e4402b Spectrum Analyzer doing funny things when the 5dB attenuator switched in.

[jwrodgers]

Hi, I recently repaired a LO Unlock error on a E4402, pictured below, by replacing the div4 chip in the LO section. It now will go through the Align Now, All Alignment process. I noticed that I was getting a good reading with the Attenuation set at -10 dB, the setting after a PRESET. But as I changed through the attenuation levels, anything with the 5 dB attenuator switched in was giving a funny reading. Screen shots are below.

Anyone got any idea what could be causing this. I believe there is a mechanical solenoid attenuator block, with I presume a 5, 10, 20 and 40 dB attenuators in them. I can hear them switch between attenuator settings.

I used a 2 GHz ref signal at -20 dBm from a signal generator for all the tests. Any signal shown with an attenuation ending in 0, e.g. 10, 20, 30, 40 etc show a power level in the signal of approx -21 dBm, (I am using a rubbish RG58 coax at 2 GHz, so a few dB loss is expected.) When the attenuation ends in a 5, e.g. 5, 15, 25, 35, 45 dB, the indicated power level is all over the place!

Is this a typical fault that could be in the attenuator block? Maybe there is another cause? Could I repair it by removing, disassembly and perhaps cleaning or adjusting the attenuation block? Maybe a new attenuation block is the only way to go?

Thanks for any ideas/comments in advance.

[jwrodgers]

And some more screen shots....

[edp]

Hi jwrodgers,
I haven't disassembled my unit that far so am not sure what the attenuator looks like, but as far as I can see from the CLIP it is discrete components - it will be enclosed in the rf deck.
Suspect the relay k5 or one of the nearby associated passives.
See snippets of interest.
ed

[edp]

oh sorry my mistake - ignore previous post - that is from the 1.5GHz rf board, 3GHz+ does indeed have separate physical attenuator unit.