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 The information provided may be out of date.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.

Parallel plate capacitance calculation problem

Please login with a confirmed email address before reporting spam

Hello,

I've been trying to model a parallel plate capacitor using 2D Axisymmetryic geometry using ComSol 4.1.

I've parametrized the whole model dimensions as well as the voltage to be applied on one of the electrodes. I've built the model geometry as follows:

1. A rectangle to represent the air
2. Vertically spaced electrodes to represent the electrodes. These electrodes were removed from the air box using the difference command to simplify the model
3. A Ground boundary condition was applied on the lower electrode
4. A terminal boundary condition was applied on the top electrode. The terminal type was "Voltage" with its value set as a previously defined parameter "V" whose value is 1 [V]

I'm facing the following problems:
1. When using the terminal voltage set as the parameter "V", the solution shows the voltage to be zero across all the solution region! Even at the terminal.
2. When setting the terminal voltage to an explicitly state value, say 1 V, the solution seems to be correct showing the expected solution of the problem. Yet, the value of the capacitance extracted from the model is far from the expected value by at least a scaling factor of 10^-9. I've double checked the geometry and made sure every thing is scaled correctly buy still in vain. I tried to change the model dimensions and I even sometimes got negative capacitance!

All your help is highly appreciated. I tried to download the model available on the model Exchange but it wouldn't open on Comsol 4.1

Regards,
Mohammed Omar

6 Replies Last Post Jan 7, 2011, 10:45 a.m. EST
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 7, 2011, 2:30 a.m. EST
Hi

for me it works rather nicely, but I'm not sure I have the same model as you ;)
take a look

--
Good luck
Ivar
Hi for me it works rather nicely, but I'm not sure I have the same model as you ;) take a look -- Good luck Ivar


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 7, 2011, 7:53 a.m. EST
Thanks a lot for your reply Ivar. :)

I've tried your model, and things work pretty well until I defined the terminal voltage in your model using a pre-defined parameter "V"; the simulation yielded zero voltage for all the domain when using this parameter. Your model with this modification is attached with the name "capa2DaxisTst_momar_1".

Also, attached is my model "parallel_plate_cap". The major difference from your model is that it includes a boundary box larger than the terminals so as to include the fringing field. The evaluated capacitance using my model is negative!

Thanks a lot for your help Ivar :)
Thanks a lot for your reply Ivar. :) I've tried your model, and things work pretty well until I defined the terminal voltage in your model using a pre-defined parameter "V"; the simulation yielded zero voltage for all the domain when using this parameter. Your model with this modification is attached with the name "capa2DaxisTst_momar_1". Also, attached is my model "parallel_plate_cap". The major difference from your model is that it includes a boundary box larger than the terminals so as to include the fringing field. The evaluated capacitance using my model is negative! Thanks a lot for your help Ivar :)


Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 7, 2011, 9:18 a.m. EST
Hi

but you call your "V" only "V", and "V" is already used by the dependent variable of your ES physics

Always use specific names, such as My_V or V_capa never single letter names as most are already used by COMSOL and COMSOl allows you to predefine them, in a 'parameter, this is not giving any error, mostly unexpected results ;)

--
Good luck
Ivar
Hi but you call your "V" only "V", and "V" is already used by the dependent variable of your ES physics Always use specific names, such as My_V or V_capa never single letter names as most are already used by COMSOL and COMSOl allows you to predefine them, in a 'parameter, this is not giving any error, mostly unexpected results ;) -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 7, 2011, 9:29 a.m. EST
Hello Ivar,

Thanks a lot for the prompt reply. I've modified the parameter name from V and now it simulated. This is really embarrassing :).

But I am still facing the problem of the wrong capacitance calculation. The capacitance is still evaluated as negative!

Thanks a lot for your help :)
Hello Ivar, Thanks a lot for the prompt reply. I've modified the parameter name from V and now it simulated. This is really embarrassing :). But I am still facing the problem of the wrong capacitance calculation. The capacitance is still evaluated as negative! Thanks a lot for your help :)

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 7, 2011, 9:47 a.m. EST
Hi

Well If I load your file, replace your V by V_ in the parameter and the Terminal node, then I have to delete your Solution
I'm with 4.1.0.112 you should use 4.1 and patch it. There are still some physics errors in the 4.1.0.88 not to say 4.0a)
I solve and it looks OK

I use an extra fine mesh I get C = 1.1887E-14, while epsilonr*pi*(2*R)^2/d = 1.11E-14 the difference being the fringe effect

--
Good luck
Ivar
Hi Well If I load your file, replace your V by V_ in the parameter and the Terminal node, then I have to delete your Solution I'm with 4.1.0.112 you should use 4.1 and patch it. There are still some physics errors in the 4.1.0.88 not to say 4.0a) I solve and it looks OK I use an extra fine mesh I get C = 1.1887E-14, while epsilonr*pi*(2*R)^2/d = 1.11E-14 the difference being the fringe effect -- Good luck Ivar


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 7, 2011, 10:45 a.m. EST
Thanks a lot Ivar :)

It works as a charm...

Thank you a lot for your support.
Thanks a lot Ivar :) It works as a charm... Thank you a lot for your support.

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.