Discussion Forum

Hi everyone,

I have a problem understanding the behavior of deformed geometry module.

In my model I used the Heat transfer with Surface to Surface radiation module and the Deformed Geometry one. I need to represent two situations, in different parts of the model:

1. Simple translation of an object while modeling the heat transfer within it.
I used here the deformed geometry with a prescribed velocity for all the boundaries involved and I obtained the wanted result (the solution is 'attached' to the object and translate with it, as it would in the real world, evolving in time).

2. Mass removal with exchange of heat at the boundary where the removal occurs. In this case only one boundary is moved with a prescribed velocity (the boundary of removal).
However I observe the solution translating within the object and not being fixed: basically the temperature increase experienced is not removed when the correlated mass is removed (the heat diffuses faster if I increase the boundary velocity, which is not physical).

Moreover I noticed that sometimes, with no apparent reason, the problem reverses and I find good physical results for the second case and a sort of 'eulerian control volume solution' in the first case.

I found the documentation on deformed meshes confusing or rather not very 'illuminating' when having to do with the comsol interface. I already went through previous posts on similar topics but I find no one with such problem.

Should I stick with the deformed geometry or should I use the moving mesh in the first case? And for the second case (the moving mesh seems not the correct solution having mass variation)?
Is there a setting I am not aware of in the deformed geometry module to solve the problem or it is just a question of boundary conditions?

Thank you very much for any solution/helpful comment you may give me.

Amedeo