during my approach of designing a mixer I encountered the problem that I get positive values for the return loss (in dB) at the small signal (RF) port. It appears that this behavior depends on the small signal power level which I find unreasonable. I tried to reconstruct the problem using the "Small_Signal_Sparameters_Under_Large_Signal" example (see attached). There I put port 2 in power sweep mode (30 to 70 dB below port 1 power). I would expect all lines in the S12 and S22 diagram to lie on top of each other, but this is not the case.
I thought it might be a problem with the error settings of the harmonic balance, but decreasing the relative and absolute error for APLAC does not change anything.
Reactivating the LSSS control block produces reasonable but different results.
Do I really need the LSSS block? Why does the Tone=2 port with small enough power level not have the same effect?
Why do I even need two schematics for all S-parameters? In the provided example the "S12_and_S22_Test_Bench" gives the same results for S11 and S21 as the "S11_and_S21_Test_Bench".
This is going to be much easier to explain with some graphics embedded in my reply, but this forum does not permit that, so I will reply to you in a direct email rather than on the forum. You should receive this email shortly...