AnsweredAssumed Answered

HP8510C/8515a has over 20 dB power slope from 1GHZ to 22 GHZ

Question asked by etroy on Oct 10, 2011
Latest reply on Oct 11, 2011 by RayH
I have an HP8510C system with an 8515A S parameter box. I started noticing some strange behavior, where I was getting NO IF error at frequencies above 20 GHz. I also noticed very bad performance at the higher frequencies. At  first I suspected my source was bad, but I confirmed that it works ok, at least when commanded by the network analyzer. (It does seem to stop sweeping at about 20 GHz when commanded to sweep from 8 GHz to 22 GHz, when using it by itself (HP8350B/83595A), but I am not sure about that issue. But, in an effort to debug the problem, I hooked the S11 and S22 ports to my spectrum analyzer and swept from 1 GHz to 24 GHz at about +7 dBm. I noticed the power level on the spectrum analyzer went from about -3 dBm at 1 GHz to about -26 dBm at 24 GHz. I thought it was supposed to be pretty flat, and that is not. So, I suspected the cable going from the signal generator to the 8515A (I did have an issue with it a few months ago.) But, when I connected the output of the signal generator directly to the spectrum analyzer and did the same sweep, the power was quite flat, and well above 0 dBm. There is obviously some frequency slope in the RG141 coax, but nothing like the nearly 25 dB of change that I am seeing at the test port connectors. (S11 and S22 are the same.)

I thought maybe it was the input RF connectors in the 8515A. They were a little loose, but tightening them made no difference. Then, I opened the unit and bypassed the external connector and the short coax that goes to the first switch, and inserted the RF directly into the switch. No difference. I also tried another cable. No difference. I went into the service manual and did the steps where you re-assign the input paths, called "Test set unratioed power troubleshooting", and those tests look normal.

I suspect the occasional "NO IF" error that I get at higher frequencies is related to this situation. I get that error even more if I try to set the signal generator level to -5 dBm.

Does anyone have an idea what the problem could be? I can't believe that the unit has a 25 or more dB slope in the test port power from 1 to 24 GHZ.