Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

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.

"zero length" element

Please login with a confirmed email address before reporting spam

I am trying to model a bolt connection in an assembly. In order to do so I would like to place a node at the center of where the bolt hole should be (all bolt hole cutouts have been eliminated) and would like to generate a "zero length" element at that node by placing another node in the same spot. One node would be attached rigidly to ground, and the other one wuold be attached to the assembly geometry. I'd like to pull load outputs from the zero length element. Is this possible in Comsol? I, so far, cannot figure out how to manually place elements, or define an element length to be zero.

Thanks.

7 Replies Last Post 24 mag 2013, 10:32 GMT-4
Josh Thomas Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 20 mag 2013, 15:41 GMT-4
Roy-

How will the 2 nodes communicate, ie how are they connected? I guess it's not clear what you need the 2nd node for? Why can't you pull loads directly from the "assembly geometry" as you call it?

My first thought is to set up an assembly situation (rather than a form union operation). Then you can generate separate meshes that have duplicate boundaries (even though they lie on the same location in space).

Best regards,
Josh Thomas
AltaSim Technologies
Roy- How will the 2 nodes communicate, ie how are they connected? I guess it's not clear what you need the 2nd node for? Why can't you pull loads directly from the "assembly geometry" as you call it? My first thought is to set up an assembly situation (rather than a form union operation). Then you can generate separate meshes that have duplicate boundaries (even though they lie on the same location in space). Best regards, Josh Thomas AltaSim Technologies

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 22 mag 2013, 15:53 GMT-4
What I am attempting to model, for simplicity's sake, is a flange bolted to a plate. And what I ultimately want to get from the analysis are the loads at the bolt interface. What I want to do is eliminate the actual holes in the solid model and instead place a rigid node at the bottom surface of the flange at the center of where the bolt hole would be. There would be a coincident fixed node at the top surface of the plate. The "zero-length" element would be between the two so that I could pull load values from the model. Is there a better way to accomplish the same thing?

Thank you.
What I am attempting to model, for simplicity's sake, is a flange bolted to a plate. And what I ultimately want to get from the analysis are the loads at the bolt interface. What I want to do is eliminate the actual holes in the solid model and instead place a rigid node at the bottom surface of the flange at the center of where the bolt hole would be. There would be a coincident fixed node at the top surface of the plate. The "zero-length" element would be between the two so that I could pull load values from the model. Is there a better way to accomplish the same thing? Thank you.

Josh Thomas Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 22 mag 2013, 16:10 GMT-4
Roy-

I think using the "Form Assembly" at the end of Geometry sequence so that the flange and the plate are 2 separate bodies, then coupling the two bodies together at a coincident point using an "Identity Point Pair" will work for you. Make sure that both the Flange and the Plate have a point defined in the geometry sequence at the same location. Then, go to Definitions>Pairs>Identity Point Pair to select the two points which will be tied to have the same displacement.

In post-processing you should be able to probe for loads at this connection.

Best regards,
Josh Thomas
AltaSim Technologies
Roy- I think using the "Form Assembly" at the end of Geometry sequence so that the flange and the plate are 2 separate bodies, then coupling the two bodies together at a coincident point using an "Identity Point Pair" will work for you. Make sure that both the Flange and the Plate have a point defined in the geometry sequence at the same location. Then, go to Definitions>Pairs>Identity Point Pair to select the two points which will be tied to have the same displacement. In post-processing you should be able to probe for loads at this connection. Best regards, Josh Thomas AltaSim Technologies

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 23 mag 2013, 15:45 GMT-4
I have tried implementing your suggestion on a simplified geometry, but when I try to pull reaction forces from the points attached to the "plate," I get zero force. The coincident points on the "flange" have reaction forces at those points. So it appears that although I tied the points together as identity pairs, nothing is being translated from one to another. As far as I can tell, I've set them up properly (checked the help section, etc). Any thoughts?

Thanks again in advance.
I have tried implementing your suggestion on a simplified geometry, but when I try to pull reaction forces from the points attached to the "plate," I get zero force. The coincident points on the "flange" have reaction forces at those points. So it appears that although I tied the points together as identity pairs, nothing is being translated from one to another. As far as I can tell, I've set them up properly (checked the help section, etc). Any thoughts? Thanks again in advance.

Josh Thomas Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 23 mag 2013, 15:50 GMT-4
Roy-

It's hard to comment without the mph file to look at. Would you be able to post the simplified model at least?

-Josh
Roy- It's hard to comment without the mph file to look at. Would you be able to post the simplified model at least? -Josh

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24 mag 2013, 08:36 GMT-4
Josh, I've attached a simplified model of hte type of geometry I'm trying to model. I also find it interesting that even though the attachment points on the 2 pieces are attached, the model fails to resolve it I put constraints on the "plate" but will resolve if i put constraints on the "flange" (loads are applied to the flange). This seems to fall into the same category of the points not communicating with each other.

Thanks in advance for your help.
Josh, I've attached a simplified model of hte type of geometry I'm trying to model. I also find it interesting that even though the attachment points on the 2 pieces are attached, the model fails to resolve it I put constraints on the "plate" but will resolve if i put constraints on the "flange" (loads are applied to the flange). This seems to fall into the same category of the points not communicating with each other. Thanks in advance for your help.


Josh Thomas Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24 mag 2013, 10:32 GMT-4
Roy -

You may want to submit this to support because I think there might be something missing in using the Identity Point Pairs. You need to somehow specify the connection between the nodes under the physics interface (not only in the Definitions node). This is available for boundary pairs by right-clicking on the solid mechanics node and selecting Pairs>Continuity -- then you can link the displacements at boundaries from different parts in an assembly that touch boundaries. But, there is no corresponding way to do this for edges or points. I'm not exactly sure why not.

A work-around is to use the more general Identity Mapping function definition and then specify a Prescribed Displacement to the points on the second side, calling the displacements from the first side. See the attached mph for how to do this.

Also, I'm not quite sure why you are specifying zero displacement on the boundary between the flange and the plate. I'd think that you'd want this boundary to be free to move. The set-up I have solved in the mph file seems to make sense to me as to what you'd like. You can get reaction forces from both "sets" of points. The points on the flange and the points on the plate.

And, I added contact between the flange and the plate as you'll see. This makes more sense physically. Again, see attached mph file

Hope this helps some.

Regards,
Josh
Roy - You may want to submit this to support because I think there might be something missing in using the Identity Point Pairs. You need to somehow specify the connection between the nodes under the physics interface (not only in the Definitions node). This is available for boundary pairs by right-clicking on the solid mechanics node and selecting Pairs>Continuity -- then you can link the displacements at boundaries from different parts in an assembly that touch boundaries. But, there is no corresponding way to do this for edges or points. I'm not exactly sure why not. A work-around is to use the more general Identity Mapping function definition and then specify a Prescribed Displacement to the points on the second side, calling the displacements from the first side. See the attached mph for how to do this. Also, I'm not quite sure why you are specifying zero displacement on the boundary between the flange and the plate. I'd think that you'd want this boundary to be free to move. The set-up I have solved in the mph file seems to make sense to me as to what you'd like. You can get reaction forces from both "sets" of points. The points on the flange and the points on the plate. And, I added contact between the flange and the plate as you'll see. This makes more sense physically. Again, see attached mph file Hope this helps some. Regards, Josh

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.