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 The information provided may be out of date.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.

How to set up the port?

Please login with a confirmed email address before reporting spam

Dear all,

I use RF module constructing 2 models, which have same structures except the length. The input power and other parameters are also the same. But after simulation, the power at the incident port of both models are not equal. I want them to have a same value in order to compare the wave and study the dissipate power along distance. I already use a scattering BC. The characteristic are similar but still not equal. So, I wonder that is there any way to set up the structure to have a same input power after the simulation?

Thank you in advance

2 Replies Last Post 02.11.2015, 00:55 GMT-5
Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 9 years ago 04.08.2015, 14:28 GMT-4
Select your wave-launching boundary as a "Port," set "Wave excitation at this point" to On, and then click (to check) the checkbox that says "Specify deposited power." This is supposed to force the input power to the value you set.

Alternatively, If your model is linear, then you can just leave that box unchecked, and scale your results based on the post-processed computed input power, which you can find by integrating the normal component of the average Poynting vector on the port. Also, if you choose this latter method, you don't have to choose your launching boundary as a "Port" at all, but could use a scattering boundary, if you prefer. But I use Ports whenever they fit the physics, since they come with convenient computational features.
Select your wave-launching boundary as a "Port," set "Wave excitation at this point" to On, and then click (to check) the checkbox that says "Specify deposited power." This is supposed to force the input power to the value you set. Alternatively, If your model is linear, then you can just leave that box unchecked, and scale your results based on the post-processed computed input power, which you can find by integrating the normal component of the average Poynting vector on the port. Also, if you choose this latter method, you don't have to choose your launching boundary as a "Port" at all, but could use a scattering boundary, if you prefer. But I use Ports whenever they fit the physics, since they come with convenient computational features.

Please login with a confirmed email address before reporting spam

Posted: 9 years ago 02.11.2015, 00:55 GMT-5
Dear Robert,
I am simulating a metal-dielectric tri-layer structure with chi3 contribution to the electric permittivity at higher power levels. I have defined the electric permittivity from Lorentz-Drude Model (L-D) and defied the epsilon (Total) = epsilon (L-D) +3*chi3*mod(E^2). Here mod(E^2) is from the selected domain probes. So for the wave excitation, i have opted the "port" and defined "specified deposited power" as 1W, and activated the slit condition downward to the simulating unit cell with domain backed conditions. In the computation I have used "Stationary solver" with "MUMPS" solver. The simulation starting to "Nonlinear solver" and iterative mode until finding a convergent result. The electric field coming out to some value with 10^(-20).

But the same simulation with unchecked box for "specified deposited power" running the simulation in "linear solver" and giving the electric field to some value of 10^(7).

So why does the electric field differs a lot. what method should i follow for the defined specified deposited power contribution to the electric permittivity.

Thank you
sriram
Dear Robert, I am simulating a metal-dielectric tri-layer structure with chi3 contribution to the electric permittivity at higher power levels. I have defined the electric permittivity from Lorentz-Drude Model (L-D) and defied the epsilon (Total) = epsilon (L-D) +3*chi3*mod(E^2). Here mod(E^2) is from the selected domain probes. So for the wave excitation, i have opted the "port" and defined "specified deposited power" as 1W, and activated the slit condition downward to the simulating unit cell with domain backed conditions. In the computation I have used "Stationary solver" with "MUMPS" solver. The simulation starting to "Nonlinear solver" and iterative mode until finding a convergent result. The electric field coming out to some value with 10^(-20). But the same simulation with unchecked box for "specified deposited power" running the simulation in "linear solver" and giving the electric field to some value of 10^(7). So why does the electric field differs a lot. what method should i follow for the defined specified deposited power contribution to the electric permittivity. Thank you sriram

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.