Friction analysis

[Anonymous]

Hello,

I'm trying to do a MES with non linear materials friction analysis as shown in the picture below.

It is a tube divided in two bodies, and a split part of a roller wheel. The lower surface of body 1, and upper surface of body 2 are in frictional surface contact using the Modified Coulomb friction law.

The constraints are:

• X's -> nodal Tx restriction
• Triangles -> surface Tx and Tz restrictions
• Circles -> surface fixed (all directions and rotation restrictions)

When I perform the non frictional analysis (without the horizontal force and adding a Tz restriction to the X's), the analysis goes ok and last about 15 minutes. But when trying to do the frictional analysis, the calculations take so long, in one hour it doesn't pass the first 5 percent of the analysis.

The question is: this is normal in this type of analysis? or it's a model problem? And if this is the case, how can I reduce the analysis time? Please, any help or experience will be well received.

Thanks!!

[John_Holtz]

Hi @Anonymous

I would say that it is normal for the analysis to be much slower with friction included.

The questions you need to investigate is these:

1. Are the results different enough with friction included to justify running the analysis for X days instead of 15 minutes without friction.
2. Is there a different way to include the effects of friction?

One thing to keep in mind: it is not possible to know how long it will take to solve based on 5% completion in 1 hour. It is possible that the analysis will speed up after the initial convergence problems. (It is also possible that it will run slower as time goes on. That is the nature of nonlinear analysis.)