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.

Applying a +/- 10 V sinusoidal signal to electrodes in air

Please login with a confirmed email address before reporting spam

Hi there

Been playing around with COMSOL in an effort to learn how to use it.

I have some success modelling my little arrangement with the electrostatics module:

a row of electrodes in the middle, surrounded essentially by a cage of ground electrodes that are 30 mm away (imagine a rectangle, with some electrodes in the center then grounds around the perimeter).

Now what I want to do is switch the electric potential signal I had, for one you might get out of a signal generator. Namely a 10 V (20 V peak to peak) Sinusoidal signal of a frequency of my choosing. How do I actually implement this now?

Do I need to change from electrostatics to electromagnetic waves from the RF-module? Or do I stay with electrostatics and do a frequency domain study?

I did try doing electromagnetic waves, but it only allows you to specify an electric field strength (V/m) and a frequency and anyway, I ended up with an error about the relative error being greater than the relative tolerance. What ever that means.

Any help would be much appreciated!

(I can attach the file later if necessary, I am not actually near the computer with the model on at the moment)

5 Replies Last Post 14 sept. 2012, 16:07 UTC−4

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 14 sept. 2012, 10:34 UTC−4
Hi Alex,

if your geometry is 10 to 100 times smaller than the wavelength of the signal you are planning to use, you can stay with AC/DC and do a frequency domain study.
If wavelength and geometry are same size or geometry is bigger you must change to RF.

Cheers
Edgar
Hi Alex, if your geometry is 10 to 100 times smaller than the wavelength of the signal you are planning to use, you can stay with AC/DC and do a frequency domain study. If wavelength and geometry are same size or geometry is bigger you must change to RF. Cheers Edgar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 14 sept. 2012, 10:49 UTC−4
Thanks for that Edgar, thats a useful tip.

The spaces between my electrodes are like 25-30mm so a 10 V sinusoidal signal of 100 kHz to 1 MHz gives wavelenghts of 300m to 3km so I think it looks like my geometry is a lot smaller. Its 10,000 times smaller than the wavelength at least.

Would that also be a problem? I mean it does work in real life, I put exactly those signals in to a geometry like that and measured current waveforms. (I hope, this might be point where I find out I have been measuring something completely different).
Thanks for that Edgar, thats a useful tip. The spaces between my electrodes are like 25-30mm so a 10 V sinusoidal signal of 100 kHz to 1 MHz gives wavelenghts of 300m to 3km so I think it looks like my geometry is a lot smaller. Its 10,000 times smaller than the wavelength at least. Would that also be a problem? I mean it does work in real life, I put exactly those signals in to a geometry like that and measured current waveforms. (I hope, this might be point where I find out I have been measuring something completely different).

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 14 sept. 2012, 10:57 UTC−4
Alex,

you can certainly use AC/DC for your problem. You will of course see capacitive coupling between the electrodes and respective current flow. No problem that the wavelength is so large. The lower the frequency the better the quasistatic approximation.

Cheers
Edgar
Alex, you can certainly use AC/DC for your problem. You will of course see capacitive coupling between the electrodes and respective current flow. No problem that the wavelength is so large. The lower the frequency the better the quasistatic approximation. Cheers Edgar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 14 sept. 2012, 11:00 UTC−4

Alex,

you can certainly use AC/DC for your problem. You will of course see capacitive coupling between the electrodes and respective current flow. No problem that the wavelength is so large. The lower the frequency the better the quasistatic approximation.

Cheers
Edgar


Cool

I guess the next question is how do I get the signal in there? I just tried setting electric potential to 10V and then putting 100e3 in the frequency parameter under the frequency domain study tab. However, it seems to compute a grid full of nothing. Not sure I am doing this right!
[QUOTE] Alex, you can certainly use AC/DC for your problem. You will of course see capacitive coupling between the electrodes and respective current flow. No problem that the wavelength is so large. The lower the frequency the better the quasistatic approximation. Cheers Edgar [/QUOTE] Cool I guess the next question is how do I get the signal in there? I just tried setting electric potential to 10V and then putting 100e3 in the frequency parameter under the frequency domain study tab. However, it seems to compute a grid full of nothing. Not sure I am doing this right!

Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 14 sept. 2012, 16:07 UTC−4
Consider specifying your input signal using a "lumped port." There are many ways to do these things wrong, by the way. So if you keep having trouble, you might want to upload your model for others to look at.
Consider specifying your input signal using a "lumped port." There are many ways to do these things wrong, by the way. So if you keep having trouble, you might want to upload your model for others to look at.

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.