Assistance Required Modeling HV to Signal Wire Coupling

James Taton Applied Physics

Please login with a confirmed email address before reporting spam

Hello All,

I am attempting to recreate an experiment in COMSOL in which a high voltage wire has an applied switching voltage (300V peak to peak) and is placed within proximity of a test signal wire to observe the coupling between them for EMC/EMI applications. I am however having an issue figuring out which physics interface(s) and domain/boundary conditions I should be incorporating into the model to achieve the expected outcome. The HV wire potential is set to a square waveform and I should be observing spikes in the signal wire potential at the rise and falls. I have played around with ES and EC to no avail and before I start poking around further on my own, I figured some guidance would be a better option to get started

All of my previous COMSOL experience has been with microfluidics reasearch so I'm a little unsure changing things up but any help would be much appreciated!


3 Replies Last Post 26.06.2025, 21:18 MESZ
Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 55 minutes ago 26.06.2025, 20:39 MESZ
Updated: 53 minutes ago 26.06.2025, 20:41 MESZ

There are no true square waves. A true square wave has infinite frequency content. What is the real-world rise-time of the signal that you wish to model? Assuming that time is and with c= speed of light, how do the dimensions of the important parts of your model compare to ? For a slow enough rise time (if your dimensions are very small compared to ), then perhaps you can use a quasi-static EM fields representation, and the AC/DC module has a number of variants of those available. Otherwise, you need to employ a time-domain RF model.

-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
There are no true square waves. A true square wave has infinite frequency content. What is the real-world rise-time of the signal that you wish to model? Assuming that time is t_{rise} and with c= speed of light, how do the dimensions of the important parts of your model compare to c*t_{rise} ? For a *slow enough* rise time (if your dimensions are very small compared to c*t_{rise} ), then perhaps you can use a quasi-static EM fields representation, and the AC/DC module has a number of variants of those available. Otherwise, you need to employ a time-domain RF model.

James Taton Applied Physics

Please login with a confirmed email address before reporting spam

Posted: 36 minutes ago 26.06.2025, 20:58 MESZ

There are no true square waves. A true square wave has infinite frequency content. What is the real-world rise-time of the signal that you wish to model? Assuming that time is t_{rise} and with c= speed of light, how do the dimensions of the important parts of your model compare to ct_{rise} ? For a *slow enough rise time (if your dimensions are very small compared to c*t_{rise} ), then perhaps you can use a quasi-static EM fields representation, and the AC/DC module has a number of variants of those available. Otherwise, you need to employ a time-domain RF model.

Rise time of the wave is around 200ns and a frequency of 10Hz so the product comes out to about 60m and the wires I'm modeling are only a meter long and have mm scale radii. Would that be within reason to use quasi-static? I believe I have essentially been attempting that with a Time-Dependent solver in AC/DC but either detecting no induced voltage on the signal wire or a non-converging solution.

>There are no true square waves. A true square wave has infinite frequency content. What is the real-world rise-time of the signal that you wish to model? Assuming that time is t_{rise} and with c= speed of light, how do the dimensions of the important parts of your model compare to c*t_{rise} ? For a *slow enough* rise time (if your dimensions are very small compared to c*t_{rise} ), then perhaps you can use a quasi-static EM fields representation, and the AC/DC module has a number of variants of those available. Otherwise, you need to employ a time-domain RF model. Rise time of the wave is around 200ns and a frequency of 10Hz so the product comes out to about 60m and the wires I'm modeling are only a meter long and have mm scale radii. Would that be within reason to use quasi-static? I believe I have essentially been attempting that with a Time-Dependent solver in AC/DC but either detecting no induced voltage on the signal wire or a non-converging solution.

Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 16 minutes ago 26.06.2025, 21:18 MESZ
Updated: 12 minutes ago 26.06.2025, 21:22 MESZ

Yes, the AC/DC module would seem to be applicable in your case. The ES formulation is for electrostatics. So I encourage you to use one of the many other formulations available in the AC/DC module. There are many examples, tutorials, videos, etc. available at various locations on the Comsol Website. Choose one of the formulations that includes the important physics that you need to capture, and choose the correct boundary conditions, etc. If you haven't set up many AC/DC models before, then you have some homework to do to come up to speed.

-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
Yes, the AC/DC module would seem to be applicable in your case. The ES formulation is for electrostatics. So I encourage you to use one of the many other formulations available in the AC/DC module. There are many examples, tutorials, videos, etc. available at various locations on the Comsol Website. Choose one of the formulations that includes the important physics that you need to capture, and choose the correct boundary conditions, etc. If you haven't set up many AC/DC models before, then you have some homework to do to come up to speed.

Reply

Please read the discussion forum rules before posting.

Please log in to post a reply.

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.