Preview only show first 10 pages with watermark. For full document please download

Flow Front Advancement In The 3d Flow Solver

   EMBED


Share

Transcript

AUTODESK® MOLDFLOW® INSIGHT 2011 VALIDATION REPORT Flow Front Advancement in the 3D Flow Solver Executive Summary The Autodesk Moldflow 2011 release introduces the Level Set flow front tracking algorithm for flow front advancement prediction in the Coupled 3D Flow solver used in both Autodesk Moldflow Insight and Autodesk Moldflow Adviser products. The Level Set method replaces the Volume of Fluid flow front tracking algorithm used in previous releases. The Level Set method improves the symmetry of flow front prediction for symmetrical geometries, flow front smoothness and the accuracy of the flow front in predicting jetting. It is faster than the Volume of Fluid method used in previous releases when inertia is not simulated. When the option to Simulate inertia effect is selected, the solution time may increase, but the flow front prediction is more accurate, especially when jetting occurs. Introduction The flow front prediction of the Coupled 3D Flow solver in Autodesk Moldflow 2011 products has been enhanced by the introduction of the Level Set flow front tracking algorithm to address shortcomings in asymmetric flow front behavior on symmetrical geometries, flow front smoothness and the accuracy of jetting predictions. Prior versions of the Coupled 3D Flow solver used the Volume of Fluid flow front tracking algorithm. Mathematical modeling The Level Set flow front tracking algorithm is described by the following equation: 𝝏𝝏𝝏𝝏 𝝏𝝏𝝏𝝏 + 𝒖𝒖 𝝏𝝏𝝏𝝏 𝝏𝝏𝝏𝝏 + 𝒗𝒗 𝝏𝝏𝝏𝝏 𝝏𝝏𝝏𝝏 + 𝒘𝒘 𝝏𝝏𝝏𝝏 𝝏𝝏𝝏𝝏 = 𝟎𝟎 [1] where F is the level set distance function and u , v , w are the velocity components in the x, y and z directions. The level set function F defines the distance to the flow front from anywhere in the solution domain. F < 0 refers to anywhere in the filled region, F > 0 refers to anywhere outside the filled region and F = 0 refers to the exact position of flow front interface (References [1] and [2]). The Volume of Fluid method also can be described by equation [1] but then F defines the filled volumetric fraction of a predefined volume in the solution domain. F = 0 refers to an empty subvolume, F = 1 refers to a filled volume and F = 0 . 5 refers to a sub-volume representing the free surface interface. Implementation The Level Set flow front tracking algorithm uses Equation [1] to solve the distance to the flow front, F , throughout the solution domain. This results in a smoother and more accurate representation of the free surface. If the option to Simulate inertia effect is selected, the Level Set method is superior in capturing jetting. Autodesk Moldflow Insight 2011: Flow Front Advancement in the 3D Flow Solver Validation Flow front smoothness and accuracy The cylindrical part shown in Figure 1 is used to demonstrate the benefits of the Level Set algorithm in Autodesk Moldflow Insight 2011 over the Volume of Fluid method in Autodesk Moldflow Insight 2010, Release 2 in regard to flow front smoothness and accuracy. This model emphasizes the improvements in jetting prediction when the option to Simulate inertia effect is selected. Figure 1. Cylindrical part used to demonstrate predicted flow front smoothness and accuracy. Snapshots of the Polymer fill region plot taken at approximately 0.04, 0.1 and 0.3 seconds are presented in Figure 2 to compare the results from Autodesk Moldflow Insight 2011 and Autodesk Moldflow Insight 2010, Release 2. It is apparent that Autodesk Moldflow Insight 2011 is capturing the jetting phenomena. The flow front disintegrated into many fragments in Autodesk Moldflow Insight 2010, Release 2, which is incorrect. The Autodesk Moldflow Insight 2011 release is much better at predicting jetting than its predecessor. Observing the last two flow front images taken at 0.3 seconds, the Level Set algorithm highlights the smoother representation of the Polymer fill region plot when compared to the Volume of Fluid method used in Autodesk Moldflow Insight 2010-R2. The smoother plot in the 2011 release is because the Level Set algorithm calculates the distance to the flow front of every node in the domain and the interpolated contour at a distance of 0, then represents the position of the flow front interface. The previous Volume of Fluid method represented the flow front interface just as a simple filled/unfilled interpolation, which was much coarser. 2 Autodesk Moldflow Insight 2011: Flow Front Advancement in the 3D Flow Solver . Figure 2. Polymer fill region plot comparison. In general, if the option to Simulate inertia effect is selected, Autodesk Moldflow Insight 2011 will run slower than Autodesk Moldflow Insight 2010, Release 2. However, for this case the 2011 release is faster than its predecessor (refer to Table 1). During the early stages of the Volume of Fluid based analysis in Autodesk Moldflow Insight 2010, Release 2, jetting should have been predicted but instead, the flow front disintegrated into smaller regions. This flow front disintegration results in very small time steps taken by the solver which result in a long solution time in Autodesk Moldflow Insight 2010, Release 2. In general, whenever the option to Simulate inertia effect is selected and jetting occurs, then Autodesk Moldflow Insight 2011 should run faster than Autodesk Moldflow Insight 2010, Release 2. However, if jetting does not occur, it is expected that Autodesk Moldflow Insight 2010, Release 2 will be faster. Flow front symmetry To demonstrate the superior prediction of flow pattern symmetry for symmetric parts in Autodesk Moldflow Insight 2011, the model shown in Figure 3 is used. The model is a symmetric family mold of two semi-circular, 700mm long, half-round pipe sections of wall thickness 3 mm. It is expected that the tips of the two pipe sections will fill at the same time. Due to the geometry of the part, it is unlikely that jetting would occur. For this reason, the option to Simulate inertia effect is not selected in these analyses. 3 Autodesk Moldflow Insight 2011: Flow Front Advancement in the 3D Flow Solver Figure 3. Symmetric model of the half pipe. Figures 4 and 5 present a direct comparison of the Polymer fill region result taken at a time of 1.38 seconds predicted by Autodesk Moldflow Insight 2011 and Autodesk Moldflow Insight 2010, Release 2, respectively. The top left plot pane shows a zoomed in selection of the left pipe, the top right pane shows a zoomed in selection of the right pipe, and the lower plot panes show both pipes in relation to the runner system. Figure 4. Symmetric Polymer fill region plots from Autodesk Moldflow Insight 2011. 4 Autodesk Moldflow Insight 2011: Flow Front Advancement in the 3D Flow Solver The Polymer fill region result for Autodesk Moldflow Insight 2011 (Figure 4) shows that the Level Set method correctly predicts similar flow pattern positions in each pipe section. The corresponding plot for Autodesk Moldflow Insight 2010, Release 2 (Figure 5) shows the right section of the pipe filling before the left section, which is incorrect. Figure 5. Asymmetric Polymer fill region plots from Autodesk Moldflow Insight 2010, Release 2. 3D Flow solution speed It has been noted that inertia simulation was not enabled when testing symmetry on the half pipe model described in the previous section (Figures 3, 4 and 5). Table 1 shows that the Level Set method of Autodesk Moldflow Insight 2011 runs more than 10% faster than its predecessor Volume of Fluid method when the option to Simulate inertia effect is not selected. This overall speed-up occurs despite other CPU intensive features added for Autodesk Moldflow Insight 2011 such as the new sink mark and weld line calculations. However, when the option to Simulate inertia effect is turned on, Autodesk Moldflow Insight 2011 runs 3 times slower on the symmetrical model shown in Figure 3 than without the inertia option. The solution slow-down caused by enabling inertia simulation is more evident in Autodesk Moldflow Insight 2011 than it was in the previous release. However, the 2011 results are far more accurate, especially in the prediction of jetting (Figure 2). 5 Autodesk Moldflow Insight 2011: Flow Front Advancement in the 3D Flow Solver Table 1. Computation times required for cylindrical and symmetric models with and without inertia. Autodesk Moldflow Insight 2010, Release 2 Autodesk Moldflow Insight 2011 Cylindrical part (inertia) 6281.96 sec * 1296.94 sec Cylindrical part (no inertia) 228.1 sec 339.3 sec Symmetric part (inertia) 634.38 sec 1667.63 sec Symmetric part (no inertia) 595.84 sec 526.25 sec *Flow front result from this combination incorrect. Timing is not representative of a typical case. Conclusion Autodesk Moldflow Insight 2011 has improved jetting prediction, symmetry prediction and smoothness of the predicted flow front of the Coupled 3D Flow solver used in the Autodesk Moldflow Insight and Autodesk Moldflow Adviser products. This has been achieved by the introduction of the Level Set flow front tracking algorithm. The Level Set flow front tracking algorithm replaces the Volume of Fluid flow front tracking algorithm of previous releases. For analyses in which the option to Simulate inertia effect is selected, the Level Set method significantly improves accuracy, but at a cost to solution time. References 1. Adalsteinsson, D., and Sethian, J. A., A fast level set method for propagating interfaces, Jour. Comp. Phys., Vol. 118, pp. 269-277, 1995. 2. Osher, S., and Sethian, J. A., Fronts propagating with curvature dependent speed: Algorithms based on Hamilton-Jacobi formulation, Jour. Comp. Phys., Vol 79, pp. 12-49, 1988. Revised 10 March 2010. © 2010 Autodesk, Inc. All rights reserved. Except as otherwise permitted by Autodesk, Inc., this publication, or parts thereof, may not be reproduced in any form, by any method, for any purpose. Trademarks Autodesk and Moldflow are trademarks or registered trademarks of Autodesk, Inc., in the USA and/or other countries. All other brand names, product names, or trademarks belong to their respective holders. Disclaimer THIS PUBLICATION AND THE INFORMATION CONTAINED HEREIN IS MADE AVAILABLE BY AUTODESK, INC. "AS IS." AUTODESK, INC. DISCLAIMS ALL WARRANTIES, EITHER EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE REGARDING THESE MATERIALS. 6