8562A Display Glitching

[rogersstuart]

I have an HP 8562A that I don't use very often. I turned it on the other day and I guess it didn't appreciate my long absence because it decided to break. The CRT was pretty sharp and didn't have any burn-in. I assume this is the result of dried capacitors? It seems to work fine over GPIB, so other than the display problem and needing to be calibrated it's fully functional. The fault is intermittent right now so I think that I should do something about it before it gets worse. Does anyone have experience with this? Is there a pot that needs to be tweaked or a specific cap that always goes bad?

[rogersstuart]

I decided to open the unit and have a poke around. My analyzer seems to be one of the original 1987 6 ROM models and reports that it has close to 10k hours on it. There was a tamper sticker which indicated some maintenance or a calibration occurred during 2019 but I can't be sure of what occurred or why this unit was defected and sold to me for such a low price. The only problem when I received it was a dead fan which was causing overheating after 30 minutes.

I looked though the service manual. Based on what I read I decided to probe the A2 board near the line generators. The fault is intermittent which is a bit of a hassle since I can't reliably trigger it. There's definitely something wrong. The waveforms don't exactly match what is shown in the manual when the unit is displaying the CRT test image. The LCHAR signal stood out. There is no change when I look at it with the test screen displayed (there should be) but it looks fine during normal operation.

I didn't see any power rail test points on the A2 board explicitly called out in the manual. I probed around until I saw a few stable voltages and then checked them in AC mode on the scope. None seemed to indicate that an LDO is oscillating from a dried cap. Even when the display was messing up. I tried tapping all the line generator components with a plastic tool. None of them seem responsive. Spraying them with freeze spray did get a response but I was unable to localize it. I can only say that spraying that region of the board with freeze spray seems to correct the severe glitching and has a stabilizing effect until the unit is power cycled. Last night I was able to stabilize it for hours. Today I found it bouncing between working and not working after being left alone for some time.

What could it be? A cracked component would have responded to the plastic tool right? As would a broken trace? Could it be a semiconductor or even a bond wire? I don't have much experience with this so I may have to live with it.