Which material model should be used in OptiStruct for EPS foam?
Altair EmployeeHi all,
Expanded Polystyrene (EPS) is usually used as a package material. A quasi-static compression test was done to test its material behavior. In order to validate the material stress-strain curve from the compression test, a simulation should be done. The velocity of the compression test was about 5mm/min, so I think a static nonlinear analysis in optistruct may be proper. However, I don't know which material model is suitable for EPS foam material which has both hyperelastic and plastic behavior. Any suggestions?
<?xml version="1.0" encoding="UTF-8"?>
Roy
Answers
-
Hi Roy,
You can use MATX33, 70 or 77.
Instead, of OptiStruct you can prefer RADIOSS.
0 -
Altair Forum User said:
Hi Roy,
You can use MATX33, 70 or 77.
Instead, of OptiStruct you can prefer RADIOSS.
1. From the reference guide of MATX33, the yield stress vs. volumetric strain curve should be defined. The difficult is how to get the curve especially the volumetric strain in engineering. Usually, we can only get the engineering stress-strain curve through uniaxial compression test. How to convert this stress-strain curve to stress-volumetric strain curve?
2. For MATX70, it can consider the strain rate effect. But this material is elastic without plastic, right? That's, when the loading release, the shape change to its original location without plastic deformation.
3. Since my test is quasi-static, the total testing time is about 10 minutes. Do you think RADIOSS is proper for this problem? How to control the CPU time?
What's more, could you please provide some examples or documents on how to use MATX33, MATX70 for foam material?
Thank you
Roy
0 -
Convert the engineering curve to true curve and use the same.
I am attaching a list of material cards for foam (RADIOSS)
<?xml version="1.0" encoding="UTF-8"?>
You can use MATX38 for instance,
0 -
1. I don't think the stress vs volumetric strain curve is the same as the normal true stress strain curve.
2. As far as I know, for this long time duration quasi-static problem, Radioss solver may be not appropriate. The CPU time would be very long.
Roy
0 -
Opti NLGEOM converts to RADIOSS Block format internally and runs RADIOSS in the back ground
0
