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.

Creating a electromagnetic pulse

Please login with a confirmed email address before reporting spam

Hey all,

I am trying to create a Gaussian pulse that simply travels across the simulation area (this is part 1). I am wondering if I anyone has experience or knows if this is correct.

I am using the RF module, using Transient Electromagnetic Wave physics. I am using a scattering boundary condition on the left side of my simulation area. Here, I set the wave to

Ez = V_max*exp(-(t-t0)^2)

However, the pulse that is generated doesn't really look like a Gaussian.

Any help is appreciated.

Thanks,

John

3 Replies Last Post 2011年6月7日 GMT-4 06:36
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 2011年6月6日 GMT-4 11:03
Hi

first of al try a search on the FORUM, there are several examples around. If you are in v4, you have the gaussian "function" predefined

--
Good luck
Ivar
Hi first of al try a search on the FORUM, there are several examples around. If you are in v4, you have the gaussian "function" predefined -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2011年6月6日 GMT-4 23:31
Thanks.

Here is a problem I am having:

Whenever I run the solver using 'initial values of variables solved for' = initial values, the pulse looks messed up. It doesn't look like a single gaussian pulse but rather has a few peaks/troughs.

However if I run the solver using the initial values = Solver 1, after running the simulation once, the pulse looks correct! The problem is that it still has the last values solved for the previous attempt, so it doesn't start out with a clean run.

I am wondering if this is because a Gaussian pulse does not actually have a non-zero value anytime (just infinitesimally small), so my initial values of 0 are causing problems.

Does anyone have experience with this problem?

Can I define a piecewise function to change the gaussian pulse to only be non-zero after time t=t_0? Or is there some other work-around?

If anyone knows if I am on the right track I would really appreciate some feedback.

Thanks,

John
Thanks. Here is a problem I am having: Whenever I run the solver using 'initial values of variables solved for' = initial values, the pulse looks messed up. It doesn't look like a single gaussian pulse but rather has a few peaks/troughs. However if I run the solver using the initial values = Solver 1, after running the simulation once, the pulse looks correct! The problem is that it still has the last values solved for the previous attempt, so it doesn't start out with a clean run. I am wondering if this is because a Gaussian pulse does not actually have a non-zero value anytime (just infinitesimally small), so my initial values of 0 are causing problems. Does anyone have experience with this problem? Can I define a piecewise function to change the gaussian pulse to only be non-zero after time t=t_0? Or is there some other work-around? If anyone knows if I am on the right track I would really appreciate some feedback. Thanks, John

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 2011年6月7日 GMT-4 06:36
Hi

if you have a time gaussian pulse, you should probably also tailor the stepping for the time, not only via the "range" command, but also the stepping method, I would propose to give many fine steps around the gaissian maximum, and less outside, then select "intermediate" and not "automatic" in the solver sub-node for the time stepping settings. This will ensure that your gaussian pulse is sampled fine enough.

Furthermore the default v4 gaissian has a mean of "0" so you might want to delay it to get a better curve starting closer to "0"

--
Good luck
Ivar
Hi if you have a time gaussian pulse, you should probably also tailor the stepping for the time, not only via the "range" command, but also the stepping method, I would propose to give many fine steps around the gaissian maximum, and less outside, then select "intermediate" and not "automatic" in the solver sub-node for the time stepping settings. This will ensure that your gaussian pulse is sampled fine enough. Furthermore the default v4 gaissian has a mean of "0" so you might want to delay it to get a better curve starting closer to "0" -- Good luck Ivar

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.