Contact problem-job will not run-no idea how to fix the errors

[nashaeroeng]

I have only been using Abaqus for a few months and need to simulate a contact problem where a normal load is applied to a long thin test-piece using a hemispherical roller. The problem includes friction in the analysis and is being modelled as quasi-static loading. I have looked at several worked examples and tried to model the problem countless times over the last week to no avail. I keep getting the same three errors:

Too many attempts made for this increment;

The required increment is less than the minimum specified;

...And an error that involves zero pivot

I have tried to fix each error but nothing I do seems to work.

I have included an image of the model (hopefully I uploaded it correctly) I am trying to make, the test piece, which is a 2D shell, is going to be pulled upwards and can only move in the y direction, the roller, which is a 2D analytical rigid body, can only move along the x axis.

Hopefully someone can help me out?? If more information is needed about the problem, I'll be happy to include it!

 

[DanStro]

Does it get through any increments or is it stopping on the first one?

Here are a some of ideas, for what they're worth:

It's hard to tell in the image but it kind of looks like you roller is interfering with the plate. If that's true then you might want to change that. Unless there is a temperature change I usually model the parts in contact so there is no severe discontinuity happening when the run starts.

Some other things that you want to check is the interactions because it sounds like you want to make sure to NOT be using small sliding.

Also, are you using Automatic stabilization in the step? I've found that this normally helps a lot with contact problems.

Dan

 

[nashaeroeng]

Hi Dan,

It usually goes through a few increments, but there has been times when it stops after the first one.

I'm not taking any temperature into account at all, this was meant to be a very basic model to give a general idea of the test before going into more detail....but it's turned out not to be very basic at all!!!!

I started using automatic stabilisation after my first few attempts but that didn't make a difference. Could you tell me what you mean as well by how the roller is interfering with the plate?

Also....is it possible that this could be caused by the BCs??

I haven't been using small sliding. Also when the error about the increment being less that what is required comes up I reduce the minimum but this makes no difference either!

 

[DanStro]

I don't know how to attach an image (that doesn't have a URL) so I can't upload a marked up image. But I'll try to explain it. If you zoom in on the image that you attached it looks like the vertex of the roller is inside of the plate. It could just be because of the quality of the image or there is another edge in the foreground but I thought I should mention it.

 

[nashaeroeng]

Ah ok cool, so how should i fix this? Should i start with the roller at some clearance from the testpiece?