Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Edge as a rigid domain in contact mechanics

Please login with a confirmed email address before reporting spam

In 2d, one can use a real 2d rigid domain or an edge (just a line) to represent a rigid domain. Which is better?


5 Replies Last Post 2 avr. 2024, 02:42 UTC−4
Henrik Sönnerlind COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 8 months ago 20 mars 2024, 08:54 UTC−4

A rigid domain is a material model that you select for a domain. If you want to achive the same thing using a rigid connector, you would have to select all boundaries around the domain (geometrically lines in 2D). In that case, you would still solve for all DOFs inside the domain, unless you remove it from the selection of the physics interface.

In almost all cases, using a rigid domain is better. You will, for example, not have an empty hole in your plots.

If you just want to make a certain boundary rigid, then it is a rigid connector that you should use. But this is a different representation of your physics than if you are using a rigid domain.

Either representation can be used for contact problems, depending on what you want to achieve.

-------------------
Henrik Sönnerlind
COMSOL
A rigid domain is a material model that you select for a domain. If you want to achive the same thing using a rigid connector, you would have to select all boundaries around the domain (geometrically lines in 2D). In that case, you would still solve for all DOFs inside the domain, unless you remove it from the selection of the physics interface. In almost all cases, using a rigid domain is better. You will, for example, not have an empty hole in your plots. If you just want to make a certain boundary rigid, then it is a rigid connector that you should use. But this is a different representation of your physics than if you are using a rigid domain. Either representation can be used for contact problems, depending on what you want to achieve.

Please login with a confirmed email address before reporting spam

Posted: 8 months ago 26 mars 2024, 01:32 UTC−4

Thank you very much Dr Sönnerlind. In the attached figure the long straight line representing the ground is what I refer to as the edge or line and what I wish to be rigid. Then I tried to add your rigid connector but I am not allowed to choose that straight line as the rigid connector, yet instead I can only choose the bottom line of the square. If the bottom line of the square isn't rigid, may I not use a rigid connector any more?

Thank you very much Dr Sönnerlind. In the attached figure the long straight line representing the ground is what I refer to as the edge or line and what I wish to be rigid. Then I tried to add your rigid connector but I am not allowed to choose that straight line as the rigid connector, yet instead I can only choose the bottom line of the square. If the bottom line of the square isn't rigid, may I not use a rigid connector any more?


Henrik Sönnerlind COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 8 months ago 27 mars 2024, 08:36 UTC−4

As I understand it, you want to simulate contact against a plane that is not really a part of your structural mechanics problem.

A quick (but not elegant) solution is to add a thin strip of material also below the long line, and proceed as for any contact problem.

But you can also just draw the line and mesh it (1 element is enough if it is straight), without having any physics attached. You can then use that line as source side in a contact pair. Note however that when the line does not have an adjacent domain, it does not have a natural 'outwards normal'. So for the contact search to work, the line must be drawn with the correct orientation (you want to have the solid on the left side of the line).

-------------------
Henrik Sönnerlind
COMSOL
As I understand it, you want to simulate contact against a plane that is not really a part of your structural mechanics problem. A quick (but not elegant) solution is to add a thin strip of material also below the long line, and proceed as for any contact problem. But you can also just draw the line and mesh it (1 element is enough if it is straight), without having any physics attached. You can then use that line as source side in a contact pair. Note however that when the line does not have an adjacent domain, it does not have a natural 'outwards normal'. So for the contact search to work, the line must be drawn with the correct orientation (you want to have the solid on the left side of the line).

Please login with a confirmed email address before reporting spam

Posted: 8 months ago 31 mars 2024, 01:37 UTC−4

But you can also just draw the line and mesh it (1 element is enough if it is straight), without having any physics attached. You can then use that line as source side in a contact pair. Note however that when the line does not have an adjacent domain, it does not have a natural 'outwards normal'. So for the contact search to work, the line must be drawn with the correct orientation (you want to have the solid on the left side of the line).

Thank you very much Dr Sönnerlind. By drawing a line, you don't mean using a rigid connector any more do you? Then I think a rigid connector can be used only when the bottom line of the square is rigid.

>But you can also just draw the line and mesh it (1 element is enough if it is straight), without having any physics attached. You can then use that line as source side in a contact pair. Note however that when the line does not have an adjacent domain, it does not have a natural 'outwards normal'. So for the contact search to work, the line must be drawn with the correct orientation (you want to have the solid on the left side of the line). Thank you very much Dr Sönnerlind. By drawing a line, you don't mean using a rigid connector any more do you? Then I think a rigid connector can be used only when the bottom line of the square is rigid.

Henrik Sönnerlind COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 8 months ago 2 avr. 2024, 02:42 UTC−4

That's correct. If there is no physics at all attached to the line, it will not move.

-------------------
Henrik Sönnerlind
COMSOL
That's correct. If there is no physics at all attached to the line, it will not move.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.