The increment size exceeds the remainder of the step and is decreased to

1

Hello,
I have been modeling a simple contact simulation with coefficient of friction and with surface to surface interaction.
I have tried to solve the problem with NLGEOM ON and OFF, as well as reducing the step increments and allowing more than INC 100 (I have tried with 1000).
In the CVG file, one can read the first step goes after several increments the force residual % and displacement correction % decreases to a low value. When this happens (and when I thought that the step will be finished) I receive the following message:
Convergence

***The increment size exceeds the remainder of the step and is decreased to 0.00000e+00***
***pressure ratio (smallest/largest pressure over all contact areas) = -7.204105e-04.***

then it says job done… but not results are stored on the .frd…

I have seen a similar thread here: Contact Pair doesn't work · Issue #54 · calculix/CalculiX-Examples · GitHub

I wonder what is the real problem, since for a contact convergence issues I would expect several cut backs or rigid body motions if contact has not been established, etc… but I have even tried to add some extra constraints to get a better posed problem…not success…It is worth mentioned that I am very used to contact simulations using Abaqus but I am just starting with contacts in calculix.

Any good advise you could give here @dhondt would be highly appreciated.
Thanks,
/J.

2

Can you share the .inp file ? It will be much easier to help and maybe identify a bug in CalculiX.

1 Like
3

Thanks for your prompt response. I can unfortunately not because it is a model with an NDA.

4

In such a case you could:

  • describe as many details of this simulation as you can disclose (type of structure and operating conditions or loadin scenario, boundary conditions, material properties and so on)
  • start removing features from the simulation to make is simpler until the error disappears (there might be a particular analysis feature causing it).
  • try to cause the same error in a simulation with different, simplified to minimum geometry (so that you can share it with us)
1 Like
5

Hi,

this is a normal message telling that the last increment is so big that it shoots beyond the final time of the step and therefore has to be decreased.

If you do not get results, something else must have happened. I suggest to run some small examples from the test suite to get experience.

Best Greetings,

Guido

2 Likes