Menu

Discussion Forum

New Discussion
フィルター

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 The information provided may be out of date.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.

Mesh nodes at rotating boundary not aligning

Posted 2019/03/07 15:13 GMT-5 Fluid & Heat, Computational Fluid Dynamics (CFD), Mesh Version 4.0a 5 Replies

Donal Murphy
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hi,

I'm using an older version of comsol and am trying to model quite a simple problem. The geomerty comprises of a representation of an offeset rotating impeller. I'm trying to model the rotation of the impeller using a rotating domain using the rotating machinery physics. I'm using form assembly. As a result the two domains share ab identity pair which I've indicated with a flow continuity node. However when I go do mesh the entire domain the mesh nodes do not aline. Aswell it seems that the two domains overlap in places and don't cover free space in other places. I've included the full model and screenshots below. Any advice is much apperciated.

Thanks, Dónal

5 Replies Last Post 2019/03/07 17:20 GMT-5
Edgar J. Kaiser Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 2019/03/07 15:26 GMT-5

Donal,

it is the essential feature of an identity pair that mesh nodes do not need to align. The physical fields are interpolated from one side to the other. The gaps and overlaps are a visualisation effect resulting from the mesh view.

Cheers Edgar

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Donal, it is the essential feature of an identity pair that mesh nodes do not need to align. The physical fields are interpolated from one side to the other. The gaps and overlaps are a visualisation effect resulting from the mesh view. Cheers Edgar
Donal Murphy
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 2019/03/07 16:16 GMT-5
Updated: 7 years ago 2019/03/07 16:41 GMT-5

Hi Edgar,

Thanks for the help but my model still won't compute. The error message says "failed to find consistant initial values". I then place a step function over the first two seconds however the solver now runs but without ever converging. There still must be something undefined in my model even though I think I've accounted for everthing. Any idea's ?

Thanks

Hi Edgar, Thanks for the help but my model still won't compute. The error message says "failed to find consistant initial values". I then place a step function over the first two seconds however the solver now runs but without ever converging. There still must be something undefined in my model even though I think I've accounted for everthing. Any idea's ? Thanks

Attachments:

Edgar J. Kaiser Certified Consultant
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 2019/03/07 16:39 GMT-5

Donal,

sorry we don't do CFD. We are using Moving Meshes in electromagnetic and structural contexts, among others. Hopefully, some other forum users can help out. You may also find examples in the application library.

Cheers Edgar

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Donal, sorry we don't do CFD. We are using Moving Meshes in electromagnetic and structural contexts, among others. Hopefully, some other forum users can help out. You may also find examples in the application library. Cheers Edgar
Donal Murphy
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 2019/03/07 16:52 GMT-5

Thanks Edgar for your help.

Thanks Edgar for your help.
Donal Murphy
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 2019/03/07 17:20 GMT-5
Updated: 7 years ago 2019/03/07 17:21 GMT-5

Update : I am now sure that the problem lies at the boundary of the two domains, as the resulting plot indicates. Anyone know why the identiity pair boundary isn't being identified by comsol ?

Update : I am now sure that the problem lies at the boundary of the two domains, as the resulting plot indicates. Anyone know why the identiity pair boundary isn't being identified by comsol ?

Attachments:

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.

Suggested Content