Why is my model taking longer to converge even though convergence is good?

Yeshwanth_ · 2025-06-19T14:37:43+00:00

There was an error rendering this rich post.

I have given DT=0.1, DTMIN=0.0001, I have not specified DTMAX.

I don't why it's taking longer time to converge, Is it because of DTMIN?

I have given only enforced displacement SPCD=30mm. It's converging but it is taking longer time to get solve.

What is the preferred DT, DTMIN and DTMAX value. How the solver takes the time increments to get solve based on the DT, DTMIN and DTMAX values.

Can anyone help me out on this?

Find more posts tagged with

HyperMesh

OptiStruct

Altair HyperWorks Platform

Accepted answers

Rajashri_Saha

NLCTRAL gives better convergence, you can add this in Loadstep.

For your run time ussue, use TRACK=CONSLI for sliding contacts in LGDISP analysis.

All comments

Rajashri_Saha

Hi,

Do share the .out and nl.out file?

Are you getting any cut back? usually we suggest to go for DT=0.01 (NLPARM card), DTMIN=1e-6, DTMAX=0.1 in NLADAPT card. You can try with NLCTRL instead of NLPARM & NLADAPT.

One more suggestion is to use MPI run by giving -np, nt. Along with this you can use -incore command which takes more memory but run time will be less.

Thanks…

Yeshwanth_

Hi Rajashri,

Yes, I'm getting cutbacks.

I have attached .out and nl.out file, please let me know what's the issue?

Single-Limb_nl.out

Single-Limb.out

Yeshwanth_

Can you please tell me what's the purpose of NLCTRL and how to add it.

Rajashri_Saha

NLCTRAL gives better convergence, you can add this in Loadstep.

For your run time ussue, use TRACK=CONSLI for sliding contacts in LGDISP analysis.