I hope you will add more of these short tutorials on how to use the network analyzer as a design tool. I wouldn't mind watching short tutorials on the use of a network analyzer as a verification tool neither. Thanks for sharing this knowledge...
Thanks Bowfinger for your nice feedback! I must admit that I've never measured more than 2 ports at the same time, but I've merged 6 separate *.s2p files into a single *.s4p file (12, 13, 14, 23, 24, 34) in order to analyze mixed-mode S-parameters. The import issue cannot be solved by increasing the MATLAB parameter like that: 1-port = 'ExpectedNumVariables',3 2-port = 'ExpectedNumVariables',9 3-port = 'ExpectedNumVariables',19 (wrong) 4-port = 'ExpectedNumVariables',33 (wrong) Because the Touchstone syntax of a *.s4p file looks like this: Freq. | S11(dB) | S11(ang) | S12(dB) | S12(ang) | S13(dB) | S13(ang) | S14(dB) | S14(ang) empty | S21(dB) | S21(ang) | S22(dB) | S22(ang) | S23(dB) | S23(ang) | S24(dB) | S24(ang) empty | S31(dB) | S31(ang) | S32(dB) | S32(ang) | S33(dB) | S33(ang) | S34(dB) | S34(ang) empty | S41(dB) | S41(ang) | S42(dB) | S42(ang) | S43(dB) | S43(ang) | S44(dB) | S44(ang) This means that you must stick with 'ExpectedNumVariables',9. But in order to help MATLAB, you should definitely delete the comments in your file (the lines with the exclamation marks). The "empty" cells should than be read as "NaN" by MATLAB. I've tested it right now (R2021a), and it works :) The readmatrix command is like a blessing if you ask me. When I was dealing with this in 2017, I had to use the dlmread command - that was like hell. Every space was read and I've written a separate MATLAB script in order to get the data organized. By the way, every VNA saves the data somehow different. In addition, I haven't looked at Touchstone version 2 yet.
![HIWWHEE | OHM [Official MV]](http://i.ytimg.com/vi/OJJLtTLqLag/mqdefault.jpg)
I hope you will add more of these short tutorials on how to use the network analyzer as a design tool. I wouldn't mind watching short tutorials on the use of a network analyzer as a verification tool neither. Thanks for sharing this knowledge...
Hey Fitzgerald,
Thank you for your feedback! I will work on making more videos about using VNAs.
Nice video! The touchstone import using readmatrix will break down once you handle more than 2 ports, though.
Thanks Bowfinger for your nice feedback! I must admit that I've never measured more than 2 ports at the same time, but I've merged 6 separate *.s2p files into a single *.s4p file (12, 13, 14, 23, 24, 34) in order to analyze mixed-mode S-parameters.
The import issue cannot be solved by increasing the MATLAB parameter like that:
1-port = 'ExpectedNumVariables',3
2-port = 'ExpectedNumVariables',9
3-port = 'ExpectedNumVariables',19 (wrong)
4-port = 'ExpectedNumVariables',33 (wrong)
Because the Touchstone syntax of a *.s4p file looks like this:
Freq. | S11(dB) | S11(ang) | S12(dB) | S12(ang) | S13(dB) | S13(ang) | S14(dB) | S14(ang)
empty | S21(dB) | S21(ang) | S22(dB) | S22(ang) | S23(dB) | S23(ang) | S24(dB) | S24(ang)
empty | S31(dB) | S31(ang) | S32(dB) | S32(ang) | S33(dB) | S33(ang) | S34(dB) | S34(ang)
empty | S41(dB) | S41(ang) | S42(dB) | S42(ang) | S43(dB) | S43(ang) | S44(dB) | S44(ang)
This means that you must stick with 'ExpectedNumVariables',9. But in order to help MATLAB, you should definitely delete the comments in your file (the lines with the exclamation marks). The "empty" cells should than be read as "NaN" by MATLAB. I've tested it right now (R2021a), and it works :)
The readmatrix command is like a blessing if you ask me. When I was dealing with this in 2017, I had to use the dlmread command - that was like hell. Every space was read and I've written a separate MATLAB script in order to get the data organized.
By the way, every VNA saves the data somehow different. In addition, I haven't looked at Touchstone version 2 yet.
Great video,.thank you ☺️