Menu

Discussion Forum

New Discussion
Filter

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 The information provided may be out of date.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.

Problems with S-parameter calculation

Posted May 22, 2012, 7:01 p.m. EDT RF & Microwave Engineering Version 4.2a 10 Replies

Esteban Zamora
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

I do not understand the behavior of the transmission parameter S21 in a rectangular wave guide excited with two Port, there are differences in behavior between S21 and dBS21, these are the following differences:

-S21 has peaks. Evident when exporting data and analyze it presents oscillation between 1 and -1

-dBS21 plot in COMSOL, the oscillatory beahavoir of S21 is not presente and what is more strange is when calculated
with the formula (dBS21 = 20 * log (abs (S21)), I get different values ​​from those used in the graphic COMSOL dBS21 and the oscillatory beahavoir is present.

If Comsol used otherwise for caculation of dBS21 parameter please enlighten me, that the results did not close.

Please help me!
10 Replies Last Post Feb 28, 2013, 2:54 a.m. EST
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 23, 2012, 3:18 a.m. EDT
Hi

have you run and studied the Model library "Branch Line Coupler" ? you might find some replies thereby. It is indeed strange that the dB and linear representastion are different, I suspect you have something else you are missing

--
Good luck
Ivar
Hi have you run and studied the Model library "Branch Line Coupler" ? you might find some replies thereby. It is indeed strange that the dB and linear representastion are different, I suspect you have something else you are missing -- Good luck Ivar
Esteban Zamora
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 23, 2012, 5:28 a.m. EDT
Thanks for your answer

I have been analyzing the data, exporting data comsol and plotting in other software. When applied to the above equation at S21 data get the same results, but after some resuts the Comsol data tend to zero while the result obtained with the equation oscillated respect to zero but the magnitude is very diferent.
I hope that there are a explanation for this dissimilitude. For the attached model the cut frecuency at TE10 mode in the theory is 6.562GHz.

Thanks again for your help.

Thanks for your answer I have been analyzing the data, exporting data comsol and plotting in other software. When applied to the above equation at S21 data get the same results, but after some resuts the Comsol data tend to zero while the result obtained with the equation oscillated respect to zero but the magnitude is very diferent. I hope that there are a explanation for this dissimilitude. For the attached model the cut frecuency at TE10 mode in the theory is 6.562GHz. Thanks again for your help.
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 23, 2012, 7:39 a.m. EDT
Hi

how do you normalise the S21 curve to get the dBS21 curve?
As you have S21 positive and negative, you canot take the log of that without some renormalisation.
Your dB curve sem to pass by "0" at each extrema of S21

Have you checked the formulas COMSOL uses (turn on the Equation view, they are perhaps listed therein)

--
Good luck
Ivar
Hi how do you normalise the S21 curve to get the dBS21 curve? As you have S21 positive and negative, you canot take the log of that without some renormalisation. Your dB curve sem to pass by "0" at each extrema of S21 Have you checked the formulas COMSOL uses (turn on the Equation view, they are perhaps listed therein) -- Good luck Ivar
Esteban Zamora
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 23, 2012, 12:49 p.m. EDT
Thanks for your answer.

You can explain how I can view the equation, because I have no idea. I used absolute value of S21 for applied the logaritmic funtion as mentioned in the RFModuleUsersGuide for S11.

Thanks for your help.
Thanks for your answer. You can explain how I can view the equation, because I have no idea. I used absolute value of S21 for applied the logaritmic funtion as mentioned in the RFModuleUsersGuide for S11. Thanks for your help.
Alexander Kuznetsov
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 23, 2012, 1:54 p.m. EDT
you should keep in mind that S21 is generally complex number. So what you S21 plot shows is only the real part of it. Make sure to take into account imaginary part as well...
you should keep in mind that S21 is generally complex number. So what you S21 plot shows is only the real part of it. Make sure to take into account imaginary part as well...
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 23, 2012, 2:13 p.m. EDT
Hi

OK if you take the absolute value of an oscillating (alternating sign) function you double the peaks, and that explains the alternances of zeros and extremas

--
Good luck
Ivar
Hi OK if you take the absolute value of an oscillating (alternating sign) function you double the peaks, and that explains the alternances of zeros and extremas -- Good luck Ivar
Esteban Zamora
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 28, 2012, 11:44 p.m. EDT
Thanks for your answer

you are right, and I would like to ask you about Port uses.
it is excite a rectangular wave guide with a port of the same type (rectangular), but at the outlet of the guide is circular. The question is: What kind of port should be use in the outlet? if mode and frequency are defined for rectangular inlet section (this problem is similar to Model Wave guide adapter but with definid mode and frecuency in the inlet section )


Thanks for your help.
Thanks for your answer you are right, and I would like to ask you about Port uses. it is excite a rectangular wave guide with a port of the same type (rectangular), but at the outlet of the guide is circular. The question is: What kind of port should be use in the outlet? if mode and frequency are defined for rectangular inlet section (this problem is similar to Model Wave guide adapter but with definid mode and frecuency in the inlet section ) Thanks for your help.
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 29, 2012, 2:13 a.m. EDT
Hi

Sorry I havent been active in that part of the physics for a few years. Back in 3.5 you needed to integrate a few equations to get the appropriate form factor to use on the Port adaptation vualues, but I cannot remmeber precisely, and They must have improved that since.

There are several good RF and optics specialist out here, hopefully one of tehm is willing to give us some explanations ;)

--
Good luck
Ivar
Hi Sorry I havent been active in that part of the physics for a few years. Back in 3.5 you needed to integrate a few equations to get the appropriate form factor to use on the Port adaptation vualues, but I cannot remmeber precisely, and They must have improved that since. There are several good RF and optics specialist out here, hopefully one of tehm is willing to give us some explanations ;) -- Good luck Ivar
AH
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 28, 2013, 2:36 a.m. EST
can you plrease tell me how to export the s parameters of the modell in touchstone format ? I did exactly the same like the hand book... choose reference impedance e.g 50 ohm and gave a name to my touchstone file and a path and I calculated after it. But it dosent work no error message or so but the software dosent give me any toucstone files !

I work with COMSOL 4.3
Thanks
can you plrease tell me how to export the s parameters of the modell in touchstone format ? I did exactly the same like the hand book... choose reference impedance e.g 50 ohm and gave a name to my touchstone file and a path and I calculated after it. But it dosent work no error message or so but the software dosent give me any toucstone files ! I work with COMSOL 4.3 Thanks
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 28, 2013, 2:54 a.m. EST
Hi

I'm not aware of any automatic Touchetone file generation (there might be though ;) but if its a standard

frequency, S11 dB + phase, S12 ...

type file, then you can generate the table with the Derived Values nodes and export the resulting table easily

--
Good luck
Ivar
Hi I'm not aware of any automatic Touchetone file generation (there might be though ;) but if its a standard frequency, S11 dB + phase, S12 ... type file, then you can generate the table with the Derived Values nodes and export the resulting table easily -- Good luck Ivar

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.

Suggested Content