Is it feasable to try and build a detector for 10MHz to 40GHz ?

[Kilo Tango]

I recently bought an R & S vector network analyser but someone has ripped out one of the detectors ( and damaged a lot of plumbing ). Is it feasable to try and build another one ?. The service manual gives a test for it as with 0dBm input, between 2-10GHz the output should be -210mV.

 Here’s a picture of the remaining one.
 
Any ideas ?

Ken

[Kire Pûdsje]

Where is it used?

It just looks like a resistive power splitter with a diode detector.
If you have the space (and do not need the bandwidth), you could try with just a splitter and a crystal detector. Crystal detectors up to 18 GHz are easily obtained on ebay.

Is it the detector for stabilizing the synthesizer output power?
In that case one of these LTC5596 demo boards with some opamp circuitry might work.
Or you could even try it open loop, connecting the bottom input and the top output.

[Kilo Tango]

I think it monitors the power out of the A400 module, which eventually goes to port 1. Here is a jury rig, pict6885 I fitted one of those Russian 3A121A GaAs detector diodes that are supposed to be good for 40GHz. shorted inputs together, pict6886 it sort of works !. In the service menu, pict6889 is the good detector, pict6890 is the jury rig one. pict6893 is a 3GHz LPF fitted between measuring ports. sort of usable !!!

Ken