Edge contact between surfaces 2

[isponmo]

Dear all,

I am trying to model a 2D pin-hole problem on Patran and Nastran. I have already followed different approaches, such as using linear gap elements, rigid MPCs and sin distributed loads applied on the hole. Now, I would like to use the option "contact" that is provided with Patran, but so far I have not managed to make it work properly for an edge-to-edge contact between surfaces.

Does anyone know how I can do this?

Some extra information:
- The problem consists on a plate with a hole into which a circle-section pin is inserted.
- The pin is not glued no the hole.
- The pin transfers a vertical force to the plate.
- The problem does not admit symmetry boundary conditions.
- I am using cquad4 elements set up with Paver.
- If possible, I would like to solve it by using linear solution instead of non-linear solution. I have seen in some MSC tutorials that this approach is feasible.
- First, I would like to do a model using isotropic elements. Then, I would like to do move to an orthotropic laminate.
- Attached you can find a schematic drawing. The real problem is much more complex, but I think the idea is the same.

Thank you very much in advance.

Regards,

I. Pons

 

[franc11HD]

To solve this problem with linear contacts, you first need at least MD Nastran or Nastran 2012 (maybe 2011, I am not sure.) with the equivalent Patran wich has the linear contact enabled.

Contact problems are case by cases they do not work all the same way.

But basically, Loads/BC's, create contact deformable, target 2D elements. create two deformable bodies one fir the pin the other one for the plate. In the load case set up you just need to edit the contact table for touch contacts. That is the basic of it, but there are so many places where the analysis could fail.

hope this helps.

 

[isponmo]

Hello franc11HD,

First of all. Thank you for your reply.

I managed to do linear contact between two solids following the basic procedure, which coincides with the steps you described in the previous post. However, when the contact is between the edges of two surfaces (and not between faces), something goes wrong and the analysis fails or the results do not appear to make sense.

What do you mean when you say "Contact problems are case by cases they do not work all the same way."? What things should I take into account when modelling a contact? What are the most common mistakes?

Thank you,

I.Pons

 

[franc11HD]

Contact tolerance is one issue, contact bias could be another one, side of contact (1st vs 2nd, 2nd vs 1st, automatic) could be another one.

For 2D contacts, MSC Support told me to use discrete contact and not analytical contact. You can turn on also the segement to segement contact instead of node to segment.

You need to figure out what is wrong. Is the contact detected ? If not, then it could be contact tolerance or bias. If it you get stresses around the hole that do not make sense? If so, then maybe it is because you two parts are in interference (or at least Nastran sees them in interference) hence creating huge stress. (You can try to turn on stres free at initial contact in the contact table.)

That is why I am saying that it depends on the case and it depends on the error you get.