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.
microwave propagation
Posted 18.12.2009, 19:22 GMT-5 2 Replies
Please login with a confirmed email address before reporting spam
hi, all
i am trying to simulate the distortion of microwave propagation by an object having different dielectric properties to its background. the microwave signal was sent by monopole or dipole antennas.
it seems a nonlinear problem to me. should i use RF time dependent analysis instead of time harmonic analysis?
if i want to simulate from 1GHz--2GHz with a step frequency of 0.1GHz, do i have to input each frequency?
the object and background have dielectric dispersions, how do i assign this dielectric dispersion to the object and background? (they have different dielectric properties at different frequencies.)
thank you for you reply. happy holiday
i am trying to simulate the distortion of microwave propagation by an object having different dielectric properties to its background. the microwave signal was sent by monopole or dipole antennas.
it seems a nonlinear problem to me. should i use RF time dependent analysis instead of time harmonic analysis?
if i want to simulate from 1GHz--2GHz with a step frequency of 0.1GHz, do i have to input each frequency?
the object and background have dielectric dispersions, how do i assign this dielectric dispersion to the object and background? (they have different dielectric properties at different frequencies.)
thank you for you reply. happy holiday
2 Replies Last Post 03.01.2010, 00:04 GMT-5
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Based on your description, my initial thought is that conventional time-harmonic analysis applies. After all, in both 2D and 3D models using the RF module, you can specify a material's conductivity (sigma), which is what determines the lossiness of the material. This is seldom a strong function of frequency, for most ordinary dielectrics. If you want to study dispersion explicitly, then you should be able to specify a frequency-dependent value (or define a frequency dependent expression) for sigma.
It gets more interesting in the time domain version. The user-interface will still let you specify sigma, but clearly you could not define it explicitly as a function of frequency, since frequency has no well-defined value in a general time domain model. In principle, you could employ a time-dependent complex conductivity function, obtained via taking the Fourier-transform of your (presumably known) frequency-dependent conductivity function. I don't know if this would actually work or not, or what kind of solver-settings would then be best but I suspect the experts at Comsol could answer this. In particular, I expect such questions tend to arise in modeling plasmas.
I hope others will comment on your question further, since I would also like to hear what they have to say.
It gets more interesting in the time domain version. The user-interface will still let you specify sigma, but clearly you could not define it explicitly as a function of frequency, since frequency has no well-defined value in a general time domain model. In principle, you could employ a time-dependent complex conductivity function, obtained via taking the Fourier-transform of your (presumably known) frequency-dependent conductivity function. I don't know if this would actually work or not, or what kind of solver-settings would then be best but I suspect the experts at Comsol could answer this. In particular, I expect such questions tend to arise in modeling plasmas.
I hope others will comment on your question further, since I would also like to hear what they have to say.
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Based on your description, my initial thought is that conventional time-harmonic analysis applies. After all, in both 2D and 3D models using the RF module, you can specify a material's conductivity (sigma), which is what determines the lossiness of the material. This is seldom a strong function of frequency, for most ordinary dielectrics. If you want to study dispersion explicitly, then you should be able to specify a frequency-dependent value (or define a frequency dependent expression) for sigma.
It gets more interesting in the time domain version. The user-interface will still let you specify sigma, but clearly you could not define it explicitly as a function of frequency, since frequency has no well-defined value in a general time domain model. In principle, you could employ a time-dependent complex conductivity function, obtained via taking the Fourier-transform of your (presumably known) frequency-dependent conductivity function. I don't know if this would actually work or not, or what kind of solver-settings would then be best but I suspect the experts at Comsol could answer this. In particular, I expect such questions tend to arise in modeling plasmas.
I hope others will comment on your question further, since I would also like to hear what they have to say.
thank you Robert. it helps a lot.
i'll also paste my question in the modeling penal.
happy new year