[TowerTalk] Test Fixture for Common Mode Chokes

[Brian Beezley]

I went through the dozen or so .s2p files I have for common-mode chokes. 
The error due to shunt port capacitance was most often restricted to the 
resistive and reactive parts of the measured impedance, not the 
impedance magnitude. This means that the pi network formed by unwanted 
shunt port capacitance in conjunction with the choke is acting like an 
all-pass network. I have no idea why.

Due to shunt port capacitance, the conventional S21 series measurement 
method overestimates effective choke resistance. After using the the Y21 
method to cancel the capacitance, choke resistance at HF typically drops 
less than 10%. I think this puts the error in the negligible category 
for most hams. I have a few measurements that go to 50 or 60 MHz. Error 
at VHF is quite a bit higher. It may not be negligible for a 6m choke.

If anyone with a VNWA wants a script for the Y21 method, I can provide 
one developed by DH0KAI. It lets you plot Y21 results right in the VNWA 
software, including shunt capacitance at each port. This lets you rework 
and improve your choke fixture without the need for a capacitance meter.

Using all four S-parameters provides best Y21 accuracy. You also need 
them to obtain separate capacitance values for both ports. If your VNA 
only provides S11 and S21, you can interchange the port cables and make 
a second measurement. Combine the two [S11 S21 0 0] files to [S11 S21 
S12 S22] with https://tinyurl.com/45xk3zum.

I'm always looking for .s2p common-mode choke files. I seem to learn 
something from each one I examine. Please forward any you may have. 
Thanks to NA6O who sent me several.

I've added Y21 shunt capacitance plots to my S-parameter plotter. You 
can see some curves here:

https://k6sti.neocities.org/sp

Brian