2008 International ANSYS Conference

Size: px

Start display at page:

Download "2008 International ANSYS Conference"

Jane Richards
5 years ago
Views:

2 Content Background Filling/packing/mold cooling/deformation analysis of thermoplastic injection molding Flowing, curing and wire sweep analysis of epoxy molding compounds for encapsulation semiconductor 2008 ANSYS, Inc. All rights reserved. 2 ANSYS, Inc. Proprietary

3 Why General CFD Increasing the accuracy of injection molding simulation Epoxy Molding of Semiconductors [1] Non-uniform Thickness [2] Inertial Effect for Complex Geometry [3] 3D flow in a runner [4] Developing physics model to simulate the complex phenomena of polymer flow Customization for new processing technology 2008 ANSYS, Inc. All rights reserved. 3 ANSYS, Inc. Proprietary

4 Challenge to General CFD Convergence Difficulty Viscosity of polymer is much larger than the viscosity of air The flow approaches to stop during packing process when the temperature of polymer becomes lower than transient temperature Calculation Time A very fine mesh is necessary near wall to describe the characteristic of polymer viscosity A small time step ( s) may be needed for convergence The calculation time may be unacceptable (weeks or months/case) 2008 ANSYS, Inc. All rights reserved. 4 ANSYS, Inc. Proprietary

5 Advantages of ANSYS/CFX Stronger convergence and accuracy The free surface analysis of polymer and air with very different viscosities could convergence based on CFX accurate discretization scheme. The solution could be obtained even when the flow almost stop during packing by a coupling multigrid solver. Powerful performance in speed and parallel Could use a relative large time step ( s) to get a convergence solution Accessible parallel scalability Acceptable calculation time (1 or 2 days/case) 2008 ANSYS, Inc. All rights reserved. 5 ANSYS, Inc. Proprietary

9 Polymer Properties μ 6 Viscosity (Cross-Exp) μ ( T, p) 0 = 1 n 1+ μ 0 * τ & γ PVT(Tait) p ν ( T, p) = ν 0( T )*(1 C ln(1 + )) + ν t ( T, p) B( T ) 1.1 log( 粘度 )[P ] ºC 200ºC 220ºC 240ºC 比容積 [cm 3/g] e6 Pa 1e7 Pa 2e7 Pa 4e7 Pa 8e7 Pa log( せん断ひずみ速度 )[1/s] 温度 [C ] 2008 ANSYS, Inc. All rights reserved. 9 ANSYS, Inc. Proprietary

Mold Temperature t=1 [s] t=3 [s] t=7 [s] t=15 [s] 2008

15 Pressure & Temperature (From Filling to Packing) V/P Switch Pressure Release Shear Heating Gate seal P inet P gate P 1 P 2 P 3 : inlet : gate : ¼ of cavity length : ½ of cavity length : ¾ of cavity length 2008 ANSYS, Inc. All rights reserved. 15 ANSYS, Inc. Proprietary

Deformation X direction (wide) Y direction (length) Z direction (thickness)

19 Polymer Properties Viscosity (Herschel-Bulkely) η ( & γ, T ) = τ ( T ) / & γ + K( T ) & γ τ ( T ) = 0 exp( T / T ) y τ y y y ( n 1) { C ( T T ) /( C + ( T T ))} K( T ) = K0 exp A g B g K 粘度 [Pa s] 10 0 = K 00 ( C1 + C2α ) { α /( α α) } gel gel Curing Curing Rate (Kamal) dα = dt m n ( K1 + K2α )(1 α) K1 = A1 exp( E1 / T ) K2 = A2 exp( E1 / T ) α~0.96 α~0.65 Curing time: 90s Gel time: 35s Shear Rate[1/s] T=145 C T=160 C T=175 C Time 50 [s] T=180 C 2008 ANSYS, Inc. All rights reserved. 19 ANSYS, Inc. Proprietary

Animation of Melt Front Passing Wires 2008 ANSYS,

Fluid Force 0.4 0.3 2.50E-03 2.00E-03 1.50E-03 Fx 力 [N ] 0.2 0.1 0-0.1-0.

00E-03 5.00E-04 0.00E+00-5.00E-04-1.00E-03 2.50E-03 2.00E-03 Fy 0 0.5 1 1.5 2 2.

00E-03 Fy 5.00E-04 0.00E+00 Fz -5.00E-04 0 0.5 1 1.5 2 2.

23 Fluid Force E E E-03 Fx 力 [N ] Fz Fy 時間 [s] Force_Y_C H IP Force_Z_C H IP 力 [N ] 1.00E E E E E E E-03 Fy 時間 [s] Force_X_W ire1 Force_Y_W ire1 Force_Z_W ire1 Fx Fz 力 [N ] 1.50E E-03 Fy 5.00E E+00 Fz -5.00E 時間 [s] Force_X_W ire2 Force_Y_W ire2 Force_Z_W ire ANSYS, Inc. All rights reserved. 23 ANSYS, Inc. Proprietary

25 Summary Filling / Packing / Mold Cooling / Deformation analysis about thermoplastic injection molding were performed using ANSYS CFX. Pressure and temperature variation during whole filling & packing process are predicted. Flowing, curing and wire sweep analysis of epoxy molding compounds for encapsulation semiconductors were performed. It shows that molding defects such as voids, wire sweeps, and so on could be predicted. ANSYS CFX have been proved to have the ability to perform the integrated simulation of injection molding and reaction injection molding. The innovate ideas in polymer process could be developed using ANSYS extensive physical models, efficient solver performance and flexible customization ANSYS, Inc. All rights reserved. 25 ANSYS, Inc. Proprietary

26 References 1. Masaki Yoshii, Yoshihiro Mizukami, Hideo Shoji Evaluation Technologies on Moldability of Epoxy Molding Compounds for Encapsulation of Semiconductors, Hitachi Chemical Technical Reposrt, No.40 (2003-1) 2. S. Nakao, H. Inoue, N. Maruyama, M. Takahashi, T. Uemura, K. Hirai, A Numerical Analysis of Injection Molding of Plastics Lens for PTV, AVC Company, Mutsushta Electric Industrial Co., Ltd. 3. Seiji Maruyama, Takahiro Okano, Application of Plastic-Injection- Molding Cmputer-Aided-Engineering Analysis, MAE Technical Report, 200 No Yoshinori Inoue, 3-D Flow Simulation of Runners in Injection Molding Using Remylop-Flow, R&D Review of Toyota CRDL Col.36 No.1, ANSYS, Inc. All rights reserved. 26 ANSYS, Inc. Proprietary

2008 International ANSYS Conference

2008 International ANSYS Conference Simulation of Added Mass Effect on Disc Vibrating in Fluid Using Fluid-Structure Interaction Coupling of ANSYS/CFX QIAN LIKE ANSYS Japan K.K. 2008 ANSYS, Inc. All rights