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.

determination of initial gap between contact tips

Please login with a confirmed email address before reporting spam

Hi,

I encountered a contact problem. The upper geometry is solid domain and the lower component is also solid domain, otherwise the intermediate component is fluid domain. The assembly of this geometry is " "union fluid-solid parts" and then "form assembly". The premise is that, I think, the contact problem is involved, "form assembly" is needed, fluid-solid interface should be de-doubled, first "union fluid-solid parts" is implemented.
If initially I set a tiny gap between contact tips, the first study step for solid mechanics module doesn't converge, even though I made various trials. But if I touch the contact tips in initial stage, it works well. But I didn't know the accuracy of this choice, whether it is reasonable or not.
Does anyone would interpret the difference for this difference and suggest one method for that?

Thanks in advance,

Zeng


6 Replies Last Post 21 set 2016, 17:45 GMT-4
Henrik Sönnerlind COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 8 years ago 19 set 2016, 04:59 GMT-4
Hi,

Note that an assembly is not a necessary prerequisite for a contact analysis. It only mandatory if the contacting parts are adjacent in the initial configuration (to avoid that they automatically form a union).

Regards,
Henrik
Hi, Note that an assembly is not a necessary prerequisite for a contact analysis. It only mandatory if the contacting parts are adjacent in the initial configuration (to avoid that they automatically form a union). Regards, Henrik

Please login with a confirmed email address before reporting spam

Posted: 8 years ago 19 set 2016, 11:23 GMT-4
Dear Henrik,

Thanks for your comments. If the contacting pair will be always in contact with each other, but the compressive stress and contact area are time-varying. Is it reasonable to "form union" at geometry setup? And how about the setup of initial gap between them, connection or tiny gap?

Thanks,

Zeng
Dear Henrik, Thanks for your comments. If the contacting pair will be always in contact with each other, but the compressive stress and contact area are time-varying. Is it reasonable to "form union" at geometry setup? And how about the setup of initial gap between them, connection or tiny gap? Thanks, Zeng

Henrik Sönnerlind COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 8 years ago 19 set 2016, 14:18 GMT-4
Hi,

If they are always in contact, then you would in principle need an assembly to avoid that the get 'welded' in the sense that the shear forces will be unlimited.

If forming an assembly is a problem, then you can make a small slit in the geometry so that Union can be used, and then use the Offset option in the Contact node to control the (mathematical) distance. Using offset is also a good way to control the initial gap, irrespective of whether Assembly or Union is used.

Regards,
Henrik
Hi, If they are always in contact, then you would in principle need an assembly to avoid that the get 'welded' in the sense that the shear forces will be unlimited. If forming an assembly is a problem, then you can make a small slit in the geometry so that Union can be used, and then use the Offset option in the Contact node to control the (mathematical) distance. Using offset is also a good way to control the initial gap, irrespective of whether Assembly or Union is used. Regards, Henrik

Please login with a confirmed email address before reporting spam

Posted: 8 years ago 20 set 2016, 12:32 GMT-4
I followed your advice and set the contact offset. The radius of adjacent sphere is R, the initial gap between them is set to a=0.01*R, then "form union" is used to assemble geometry. In Structural mechanics module, both destination contact offset and source contact offset are set to 0.01*a. I think it's really small compared to the radius of sphere. But when I run only structural mechanics, It still didn't converge. In segregation solver plot, contact pressure converge well, otherwise displacement suspend at 10^-2 a long iteration.
I didn't figure out the problem of this divergence. Is the value of contact offset reasonable in my setting?

I followed your advice and set the contact offset. The radius of adjacent sphere is R, the initial gap between them is set to a=0.01*R, then "form union" is used to assemble geometry. In Structural mechanics module, both destination contact offset and source contact offset are set to 0.01*a. I think it's really small compared to the radius of sphere. But when I run only structural mechanics, It still didn't converge. In segregation solver plot, contact pressure converge well, otherwise displacement suspend at 10^-2 a long iteration. I didn't figure out the problem of this divergence. Is the value of contact offset reasonable in my setting?

Please login with a confirmed email address before reporting spam

Posted: 8 years ago 21 set 2016, 12:42 GMT-4
Does anyone give me more clue on it? Even the contact offset are set, the contact of adjacent sphere didn't converge.
Does anyone give me more clue on it? Even the contact offset are set, the contact of adjacent sphere didn't converge.

Please login with a confirmed email address before reporting spam

Posted: 8 years ago 21 set 2016, 17:45 GMT-4
I posted the error plot in the attachment. Does anyone give any advice about it?
I posted the error plot in the attachment. Does anyone give any advice about it?

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.