# Discussion Forum

## Define an arbitrary EM wave

 Topics: 4.3b, RF Module, Heat Transfer
RSS feed   |   Turn on email notifications   |   4 Replies   Last post: August 18, 2013 10:04am UTC

John Smith

August 7, 2013 2:19pm UTC

Define an arbitrary EM wave

Hello
I am new in Comsol and i have a problem :
i want to calculate SAR for SAM phantom for different types of EM waves. but the source of EM waves ( signal generator ) isn't important. is there any wave to just define wave equation ? for example just define an incoming Gaussian wave and see its effects ?
Can i use Comsol model (SAR in human head) and remove the antenna and define an incoming EM wave instead of it ?

Robert Koslover

August 12, 2013 1:56pm UTC in response to John Smith

Re: Define an arbitrary EM wave

I think you should be able to do this using the RF module in the transient (aka, time-domain) formulation. That way, you can specify a wave as a function of time. And you can even use the built-in gaussian function.

John Smith

August 13, 2013 2:21am UTC in response to Robert Koslover

Re: Define an arbitrary EM wave

Dear Robert
Thank you. i see but to specify a spacial direction , we should used Ports ?

Robert Koslover

August 16, 2013 1:37am UTC in response to John Smith

Re: Define an arbitrary EM wave

If you define a geometric face as a "scattering boundary condition," you can specify the individual vector components of the fields and also of the k vector on that boundary, thus giving you general control of the wave direction, field amplitude and phase distribution, etc. Just be careful to do all your trigonometry and complex-fields-defining correctly. The "port" boundary conditions offered in the RF module appear to be of more-limited utility in the time-domain than in the frequency-domain. In the frequency domain, I use ports for most of my RF sources.

Added later: I regret implying that you should or must use complex fields. Using real-valued fields in time-domain mode should work fine.

John Smith

August 18, 2013 10:04am UTC in response to Robert Koslover

Re: Define an arbitrary EM wave

Thank You
I got it

Rules and guidelines