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.
Doubt regarding periodic boundary condition error
Posted 2 sept. 2020, 07:11 UTC−4 Electromagnetics, Mesh, Studies & Solvers Version 5.5 6 Replies
Please login with a confirmed email address before reporting spam
Hello,
I am trying to make a calendered mesh and supply some electric voltage to it to see electric field contours in the nearby region. Instead of making the full (large) mesh, I made a unit cell and thought of providing periodic boundary conditions. But I get this error when I compute the solution to the problem:
"- Feature: Compile Equations: Stationary (sol1/st1)
Error in multiphysics compilation.
Transformation does not map source onto destination.
- Boundary similarity: dst2src_pc2_sim5
- Detail: Source and destination are not compatible."
Please let me know what to do. I know why the error is coming, but I do not know how to fix it. I also tried building the mesh myself, to suit the periodic boundary condition, but that isn't working too. I have uploaded my comsol file for reference.
Attachments:
Please login with a confirmed email address before reporting spam
Hmm. This appears to be an electrostatic model. I've noticed that you appear to be applying potentials to complicated surfaces that fully enclose good conductors (copper). Not only does this increase your mesh count a lot, but there is no purpose that I can think of in attempting to compute electrostatic (note the word "static" here!) fields inside copper! Those fields are zero. Now, you can specify potentials on those metal surfaces if you like, but you should subtract (i.e, use the boolean difference operation) all of the copper material parts from your model entirely, keeping only (as computational volumes) those regions that are non-conductive, along with surfaces that are PECs (perfect electric conductors) to represent the copper. Likewise, you will only mesh the aforementioned volumes and surfaces. You will not mesh the interiors of the copper parts, since the fields are zero there and literally irrelevant to the computation. This approach should also ease your accurate periodic mapping of the meshes between faces, since the copper faces on the periodic boundaries are not to be meshed anymore. That said, you may still have problems remaining, but you will be much closer to having a good model. So I suggest you try all that, and then, if you still have issues, post your updated/improved model (and any updated questions) again. Good luck!
-------------------Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
Please login with a confirmed email address before reporting spam
Ok @Robert, I made some changes and it looks like the previous error isn't arising anymore. But I'm now getting:
Failed to find a solution. Singular matrix. Returned solution is not converged. Not all parameter steps returned.
I've attached the new file, could you have a look into it and let me know if you can fix the problem? It'd be of great help. Thank you.
Attachments:
Please login with a confirmed email address before reporting spam
Ok, I almost never do this, but I found your problem's odd failure to solve intriguing, so I played around with it a bit, making many changes, until I got it to solve properly. The attached model yields physically-reasonable results. Among the changes, you will notice that I simplified/ removed some unnecessary features (i.e., items that would make no difference to the basic calculation). You may wish to put those things back, if you have plans for them to actually matter to the physics later on. But, if you do that, be sure to put them back piece by piece, making sure that the model still executes, as you do so.
Enjoy!
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
Attachments:
Please login with a confirmed email address before reporting spam
Ok it is working now. I truly am very thankful for this Robert.
P.S - I know I've troubled you enough, but I just want to ask one last thing. I got a warning after I computed the solution. "Inconsistent unidirectional explicit constraints are not supported". Should I be worried of this?
Please login with a confirmed email address before reporting spam
I got a warning after I computed the solution. "Inconsistent unidirectional explicit constraints are not supported". Should I be worried of this?
Maybe! I hadn't checked/noticed that in the log, but I see it now. Evidently, there are still some subtleties I missed in specifying the periodic boundary conditions. I suggest you experiment with the various settings some more. And, perhaps another user can take a look at the file and offer a suggestion?
-------------------Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
Please login with a confirmed email address before reporting spam
I got a warning after I computed the solution. "Inconsistent unidirectional explicit constraints are not supported". Should I be worried of this?
Maybe! I hadn't checked/noticed that in the log, but I see it now. Evidently, there are still some subtleties I missed in specifying the periodic boundary conditions. I suggest you experiment with the various settings some more. And, perhaps another user can take a look at the file and offer a suggestion?
Sure, thanks for you help anyways Robert.