Electronics > Repair
Another TDS 5xx/6xx restore/repair
Russ.Dill@gmail.com:
So I needed a first decent scope, and noticed that capacitor problems that TDS 5xx/6xx are having and figured I'd pick one up for cheap and repair it.
The description was pretty vague:
--- Quote ---The unit is not in working condition.When I turned on the unit all indicates of panel are lighted up then the system kept on status for 24 hours.The screen showed some words failed initial calibration.I think the unit need some repair jod in order to make it function.
--- End quote ---
But I was able to get it for $250.
From what I understand with bad caps, they can overload whatever working components or power supplies there still are, so I didn't risk turning it on, but instead took it apart...and ran into a problem:
I'm not sure what happened to bend this steel like this, but the case would not come off. I basically had to take apart enough of the scope with the case still on before being able to get the cover off.
The next thing I noticed was that plastic components, headers, and even the top of the front panel were melted. No idea what happened here, but people, please, stop abusing your test equipment:
So the next obvious thing was the caps, or more accurately, the electrolyte from the caps. It was pretty much everywhere (in the last image, caps have already been removed, the deformed header is actually due to the before mentioned heat damage):
So next was removing all the electrolytic caps, about 80 of them. I figured I'd speed things up by using hot air. That'll go fast, right? ABSOLUTELY! The caps literally leap off the board ... at your face ... at high velocity. This was a really bad idea. Electrolytic caps have a tendency to explode, old leaky ones even more so. Mental note, when desoldering components, wear safety goggles.
For the remainder of the capacitors, I poked a hole in vent hole in all of them, and *then* removed them with hot air, which actually went pretty smoothly.
Ok, so next step, cleaning the electrolytic off the boards so it does no further damage. The first step was a scrub down with an ESD brush and a 1% liquinox detergent. Next, thorough rinsing with distilled water. Then, another scrub, but this time with 91% isopropyl alcohol. And again, a rinse with distilled water. This was followed by air drying with warm air (150C). There was *alot* of board to clean. Two main PCBs: (Before decap and cleaning)
There is also a serial/parallel interface board, and the front panel board that also have electrolytics that needed replacing. For the front panel board, I removed the rotary dials and the speaker before cleaning.
After that was a lot of work with a good chisel tip and a desoldering braid to clean old solder off the pads:
Next I probed all the capacitor pads to check for any that had broken traces or vias. I found a single bad via on the front panel board:
I don't have a pic, but I scraped back the soldermask on the trace and pretty much all of the copper underneath it was eroded away. Anyway, I got some wire, passed it through, and soldered it on both sides:
...And then it was time to solder 80 caps. Joy. But then hooray, I was done, time to power it on. Ok, click, whirr.....all the lights come on....but stay on, and nothing shows up on the screen....boooo....
Investigating the service manual showed that the firmware update switch was in the program instead of run position. So that was a simple fix. Try again, and yay, it powers up, but the diagnostic screen is a what's what of things that can go wrong with your scope. The primary concern was that the large acquisition board on the bottom of the scope could not be reached. The service manual recommended swapping out the acquisition board to see if that fixes it, and if it doesn't, swapping out the processor/display board instead. ...that's not happening.
I checked the connection between the boards, and it seemed tight, but I reseated them anyway, and viola, that error was gone. But, not only was calibration failing, but there were other failures on the aquisition board as well. My 1kHz test signal looked like this:
...one of these signals is not like the other....Anyway, I took a peek at the error log, I could at least see that my work on the scope had eliminated some of the errors that were occurring previously on the scope, before I got it. Anyway, it was telling me:
--- Quote ---ERROR: diagnostic test failure
Calibration failed
ERROR: diagnostic test failure
checkIfCalFailed, Calibration failed
ERROR: diagnostic test failure
digA2DDiag, failure in checking D4D7 bins
--- End quote ---
The last one sounded particularly bad, and I couldn't find reference to it anywhere. I did notice the "A2D" and the D4-D7 though. So some A2D converter was having trouble with some of its signals. Ok, time for a close visual inspection of the Acq board:
Ouch. The via on the bottom is just gone, well, its still there, its just now a pretty looking blue powder. Ok, time for some more via repair, and since the top one is looking sad too, why not fix that one as well:
And power it up again:
--- Quote ---ERROR: diagnostic test failure
Calibration failed
ERROR: diagnostic test failure
checkIfCalFailed, Calibration failed
--- End quote ---
Well, I certainly fixed the D4D7 thing. Reading up on the remaining two errors, I found out that all I needed to do was wait till the scope warmed up, and re-run the calibration sequence. So i did that, and:
Tada, working scope. You might notice that the display is dim and there is some color splotching. This is not due to potato. These CRTs tend to get dim over time, and the LCD shutters in front of the CRT that provide color can be damaged:
(Taken during startup sequence when screen is flashing through multiple colors). Ouch, again, someone did not take good care of their test equipment. No matter, I was already replacing the CRT with a LCD anyway. In the meantime, the scope still works fine with a cruddy screen, and I can also hook up a VGA monitor to it. As far as the case, hopefully someone I know who does auto body repair can do something to smooth it out.
Its certainly a fun project, and you can get a lot for your dollar, but of course, you could also end up with a completely worthless paperweight. When my LCD parts come in, I'll follow up with some images from that project.
nctnico:
This model is pretty useless though. It doesn't have peak detect so it misses short pulses on low sampling rates.
One of the problems with replacing the capacitors is that there are two values: 10uf/35V and 33uf/10V. Some of the 10V capacitors have 12V or 15V on them so choosing capacitors with a slightly higher voltage rating doesn't hurt.
Hydrawerk:
I like that 7 segment LED display.
Russ.Dill@gmail.com:
--- Quote from: nctnico on December 31, 2013, 12:10:52 am ---This model is pretty useless though. It doesn't have peak detect so it misses short pulses on low sampling rates.
One of the problems with replacing the capacitors is that there are two values: 10uf/35V and 33uf/10V. Some of the 10V capacitors have 12V or 15V on them so choosing capacitors with a slightly higher voltage rating doesn't hurt.
--- End quote ---
It was actually cheaper just to order 100 33uF 35V caps than to try to mix and match.
nctnico:
The problem is that some capacitors are not used for decoupling but in analog circuits like the ones on the frequency synthesizer. You really have to replace the capacitors with the right values in these scopes. The same goes for using ceramic multilayer capacitors. It is tempting to use those but some of the circuitry is designed for capacitors with a high ESR.
The 7 segment display is for diagnostic purposes.
Navigation
[0] Message Index
[#] Next page
Go to full version