Can it be disabled or reduced ? Posting many images is the standard here. Unless it’s only for new users and stops working after some time.
@Calc_em Getting that stupid error message again? I was trying to share the Google Drive link: “An error occurred: Sorry you cannot post a link to that host. usp=sharing”
It’s only for new users and a default setting.
Ok, I guess we can live with that then.
1 Like
What do you mean ? It doesn’t converge ?
It fails. I open the same PMX from yesterday, no changes, but result is:
average force= nan
time avg. forc= 0.010000
largest residual force= 0.000000 in node 1429 and dof 3
largest increment of disp= 4.646660e+03
largest correction to disp= 8.506872e+02 in node 201 and dof 2
Job failed - no results exist.
Process elapsed time: 46.891 s
Maybe you should try resubmitting the analysis and checking if it’s the correct version of the model. And look for the .frd file from yesterday in your work directory - it should still be there.
Oh sorry …
*HYPERFOAM, N=2
0.02, 3.0, 0.02, 3.0, 0.05, 0.05
1 Like
Good news - CalculiX devs are already working on a fix for this bug: Hyperfoam material model showing incorrect deformation with no stresses · Issue #188 · Dhondtguido/CalculiX · GitHub
It most likely won’t help with convergence issues, though. So it’s necessary to have reasonable constants for this model.
1 Like
@Viertelstein , I don’t know from where you have your material parameters for your foam, but as described by @Disla HYPERFOAM Model Issue - #29 by Disla when you insert twice the material parameters, you will also double the energy in the structure, so to correcting this issue, you will need to half the material energy parameter why these values will be more correct
*HYPERFOAM, N=2
0.01, 3.0, 0.01, 3.0, 0.05, 0.05
compared to these values which will double the energy content
*HYPERFOAM, N=2
0.02, 3.0, 0.02, 3.0, 0.05, 0.05
Yeah, the equation for the strain energy potential in this model is:
where N is the parameter that you can change on the keyword data line. So, when you increase N, you also increase the number of terms added in this sum and thus add more strain energy.
1 Like
Just make sure that you apply concentrated force only to reference points, not to entire faces.
Tied contact is fine, but tie constraints usually works better.
This is 1.9x scaled deformation, check true scale. But maybe your contact stiffness is too low.
1 Like
Sorry, I confused Hyperelasticity with Hyperfoam. The phone has a small screen.
I tried different settings in SURFACE INTERACTIONS. TIED with different values (which still works best in generell), HARD, LINEAR and EXPONENTIAL, all fail sooner or later.
TIED still shows overlapping geometries and a “finished job” without reaching TOT TIME 1.0 (is this because of CONCENTRATED FORCE (=no specified penetration value)?
Displacement control would be better, at least for convergence. But it shouldn’t penetrate so much with force control either. Make sure that contact is properly defined - correct surfaces are selected, master has a coarser mesh and belongs to the stiffer or rigid part. If it still doesn’t work, please share the file here. Maybe it’s still an issue with this material model.
One big mistake from my side! I have to mark ALL surfaces for propper contact! Together with 10 mm mesh for the stamp and 3 mm for the latexblock it looks pretty good!
Yeah, that would definitely explain it. Contact regions shouldn’t spread too far, but they must cover all areas that may touch during the deformation.
Assuming that your step time is the default 1 s (better check that), it might be an issue with reading the CalculiX results in PrePoMax. Sometimes this may happen. You can try rerunning, check the CalculiX log or even run the analysis from the .inp file outside of PrePoMax (using standalone CalculiX).
But how did you load it in the analysis ? So far you’ve been using prescribed displacement.