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 The information provided may be out of date.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.
Unable to calculate complex field for transient electromagnetic waves module
Posted May 14, 2014, 10:57 a.m. EDT RF & Microwave Engineering, Modeling Tools & Definitions, Parameters, Variables, & Functions Version 4.3b 3 Replies
Please login with a confirmed email address before reporting spam
Hi all,
I'm trying to simulate a Gaussian beam propagation in time domain for 780 nm light.
I'm using the Electromagnetic waves transient RF module.
And I use scattering boundary condition for incident electric field (source), which is very similar to COMSOL's example on second harmonic generation:
www.comsol.com/model/second-harmonic-generation-of-a-gaussian-beam-956
My problem is, I cannot calculate complex field solutions (the COMSOL's example on second harmonic generation also only calculate real part of the field). There will be error saying incorrect initial values.
I would want to get phase information of the propagating field, as well as the later case involving materials interaction with complex refractive index (which will not be possible to calculate with only real part of electric field).
Does anybody encounter similar problem?
I managed to get a solution by calculating the real part of the field first and then the phase shifted imaginary part later, and combine them together in another program, but that only works for free propagating field without any materials interaction.
Thank you all
Rico
I'm trying to simulate a Gaussian beam propagation in time domain for 780 nm light.
I'm using the Electromagnetic waves transient RF module.
And I use scattering boundary condition for incident electric field (source), which is very similar to COMSOL's example on second harmonic generation:
www.comsol.com/model/second-harmonic-generation-of-a-gaussian-beam-956
My problem is, I cannot calculate complex field solutions (the COMSOL's example on second harmonic generation also only calculate real part of the field). There will be error saying incorrect initial values.
I would want to get phase information of the propagating field, as well as the later case involving materials interaction with complex refractive index (which will not be possible to calculate with only real part of electric field).
Does anybody encounter similar problem?
I managed to get a solution by calculating the real part of the field first and then the phase shifted imaginary part later, and combine them together in another program, but that only works for free propagating field without any materials interaction.
Thank you all
Rico
3 Replies Last Post Jan 15, 2015, 6:57 p.m. EST
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
I haven't looked at the example you cited, but since no one else has replied to you, here's my thoughts: First, remember that physical electric and magnetic fields, just like all other measurable quantities, are real numbers (or real vectors, etc,) with no imaginary parts. The use of complex representations of electric and magnetic fields is advantageous when computing in the frequency domain, because the cosine(omega*t) and sine(omega*t) factors can be replaced by exponentials, which make the derivatives very convenient and gives a well-defined meaning to "phase." In a time domain formulation with non-sinusoidal field variations, phase is not a simply-defined quantity anymore. A non-sinusoidal waveform can be decomposed into sinusoidal parts via Fourier analyses, but there is no longer any single/unique phase at any place on the overall time-domain waveform. So generally, in the time-domain, one uses real-valued fields. That said, for time-domain waveforms that are nearly-sinusoidal, it is obvious that phase could have a meaning. If you have computed field-strength vs time data, you should be able to fit sinusoidal expressions to it (over some finite segment of time) and observe the phase that way, or you could apply Fourier transforms to your real-valued data to study the frequency-domain characteristics there.
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Thank you Robert for the reply,
I think I get what your explaining.
Any ideas on how to include materials complex dielectric function in the time domain EM calculation?
The equation in the module should have worked in my opinion, but I'm not sure.
I think I get what your explaining.
Any ideas on how to include materials complex dielectric function in the time domain EM calculation?
The equation in the module should have worked in my opinion, but I'm not sure.
Please login with a confirmed email address before reporting spam
Posted:
10 years ago
Thank you Robert for the reply,
I think I get what your explaining.
Any ideas on how to include materials complex dielectric function in the time domain EM calculation?
The equation in the module should have worked in my opinion, but I'm not sure.
Typically complex permittivity follows a Debye model, which is parameterized by first order exponential relaxation time constants for the polarization field. Perhaps you could look at it that way. Any time constant that was very fast, or very slow, relative to you pulse time would represent either a static permittivity (fast) or nothing (slow). Only time constants on the same order as your pulse time would need time domain modelling. I don't know if COMSOL handles polarization field relaxation time constants, but it's worth checking for this. Otherwise you'll need to add this first order DiffEq yourself.