TDS520 disappearing waveform

[MichalPL]

1. I've done my homework
Hi guys,
I bought a dead Tektronix TDS520 500MHz Oscilloscope for $60 (yup!) and I brought it back to life after reading everything I could ever find on the Internet regarding TDS520 or TDS540.

2. But still
Anyway, despite my efforts I’m still getting:
Acq/Proc interface FAIL
and
Acquisition FAIL
at boot up and I’m also getting a FAIL when I try to perform the SPC (signal path compensation) procedure.

The error log, as read using GPIB, shows just "ERROR: diagnostic test failure, acqdtaconf, ** addr = 0x7300000 exp data = 0x7fff actual = 0xffff, *** WAS TESTING DIG A ***"

3. Waveform intermittently disappears
Interestingly, only once in every 3-4 boot ups I can see the waveforms on both channels 1 and 2 and the scope seems to be working correctly (although still with the same errors and probe treated as if it was a x1 one). On the other boot ups, I see no waveform on any of the channels, just the grid.

4. What I've done so far
I’ve recapped all boards and repaired tracks in (hopefully) all layers of the PCB where a cap exploded and left an ugly hole. I’ve also washed everything a couple of times using lots of isopropyl alcohol and ESD safe brush. Apart from that, NV RAM was replaced with a brand new one and everything was reset to factory defaults (yes, I’ve also deleted the calibration values after making a backup of them).

I even tried to swap the original acquisition board with the one from TDS540 but at boot-up the scope protested that it can see 4 channels whereas it’s been designed for 2. The error stated ‘hardware incompatibility’ or something like that. If you know a way to trick it into working with a TDS540 acq boards, please let me know.

Anyway, before I start desoldering ICs at random (to replace them and/or check traces that run underneath them), I thought I’ll ask you guys for advice. Maybe this intermittent problem with waveform display will be particularly helpful at arriving at the cause? Please help.

5. What the voltages look like
I’ve also measured voltages on all caps and most ICs on the Proc and Acquisition boards.
In most cases it’s 5.1V, 10V, 15V or 30V, so normal, I assume. Also, if somewhere it’s 4.9V, I guess it’s acceptable.

I’m attaching photos of both boards, where I marked components which voltages seem dubious to me. Maybe they’re normal, I have too little experience with these scopes to tell, but they definitely differ from the voltages mentioned above.

[humgar]

Your board is very corroded so there might be weakly conducting paths or flimsy traces that you overlooked.

First, try washing the board thoroughly again with de-ionized water, then dry it with a leaf blower, then spray 99% alcohol all over it to pull remaining water out, then dry it again with a leaf blower, and leave it under the sun for half a day.

In the past, I had a board or two that I thought it couldn't be repaired but it was just me being too lazy to wash the entire board: I only cleaned the surrounding impacted area with rubbing alcohol like you did.

Don't get your hopes up though, but you might get lucky if the electrolyte hasn't eaten into the traces enough to be symptomatic. It's not likely to be a chip you need to replace unless you saw visible damages. If you see visible chip damages, it's likely because somebody kept using it for a while after the electrolyte leaked, shorted certain paths and overloaded certain chips.