HP 8720C VNA - A58 Modulator / Amplifier / Detector / Splitter - how to test?

[John_ITIC]

Hello, I'm continuing the repair of my HP 8720C VNA. Per the attached images, I'm seeing good input to the A58 M/A/D/S module but no output. I'm trying to determine whether it is the A58 module or the A59 PCB that controls it that is faulty. The attached images contain the voltages and pulses I'm seeing on both A58 and A59 sides. One of the pulse trains seem to have odd high/low signal levels (-5V and -13V) while the other has 0V and -5V. This could be correct if using some strange logic technology. But it looks strange to have two different sets of voltage levels. I'm not familiar with this unit so i'm hoping that someone could help me confirm whether something looks incorrect. I also specify the resistance I'm measuring on the various A58 pins.

Also, has anyone opened one of these A58 M/A/D/S units to see whether it is serviceable? It looks like it has a lid that is glued, double-side taped or potted shut.

Thanks!
/John.

[John_ITIC]

I found the issue. The A58 M/A/D/S unit was not powered by the A59 Source Interface Board due to a burnt resistor which was open.

The burnt resistor is a 2.2 Ohm, 3W, wire wound resistor and is on the input on the LM117 linear voltage regulator. VIN=15V, VOUT=9V.

The only reason I can see to have such a resistor on the input-side of the linear regulator is to filter out high-frequency components in the power rail before using the 9V to feed the amplifier in the A58 unit. The resistor must be used as an inductor.

An interesting finding was that the A58 M/A/D/S draws 1.3A from the 9V VOUT from the LM117. Given ohms law, 3.7W should be dissipated by the 3W resistor. It seems odd that that extra 24% of power could have caused the resistor to fail. I ordered a 5W replacement resistor, just in case.

Anyways, for now, I have replaced with a piece of solid wire and the instrument is able to do basic measurements. The attached image shows a simple (uncalibrated) S11 measurement of my 18 GHz cable going into my 22 GHz spectrum analyzer. I will next continue checking it out via detailed self-tests and do any needed adjustments.

/John.

[mankan]

Thanks John for documenting your repair here. I'm currently looking into phase lock issues in my 8720C and I've narrowed it down to A58 M/A/D/S having way too low output, it's input is good. So its either A58 or A59 source control board or both that are faulty. When replacing MADS with a 6dB splitter the phase lock is good, i.e. the R sampler gets enough power both in low and high band.

So I wonder if you have any additional info about A58/59 to share?

At the moment the work got interrupted by the Rifa X capacitor in the PSU started letting out the magic smoke but is was caught quickly.

[mankan]

I've written down my A58 M/A/D/S repair adventure, here.