The mux can tolerate small negative excursions without creating significant distortion. The VNA signal may be small enough to get by. In any case? maintaining the bias is ' good housekeeping'.
The Tayloe impedance is a trickier issue. Since, at 12 KHz separation to the clock frequency we are operating on the steep side of a slope, it's difficult to know what that impedance is in reality. One might think "Well, I can just measure that with my VNA!".
?Not so simple! The VNA port 2 is essentially a receiver listening for amplitude and phase information in relation to port 1. Clock feedthrough at the mux introduces yet another signal to that receiver and can bollix your measurement!
Thank you for coming back! I should indeed consider the input impedance of the Tayloe detector when doing this.
As for the follow-up questions: Yes the antenna port was terminated as per the instructions of the manual when the "internal" RF sweep was done, but the DC bias on the switch output was not there(!). (I somehow assumed the T2 primary to be grounded but quite obviously it is not). I guess the bias ensures that the signal voltage remains within the rail voltage, and thus I am quite surprised that I did not see gross distortion of the signal at the switch output. In any case, I should repeat this after re-establishing the bias and hooking the VNA through a DC block.
I will do the above out of curiosity, but that will not remove the impedance issue you mentioned. I went down the rabbit hole because the figure reported by the FT8 software was comparable between the 20m and the 17m setting during normal operation -- that did not seem to be in line with what the RF sweep showed, raising doubts on said sweep. After all this, I should probably re-do that comparison more carefully first.
I will let you know how that goes Thanks again and regards! Matt
