Which are the differences in contact solution Abaqus 6.10 vs 6.14?

[victorroda]

Hello all,
Some time ago I solved a transient thermo-mechanical problem using Abaqus 6.10, and I was quite satisfied with the observed results (frictional dissipation energy and power loss due to friction).
Now, I am solving the very same problem with Abaqus 6.14 (using the very same *.inp file), but the obtained results are quite different from the ones I got in first instance.

Any one has any idea why this could happen?

I provide graphical results of the problem, and it can be observed that:
[ul]
[li]The frictional dissipated energy (ALLFD) is quite different from one case to the other (and so it is the power loss).[/li]
[li]The power loss (delta_ALLFD/delta_TIME) is much more smooth in the results obtained with Abaqus 6.10 than the ones obtained in Abaqus 6.14.[/li]
[li]Work of the external forces (ALLWK) and internal energy (ALLIE) are quite similar[/li]
[li]Total energy (ETOTAL) is similar in both cases, but there is more variation in 6.10 than in 6.14 (it is supposed to be constant).[/li]
[/ul]
Since the frictional dissipated energy depends on the coefficient of friction, slipping velocity and contact pressure, I assume that there is some kind of issue with the contact pressures

Link to file: Link

 

[IceBreakerSours]

I believe how the area gets computed changed in one of the versions which will have an influence on the contact pressure. Check the documentation of the two versions to be sure.

*********************************************************
Are you new to this forum? If so, please read these FAQs:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[victorroda]

Dear IceBreakerSours,

I really appreciate your response.

After some research, I have found that the difference between the results in ABQ6.10 and ABQ6.14 is caused by the contact constraint discontinuity work (CCDW).
When CCDW is summed to the frictional dissipation (FD) in ABQ6.14, a similar results as the one obtained for FD in ABQ6.10 is obtained. To be honest, it does not seem coincidence (see attached results). And this is also extended to the power loss calculations.
What I am not sure at this point is the reason why this happens, and two questions arise:
Was CCDW included in the FD results in ABQ6.10?
Was ABQ6.10 using a contact formulation where no work was dedicated to the contact constraints?

 

[IceBreakerSours]

Nice catch. There is an example in the documentation that discusses the importance of contact constraint discontinuity work (ALLCCDW or something like that). I guess contact developers at Simulia must have realized the importance of further breaking down the contact history output in terms of physical and non-physical components after v6.10.

You could try to reduce the max. time step and the run job in both versions to see how close the results look. ALLCCDW should approach zero as time increment goes to zero (and, I suspect, as the mesh is refined as well).

*********************************************************
Are you new to this forum? If so, please read these FAQs:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083