Error Help

[a2joe]

Hi Guys,

I'm new to Abaqus and I was hoping someone could help me out with the following error that I keep getting;

"Too many attempts made for this increment"

Lowering the Minimum Increment size doesn't have any effect. I can't think of anything else.

Any help would be appreciated.
Thanks

 

[rstupplebeen]

You can also try lowering the maximum increment. Are you doing a nonlinear analysis (contact, large deformation, materials properties...)? Do any steps converge? More information on your specific model would allow me to help more.

Rob Stupplebeen

 

[a2joe]

Thanks for the reply.

I tried lowering the max increment, but still get the same error. I am trying to run a sphere-plate contact (surface-to-surface contact) model. Both parts are 3d deformable shells (i tried solids at first but wasn't getting good results) with specified thickness and material properties. I have applied a rigid body constraint to the plate which is fixed. The sphere has a negative vertical displacement to simulate penetration. I wanted to compare Abaqus results with theoretical so I can check if my model was right

 

[GPapazafeiropoulos]

Dear a2joe,

Are you doing dynamic analysis in ABAQUS?

You can look at the *CONTROLS or the *CONTACT CONTROLS options in the Abaqus Keywords Reference Manual.

The *CONTACT CONTROLS option provides additional optional solution controls for models involving contact between bodies. Be very very careful with the parameters you are going to alter.

Be aware that the *CONTACT CONTROLS option must be used in conjunction with the the *CONTACT PAIR option in Abaqus/Explicit analyses.

The *CONTROLS option is more general and you can modify various parameters with this. The parameter ANALYSIS=DISCONTINUOUS would be a good initial option, but if it doesn't work, then you have to reset some convergence parameters in the *CONTROLS option.

I used ABAQUS for performing dynamic analyses in cases of earthquake-loaded soil layers, lying on rigid bedrock and I encountered serious problems, watching a notification "Too many attempts made for this increment" written in the .msg file. This can be corrected in the way I stated above, but I want to point that you must be careful because excessinely weak convergence criteria can impair the program's results so that they won't be reliable any more...

Best regards,

George Papazafeiropoulos
_______________________________________
First Lieutenant, Hellenic Air Force
Civil Engineer (M.Sc.), Ph.D. Candidate

 

[a2joe]

Thanks George,

I'm doing a static analysis.

Rob, how do I check for convergence??

Thanks

 

[Elabbasi]

This Abaqus message tells you that the analysis is not converging. There are so many potential causes for this including insufficient constraints, time step too big, unstable model, material failure, and contact.

You should look at the information Abaqus provides in the .msg and .dat files. There is a lot of info there to help you find the specific cause of non-convergence. It can help you find mistakes in your model. Probably the most important indicator is whether there were any converged increments or not.

If the info in these two files does not help you can reduce the nonlinearity of the problem (temporarily) to help find the cause of the problem. For example, you can switch to frictionless contact, or even small sliding contact. If you have nonlinear materials, you can also convert them to linear elastic.


Nagi Elabbasi

http://www.veryst.com