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)

Please login with a confirmed email address before reporting spam

Hello,

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

13 Replies Last Post 23 nov. 2016, 02:44 UTC−5
Sven Friedel COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 4 mars 2014, 08:54 UTC−5
Hi Ping Jung - can you send your model - I think we can fix that quickly.

Regards,
Sven
Hi Ping Jung - can you send your model - I think we can fix that quickly. Regards, Sven

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 4 mars 2014, 11:46 UTC−5
Dear Sven,

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
Dear Sven, 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

Sven Friedel COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 4 mars 2014, 12:07 UTC−5
Hi Pin Jung,

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
Hi Pin Jung, 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

Posted: 1 decade ago 4 mars 2014, 12:16 UTC−5
Dear Sven,

I think it should work now.
Thank you very much.

Pin Jung
Dear Sven, I think it should work now. Thank you very much. Pin Jung


Sven Friedel COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 5 mars 2014, 04:09 UTC−5
Dear Pin Yung,

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


Dear Pin Yung, 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

Sven Friedel COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 5 mars 2014, 04:11 UTC−5
I send the model to your personal email address.
I send the model to your personal email address.

Dr. Rajabu Hamisi Water and Land Resources Engineering

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 9 mars 2014, 19:01 UTC−4
Hello friend,

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
Hello friend, 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

Posted: 1 decade ago 14 mars 2014, 18:25 UTC−4
Dear Sven,

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
Dear Sven, 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


Sven Friedel COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 17 mars 2014, 06:08 UTC−4
Dea Pin Jung,

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
Dea Pin Jung, 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

Posted: 1 decade ago 19 mars 2014, 19:27 UTC−4
Dear Sven,

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
Dear Sven, 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


Sven Friedel COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 20 mars 2014, 05:27 UTC−4
Dear Pin Jung,

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
Dear Pin Jung, 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. http://www.comsol.com/video/browse/howto Best regards, Sven Friedel


Please login with a confirmed email address before reporting spam

Posted: 9 years ago 11 sept. 2015, 08:45 UTC−4
HELLO
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
HELLO 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

Posted: 8 years ago 23 nov. 2016, 02:44 UTC−5
Hi I am currently doing a 3 D analysis to find the electron distribution in plasma module. But COMSOl shows following error
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
Hi I am currently doing a 3 D analysis to find the electron distribution in plasma module. But COMSOl shows following error 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.