Antenna matching network efficiency

Hi everyone,

I am currently designing a matching network with DTCs (digitally tunable capacitors) and I am struggling to properly calculate the corresponding S21. Correct me if I am wrong, but it seems to me that it is not correct to simulate the matching network S21 in a 50 Ohm environment while the antenna input impedance is different from 50 Ohm, simply because the currents and voltages in the matching network are different when it is loaded with the actual antenna input impedance. I know some softwares (ADS or Microwave Office for example) have built-in features that help sort this out, however I am using CST Microwave Studio in which no such tools are available. So, my question is, is there a simple way to overcome this problem ?
Thanks

Yann

I think it is necessary to focus on impedance matching.
You have matching network (passive device) with s11, s21, s12, s22 (Zout) coefficients.
Antenna with some input impedance Za (reflection coefficient Ga), and some generator or receiver with impedance Zs. In your case i would only consider Zs, s22 (Zout) and Za (Ga).
Gout=S22
Zout=50*(1+Gout)/(1-Gout)
Then make Zout = Za* (complex conjugate of Za).
And obtain Gout=(Za*-50)/(Za*+50)

If Gout is matched to Ga, then there should be no reflections. Then you can directly use s21 graph. The only thing left is antenna radiation efficiency. Or you can nalyze imperfect match of Ga to Gout=S22.

Also:
https://www.antenna-theory.com/basics/impedance.php

I hope i did not made any errors in my conclusions.
Your question is mind-boggling, i am really not sure if it should be S22 to Ga matching. Hope someone with more knowledge in this area correct me if i am wrong.

To summarize:
You can directly calculate S21 with any CAD.
Then only place is left to analyze is S22 to antenna Ga mismatch.

Dirtyfighter said:

Hi everyone,

I am currently designing a matching network with DTCs (digitally tunable capacitors) and I am struggling to properly calculate the corresponding S21. Correct me if I am wrong, but it seems to me that it is not correct to simulate the matching network S21 in a 50 Ohm environment while the antenna input impedance is different from 50 Ohm, simply because the currents and voltages in the matching network are different when it is loaded with the actual antenna input impedance. I know some softwares (ADS or Microwave Office for example) have built-in features that help sort this out, however I am using CST Microwave Studio in which no such tools are available. So, my question is, is there a simple way to overcome this problem ?
Thanks

Yann

Click to expand...

When you synthesize your matching network with two 50 Ohm ports you always try to synthesize it to your Optimum impedance which gives you the best performance. So, when you replace one of your ports with the input impedance (which is obviously a conjugate of your source impedance) and looking at S11 should show a good return loss. The same applies for insertion loss or gain.

P.S: Terminator 3 is spot on.