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.
Fail to evaluate variable. (material.epsilonr13)
Posted 4 mars 2014, 07:29 UTC−5 Plasma Physics, Modeling Tools & Definitions, Parameters, Variables, & Functions, Studies & Solvers Version 5.2 13 Replies
Please login with a confirmed email address before reporting spam
I would like to calculate the electric field within a chamber with gaseous xenon filled out all the space. In the meanwhile, there is another hollow cylinder inside the chamber which is made up of PTFE with 14 copper rings surrounding. So far, I have only used a physic "Electrostatics" to build this simulation. After adding the mesh type "Free Tetrahedral" to all the geometries, I kept receiving the following errors which prevented me from simulating the electric field.
Failed to evaluate variable.
- Variable: material.epsilonr13
- Geometry: 1
- Domain: 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 ......
(There are lots of domains included but I just leave them aside to make the message cleaner)
Failed to evaluate variable.
- Variable: comp1.es.epsilonrxz
- Defined as: material.epsilonr13
Failed to evaluate variable.
- Variable: comp1.es.chixz
- Defined as: comp1.es.epsilonrxz
Failed to evaluate variable.
- Variable: comp1.es.Px
- Defined as: (((comp1.es.Ez*comp1.es.chixz)+((comp1.es.chixx*comp1.es.Ex)+(comp1.es.chixy*comp1.es.Ey)))*epsilon0_const)
Failed to evaluate variable.
- Variable: comp1.es.Dx
- Defined as: (comp1.es.Px+(comp1.es.Ex*(comp1.es.Ixx*epsilon0_const)))
Failed to evaluate expression.
- Expression: d((-(comp1.es.Dx*test(comp1.Vx)+comp1.es.Dy*test(comp1.Vy)+comp1.es.Dz*test(comp1.Vz))*comp1.es.d)*(dvol),{test@4})
As for me, I think the most serious problem comes from the first error but I am not able to solve it.
I would really like to ask for your favor to give me some tips to solve these errors.
Thank you very much.
Sincerely,
Pin Jung
Please login with a confirmed email address before reporting spam
Regards,
Sven
Please login with a confirmed email address before reporting spam
I have been trying to attach the ".mph" file but it showed that there was a size error which made me unable to load the file. How can I attach the file then to avoid this problem?
Thank you very much.
Pin Jung
Please login with a confirmed email address before reporting spam
you can delete all meshes and all solutions.
If you are working with version 4.4, go to the "ribbon" at the Top
Mesh > Clear all Meshes
Study > Clear all Solutions
If you are working with earlier versions, go to the Main Menu and Edit > Clear Meshes / Clear Solutions.
Best regards,
That should create a much smaller file.
Sven
Please login with a confirmed email address before reporting spam
I think it should work now.
Thank you very much.
Pin Jung
Attachments:
Please login with a confirmed email address before reporting spam
a couple of comments on your model:
1) The error message arises from the fact that actually none of the domains is assigned to a materiual, therefore, among others, the material proiperty epsilon is missing. You have assigned your all your materials to *boundaries* which is not corrct.
2) In electrostatic simulations, the calculation of the fiels within metals should be avoided, as their high conductivity would make them anyway having a constant potential inside. See for instance our library models on capacitors.
3) You did a great job drawing this complex geometry, but you could have made your live much easier by using "array" functions that allow you to simply multiply entities.
4) In your model there are a couple of large domains that come very close to each other. Just pressing the mesh button would fail because the mesher may have difficulties in connecting large triangular meshes formed on large surfaced but coming very close together. You have chosen to make a very fine mesh
in general - whic is a bit of a overkill. Consider adjusting only the mesh on specific surfaces and mesh only the insulator domain(s) where the potential will be calculated.
I attach a "dummy" model with all of the above things exemplified. Note also how I have used "selections" with "adjacent" boundaries to easily address all surfaces of a certain body.
Best regards,
Sven Friedel
Please login with a confirmed email address before reporting spam
Please login with a confirmed email address before reporting spam
I am building a model for phosphorus removal. But it failed to evaluate the variable in the chemical reaction engineering module. Is one of you know how to solve this problem.
Thanks
Please login with a confirmed email address before reporting spam
After modifying the parts which caused problems that you told me above, I can now successfully calculate my model. However, there is still another question I would like to ask for help. Although I would like to calculate the electric field in the whole space within the chamber, it only shows the field of the domains inside of the chamber rather than of all the space. What I want to plot is the electric field distribution of the whole space of the 2D cut plan. Since I'd already defined all the space as material "Xenon", is there any other thing that might be wrong refers to this problem? Thank you very much.
Sincerely,
Pin Jung
Attachments:
Please login with a confirmed email address before reporting spam
I have changed the follwoing setings in your model:
1) Study 1 > Solver configurations A> Delete Solvers
This will, among other things, clear all old solution data sets. I want to have a fresh start with that model.
2) In Electrostatics: Select only your insulator Domains 1, 30-31, 65, 67. ONly there the field needs to be calculated as metals will cary the same potentials within their domains. Those are defined by their boundary conditions. (either fixed potentials or in case of passive domains, floating potential).
3) Add epsilonr = 1 (or whatever permittivity is correct) to Xenon. The material property is missing.
I hope this helps to clarify the matter. I send the model to your e-mail, as I for some reason cannot attache it here.
Best regrads,
Sven Friedel
Please login with a confirmed email address before reporting spam
I've tried to workout my model by comparing with the one you showed me. However, there is a question that how can I modify mine to let it show the results inside the chamber which looks like transparency as your model does but not only the field on the surface. I tried to add a "view" in "Definition" to indicate that some of the entities are hidden but it didn't help. I also defined a "selection" in "Results" -> "data sets" -> "solution 1" but it also seemed invariant. In contrast, there is nothing special defined in the model you showed me but it exhibits in this way apparently. What's the main idea of this part?
Thank you.
Sincerely,
Pin Jung
Attachments:
Please login with a confirmed email address before reporting spam
To generate cool plots for this model I think it ius essential to understand the differences of
volume plot, surface plots and slice plots.
So far you have used a volume plot which I think is not the best option, even if you add transparency with the transparency icon on the graphics window.
If you add a surface plot, the "inside" is not displayed at all because only the surfaces are sghaded which in your case have fixed potentials.
A much better view you obtain with slice plots.
In my opinion the best of two worlds you obtain when combining slice with surface plots - see my attached example.
There a many other ways for postprocessing of course - check out the postprocessing and advanced visualization videos in our video section to capture the concept.
www.comsol.com/video/browse/howto
Best regards,
Sven Friedel
Attachments:
Please login with a confirmed email address before reporting spam
I HAVE SOME PROBLEM REGARDING MATERIAL PROPERTY
HOW CAN WE REMOVE RED MARKED PROPERTY FROM PROPERTY LIST IF WE DON'T HAVE THAT VALUE?
PLEASE ALSO CHECK MODEL LIBRARY OF PIEZOELECTRIC SHEAR BENDER PROBLEM
IF WE FOLLOW THAT PROCEDURE IT DOESN'T WORK CORRECTLY
Please login with a confirmed email address before reporting spam
Undefined variable.
- Variable: comp1.dc.Re
- Geometry: geom1
- Domain: 1
Failed to evaluate Jacobian of expression.
- Expression: dvol*comp1.dc.Re
Failed to evaluate Jacobian of expression.
- Expression: (comp1.dc.Re*test(comp1.Ne))*(dvol)
Please explain
Regards,
Vignesh
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
- BLOG How to Obtain Oil Cost-Effectively with Multilateral Wells
- KNOWLEDGE BASE Failed to Evaluate Variable
- KNOWLEDGE BASE Computing Time and Space Integrals
- KNOWLEDGE BASE Failed to Evaluate Variable
- FORUM Model fail to converge