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 Gaussian pulses using RF Module

Please login with a confirmed email address before reporting spam

Hi everyone,

I am using the RF module- Electromagnetic Waves, Frequency Domain, as shown in the attached COMSOL file.

I want to create Gaussian pulses within a period of time (for example 1 ms). These pulses should be applied from an input Port, (Port 1 is a port input power). I tried to write the below equation in the Port Settings: Pin = P*exp(-(t-t0)^2), But it does not work

I have searched on the FORUM, and seen there are several examples and tutorials around, but I can't figure out a way to apply pulses by Port1.

Would you please show me how to create a number of different Gaussian pulses?

Thanks in advance, Manal



6 Replies Last Post 16 set 2022, 11:09 GMT-4
Dave Greve Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 2 years ago 12 set 2022, 11:24 GMT-4

If you want non-sinusoidal input signals you need to work in the time domain (temw).

If you want non-sinusoidal input signals you need to work in the time domain (temw).

Please login with a confirmed email address before reporting spam

Posted: 2 years ago 12 set 2022, 12:47 GMT-4
Updated: 2 years ago 12 set 2022, 13:15 GMT-4

Hi Dave

Thank you for your response,

my work requires me to apply several pulses by Port 1 in the frequency domain.

Would you please give me an equation I can put it into Port Settings to apply several laser pulses (non-sinusoidal input signals)?

Best regards, Manal

Hi Dave Thank you for your response, my work requires me to apply several pulses by Port 1 in the frequency domain. Would you please give me an equation I can put it into Port Settings to apply several laser pulses (non-sinusoidal input signals)? Best regards, Manal

Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 2 years ago 12 set 2022, 13:07 GMT-4
Updated: 2 years ago 12 set 2022, 13:07 GMT-4

Dave Greve is right (as he usually is). I encourage you to learn about and use the time-domain RF modeling capabilities, which are included in the same RF module that you are using for frequency-domain modeling. You might want to consider searching for the word "transient" in the Application Library, and then look at the examples for that, listed under the RF module.

-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
Dave Greve is right (as he usually is). I encourage you to learn about and use the time-domain RF modeling capabilities, which are included in the same RF module that you are using for frequency-domain modeling. You might want to consider searching for the word "transient" in the *Application Library,* and then look at the examples for that, listed under the RF module.

Please login with a confirmed email address before reporting spam

Posted: 2 years ago 12 set 2022, 15:21 GMT-4
Updated: 2 years ago 12 set 2022, 15:23 GMT-4

Dears Robert and Dave

Many thanks for your response.

I followed your advice and have changed my simulation to be worked in the time domain (temw). I followed the steps in the Example "Transient Modeling of a Coaxial Cable" in the Application Library. However, I have got the attached error message.

Would you please tell me how I can solve this problem?

I think the problem is that in the Zref text field, I typed the below equation: (Z0_const/2/pi)*log(width of the core).

Best regards, Manal

Dears Robert and Dave Many thanks for your response. I followed your advice and have changed my simulation to be worked in the time domain (temw). I followed the steps in the Example "Transient Modeling of a Coaxial Cable" in the Application Library. However, I have got the attached error message. Would you please tell me how I can solve this problem? I think the problem is that in the Zref text field, I typed the below equation: (Z0_const/2/pi)*log(width of the core). Best regards, Manal


Dave Greve Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 2 years ago 13 set 2022, 12:42 GMT-4

Based on your previous posts, my guess is that you are trying to simulate heating of the absorber in response to a train of Gaussian pulses.

If so, you want a transient simulation but of a particular type. If you simulate the Gaussian pulse itself, you need several time steps per (optical) cycle, requiring an absurd computation time.

I think you will get the results you want if you use the preset study Frequency-Transient. That study becomes available if you select both emw and ht physics and then select Electromagnetic Heating physics. This physics solves the EM equations at the frequency specified and applies the time-averaged EM heating for the heat transfer problem.

In emw you will be able to write for the port power, say, 1[W]*f(t) where f(t) is the envelope for your Gaussian pulses.

Be aware that these multiphysics problems may not converge well and may require non-default settings for the solvers for reasonable computation times.

You may want to set up a Frequency-Stationary study first as this calculation is faster and easier to troubleshoot.

Based on your previous posts, my guess is that you are trying to simulate heating of the absorber in response to a train of Gaussian pulses. If so, you want a transient simulation but of a particular type. If you simulate the Gaussian pulse itself, you need several time steps per (optical) cycle, requiring an absurd computation time. I think you will get the results you want if you use the preset study Frequency-Transient. That study becomes available if you select both emw and ht physics and then select Electromagnetic Heating physics. This physics solves the EM equations at the frequency specified and applies the time-averaged EM heating for the heat transfer problem. In emw you will be able to write for the port power, say, 1[W]*f(t) where f(t) is the envelope for your Gaussian pulses. Be aware that these multiphysics problems may not converge well and may require non-default settings for the solvers for reasonable computation times. You may want to set up a Frequency-Stationary study first as this calculation is faster and easier to troubleshoot.

Please login with a confirmed email address before reporting spam

Posted: 2 years ago 16 set 2022, 11:09 GMT-4

Hi Dave

Thank you for your reply,

I have tried your solution and currently I am working on that, I hope I can acheive my targets.

best regards, Manal

Hi Dave Thank you for your reply, I have tried your solution and currently I am working on that, I hope I can acheive my targets. best regards, Manal

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.