0
\$\begingroup\$

I measured 100 S2P parameters using a Vector Network Analyzer (VNA), and now I want to have an average of these 100 parameters in ADS.

What should I do? Should I define the SnP parameter like below for each of them? Or is there an easier way? Any suggestion is appreciated.

enter image description here

Here is an example of three of them.

\$\endgroup\$
4
  • \$\begingroup\$ If you do average them, make sure the representation is linear complex numbers rather than dB and degrees. \$\endgroup\$
    – Neil_UK
    Commented Feb 13, 2023 at 21:16
  • \$\begingroup\$ @Neil_UK how do you recommend me to do the average? \$\endgroup\$
    – Ferrofluid
    Commented Feb 13, 2023 at 22:32
  • \$\begingroup\$ Perform sum of N readings, then divide by N. What format are the readings currently in? Presumably a file? Paste a section of the file into your question so we can see what your difficulty is. \$\endgroup\$
    – Neil_UK
    Commented Feb 14, 2023 at 5:48
  • \$\begingroup\$ Hi @Neil_UK thank you very much for the response. I added a drop box link and added three of my results there. The files are in S2P format. How do you recommend me to take an average? Is it possible in ADS software? Any suggestions appreciated. \$\endgroup\$
    – Ferrofluid
    Commented Feb 14, 2023 at 16:18

1 Answer 1

Reset to default
0
\$\begingroup\$

This is an example of part of the start of one of your files

!Date: Sunday, January 29, 2023 12:50:38
!S2P File: Measurements: S11, S21, S12, S22:
# Hz S  dB   R 50
300000 -18.862196 4.9757056 -1.0350283 -0.70020109 -1.0443226 -0.70608562 -18.841742 5.5203471
11549625 -18.507008 1.060286 -1.1138686 -1.1768543 -1.1045679 -1.0994619 -18.402775 0.8063547
22799250 -18.426577 2.0101566 -1.1181669 -2.1032612 -1.1124948 -2.1241975 -18.416607 1.4881673
34048875 -18.370947 2.3688161 -1.1192718 -3.210099 -1.1200194 -3.2442918 -18.300875 2.537688
...

You have a row per frequency (first row 300 kHz, next row 11.5 MHz etc) of 4 columns of S parameters, in the order S11, S21, S12, S22, in dB/angle format.

This representation is not suitable for averaging as it's non-linear, the scale varies with radius. They must be converted to complex numbers, using the standard trig conversion \$Z = a.cos(theta) + j.a.sin(theta)\$, where \$a = 10^{\frac{dB}{20}}\$

Taking a worked example of S11 at 300 kHz, this means a = 0.11399615, so Z = 0.113566 + j0.009887.

I'm not sure what you want to average with what (between files, or over a small range of frequencies (but a curve fit might be a better operation)), but, do this to all the measurements, add them together as complex numbers, divide by the number of measurements you have added together, and the result is the average. You will probably want to transform the average from complex to dB/angle at the end for more familiar units.

Use your favourite computing tool. Excel will manage it happily, I'd use python, Matlab would be fine, almost any other language that can read text files would do. Beware that the angles in the s2p file are in degrees, and most implementations of trig functions want them in radians.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.