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 The information provided may be out of date.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.

Breakdown voltage of parallel capacitor.

Please login with a confirmed email address before reporting spam

Hi there.

I am designing a simulation of the breakdown voltage of a parallel capacitor. So what I'm trying to figure out is, how the easiest way to to this is. For starters I guess it could be interesting to plot an I-V curve for the current between the two plates and then see where the current spikes. However, I am quite new at Comsol and I'm at a loss at how to make such a plot. I've designed the geometry and so forth, but how would I design the plot?

Anything helps.

Thank you

Tore.

7 Replies Last Post 13 nov. 2012, 01:25 UTC−5

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 8 nov. 2012, 03:25 UTC−5
Hi,

I don't think there is a straightforward way to model dielectric breakdown in COMSOL. You would need to implement the set of equations describing this extremly nonlinear process.
However you can model the electric field in the dielectric and see where it exceeds the break down field strenght of the material. However this doesn't give you a I-V curve.

Cheers
Edgar
Hi, I don't think there is a straightforward way to model dielectric breakdown in COMSOL. You would need to implement the set of equations describing this extremly nonlinear process. However you can model the electric field in the dielectric and see where it exceeds the break down field strenght of the material. However this doesn't give you a I-V curve. Cheers Edgar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 8 nov. 2012, 07:47 UTC−5
Thanks for your response Edgar.

I have more or less concluded that you are right. I think I can live with the solution with the electric field, and then do parametric sweeps on the voltage and the gap to get me an indication of where the breakdown is exceeded.

Thank you

Tore.
Thanks for your response Edgar. I have more or less concluded that you are right. I think I can live with the solution with the electric field, and then do parametric sweeps on the voltage and the gap to get me an indication of where the breakdown is exceeded. Thank you Tore.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 8 nov. 2012, 09:07 UTC−5
Tore,

and even then you must pay a lot of attention to the edges and corners in the model because of the field enhancements there.
So, try to avoid sharp corners and edges and take care to mesh fine enough. However, you will probably gain some insight into your problem even without directly modeling the breakdown.

Cheers
Edgar
Tore, and even then you must pay a lot of attention to the edges and corners in the model because of the field enhancements there. So, try to avoid sharp corners and edges and take care to mesh fine enough. However, you will probably gain some insight into your problem even without directly modeling the breakdown. Cheers Edgar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 9 nov. 2012, 09:46 UTC−5
Dear Edgar.

Thank you for your answer again.

I am aware of the issues in terms of the corner and I will delve into the issues. I'll let you know what I find out.

Best regards

Tore
Dear Edgar. Thank you for your answer again. I am aware of the issues in terms of the corner and I will delve into the issues. I'll let you know what I find out. Best regards Tore

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 10 nov. 2012, 05:01 UTC−5

Tore,

and even then you must pay a lot of attention to the edges and corners in the model because of the field enhancements there.
So, try to avoid sharp corners and edges and take care to mesh fine enough. However, you will probably gain some insight into your problem even without directly modeling the breakdown.

Cheers
Edgar


Hi agan Edgar.

I'm trying to do a 2D simulation of the problem at the moment. But while I'm trying to use fillets for smoothening the corners, I still can't completely optimize the issue of localized concentrated charge. If you see the model I've uploaded, you'll notice that for the fillet on the left hand side of the model, there still is a large concentration of charge because the fillet still amounts to a 90 degree angle onto the substrate. How can I best solve this?
[QUOTE] Tore, and even then you must pay a lot of attention to the edges and corners in the model because of the field enhancements there. So, try to avoid sharp corners and edges and take care to mesh fine enough. However, you will probably gain some insight into your problem even without directly modeling the breakdown. Cheers Edgar [/QUOTE] Hi agan Edgar. I'm trying to do a 2D simulation of the problem at the moment. But while I'm trying to use fillets for smoothening the corners, I still can't completely optimize the issue of localized concentrated charge. If you see the model I've uploaded, you'll notice that for the fillet on the left hand side of the model, there still is a large concentration of charge because the fillet still amounts to a 90 degree angle onto the substrate. How can I best solve this?


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 12 nov. 2012, 03:47 UTC−5
Tore,

to my taste it doesn't look too bad. You can try to use a finer mesh,

Cheers
Edgar
Tore, to my taste it doesn't look too bad. You can try to use a finer mesh, Cheers Edgar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 13 nov. 2012, 01:25 UTC−5

Tore,

to my taste it doesn't look too bad. You can try to use a finer mesh,

Cheers
Edgar


I actually tried with a finer mesh, but for some reason it actually only seems to make it worse. Something tells me a more coarse grid will be better, but I'm yet to look into it.
[QUOTE] Tore, to my taste it doesn't look too bad. You can try to use a finer mesh, Cheers Edgar [/QUOTE] I actually tried with a finer mesh, but for some reason it actually only seems to make it worse. Something tells me a more coarse grid will be better, but I'm yet to look into it.

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.