom to see if “everybody” is ready to proceed. 452 Table of Contents Notes and Reflections While Shaving Wood Afterword A lmost 20 years have passed since I started this book. I started it when I began to learn about this craft, as notes to myself. I wanted to hang on to what I had learned. A lot has happened in those years, mostly good. I have aged, moved, and downsized my shop. I have continued to learn, developed new ways to work with the tools available. I have changed my opinion about some things and confirmed it about others. I have reviewed this book as I prepared it for printing. Many times, in fact. It’s what I want to say, but be careful. I was never infallible and that has become more apparent as the years have disappeared behind me. As you can surmise, I am still woodworking and still writing about it. (I don’t spend forty hours a week in my shop anymore, however.) I still have more to say. But then, I will always have more to say until I am no longer able to learn. When that day comes, I may also be unable to write. So, I have to stop somewhere or I would never finish, would I? I reluctantly bring this, the longest woodworking project of my life, to an end. Maybe there will be a second edition. Who knows? Table of Contents 453