Discussion Forum

Hey everybody!

My goal is to simulate the stresses during a load cycle in a bearing shell.

The problem:
I assembled a journal bearing in con-rod from which i cut off the major part so that the calculation time can be reduced and defined this as union. So I don't have to bother about contact problems. On the surface of the journal bearing I solve the hydrodynamic with help of the thin film module and a PDE which handles the equilibrium of an external force. On the top of the con-rod I fix the structure what seems to me as the best way. Assuming a rigid structure in the first case the simulation works fine until it reaches the physical borders when the time dependent load becomes to high and the fluid film breaks. Therefore I coupled the hydrodynamic with elastomechanic so that the bearing shell can deform and a higher load carrying capacity is possible.
Due to initial value problems I set up an extra study where I calculate an inital status, which acts as an inital solution for instationary time dependent main study.


I made first tests with small loads, checked how to link both physics best,... and it worked fine. Now I have the problem, that I can't reach convergence with the load I'd like to simulate.
The calculation takes very long and it stops at 4 %. Until this point the convergence behaviour looks very good but then it changes, see attached screenshots, and I can't figure out why. In my opinion its not a physical problem because at the time, when the calculation stops the maximum fluid pressure is about 20 bar, the minimum filmthickness at about 20µm whereby the stresses are quite low. I think the problem must be the way the solvers work. I dont have enough experience to cope with this kind of problem. Thats why I would like to ask help from u.

I add some screenshots of the model so u can imagine how it looks like and a screenshots of one of the two solvers. (both have the same specification)


With kind regards,
Philipp