Discussion Forum

Hello,

I have a little Problem with my project and I would be very glad to get some help. I want to model two blocks, which have the same size and are only displaced in vertical direction (small gap). The outer boundary of the lower object is fixed in all directions, the one of the upper object is free in vertical direction. In this direction a boundary load acts on this boundary, so the two objects should be pressed together. I used a contact pair between the two boundaries which should get in contact. The upper contact boundary differs from the lower one. I used Structural Mechanincs with the stationary solver, the geometry is finished as an assembly. I have not very much experiences with comsol (need this projekt for my bachelor thesis), so I have several questions.

1. I generate the objects by defining a surface (param. surface) and extruding. The outer boundaries have the profile of the surface, too. How can i made the outer one plane and parallel? Can i cut a part of the object? Or is there another solution by using extruding?

2. I'm not really sure if I use the contact pairs in the right way. May you please take a short look in my project to check if I chose the right preferences. It works with this preferences with two plane cubes without a gap. But for this example with different surfaces, I need to start with this little gap, don't I?
How can I determine which boundary is destination and which one is the source? the z-coordinate of the surfaces are fourier series with random coefficiant, but same average rougness

3. How fine should the mesh be (especially in the contact boundaries)? Should there be a ratio between element max. size in source and destination boundary (the help said >2)?

4. Do I have to chose a different solver? I use now the MUMS solver with default settings.

It would be realy nice if someone can reply at least to some of this questions. If you open the project, you won't see that the surfaces aren't plane. The unevennesses are very small compared to the dimensions of the blocks. If you want to visualize this rough surfaces, you have to increase the factor for the z-expression in lower and upper surface. Unfortunately the simulation is stopped because the maximum number of newton iteration is reached and it does not converge. Thank you very much for your help!

I wanted to upload the file direct in this forum, but is a little bit so large (5.87 MB). So it is uploaded under the following link (availible until 7/19, but I can re-upload it if necessary): gigamove.rz.rwth-aachen.de/download/id/47a4msbW4PgbC8 (I changed the max. element size of the mesh in the destination boundary to 0.01 mm after uploading it)