AnsweredAssumed Answered

S21 time domain response

Question asked by antennas4space on Mar 1, 2007
Latest reply on Mar 2, 2007 by Dr_joel
I have some questions about the time domain response of an S21 measurement.

I performed a transmission response calibration from 22 to 28 GHz over 1601 points with a stepped sweep, 0 us additional dwell time, 1 kHz IFBW, default sweep time, from port 1 to port 2 using the standard PNA cables.

When I measure a 10 foot cable, the S21 time domain correctly looks like an impulse at the delay length, with obligatory sidelobes.

With a much longer cable, the S21 time domain response becomes "smeared" in time. See attached. Actually for this case, I was going through our anechoic chamber, but the response effect is the same.

I am also pretty sure that I have enough points such that the response is not aliased. The calculated time seems correct, and there are no other repeated responses.

I assumed this was a "measuring electrically long device" issue. I have tried slowing the sweep time with no effect. Tried using a reference cable into R1 from a directional coupler down-stream but got data that didn't make sense at all.

Does this look like an long device problem, or is this indicative of something else? And, if so, are there any other recommendations to yield the impulse response that I am expecting?

The only other unusual item I see is that sometime when I change a parameter, such as the tweaking the stop time of the transform window, it does show an impulse for one sweep, but then returns to the attached configuration.