TDS220 Repair Report

[george.b]

I got my hands on a faulty TDS220 recently, which I managed to repair, and thought I'd share a report on what was wrong, in case it's of any use to someone else out there.
There was no probe comp signal present. Power rails were good. Main clock, CPU clock and display clocks were present, there was data flowing to/from RAM, but none flowing to/from the CPU or ROM.
Probing around, I noticed the 68000 would go into halt 10µs after /RESET went high. I presumed it was fetching invalid data from ROM, throwing an exception, then throwing another exception when trying to handle the first exception, which caused it to finally halt. I'm not sure this is what the 68k does, as the manual isn't terribly specific about what causes it to assert /HALT - in any case, that's the hypothesis I went with.
The main board is a newer revision and has a BGA EEPROM chip. Reflowing the EEPROM seemingly brought the board back to life; however, there was still nothing on the display, despite there being data flowing on the display data lines.
I noticed the -24V line going to the display was dead, even though the -24V line from the power supply was good. I couldn't be arsed to figure out the circuit for the LCD -24V line so I just bodged it straight to the -24V line coming from the PSU.
That made some stuff show up on the screen, as seen here:



I reckoned there was something wrong with the display - and indeed there was: a faulty electrolytic cap. After replacing it, the scope came back to life, as shown in the attached picture. I'll see about replacing the display polarizer, as it's in a sorry shape. Yes, the display is inverted - lowering the contrast makes the colors the right way around, but the middle area becomes almost unreadable, due to the bad polarizer.

[george.b]

Just a small follow-up on the display polarizer: I replaced both polarizer sheets with polarizer from a broken LCD monitor. The colors are a bit off, I think TFT and STN polarizers are different somehow. It's not stellar by any means, but at least it's usable now

[feedback.loop]

Interesting. Once I replaced polarizers in HP ESG-D2000A and the repaired display also has this pinkish tint to it. I wonder what is the reason for this and how to find the right polarizing film to restore the original look. I bought a sheet of a self adhesive film of the suitable size on eBay from China.

[george.b]

I have repaired other LCDs by replacing the polarizer before, and the color was also off. The HP 200LX screen I repaired got a green tint. I don't know whether it's the polarizer, or the adhesive they use, or what. Would be good to know.

[george.r]

George #2 chiming in, owner of the 1st Tek scope from the same batch, also brought back from a similar causa mortis (no VBias to LCD).

On polarizers:
I did notice that sticking a piece of clear tape to the polarizer did shift its polarization angle and the hue of the light passing thru - this was a bit of a serendipitous discovery as I was simply using polarizer as backing for a QFP chip. Peeeerhaps buying the proper self-adhesive stuff, or using proper optically clear adhesive rather than having air in the interface will solve that. We may try this on George's present LCD if we get the 3rd, better-but-currently-damaged panel up and running

Previous research by me while trying to figure out toasted LCDs (was this also during HP 200LX repairs?) had shown quite a few different types in composition and construction to exist, their application varying with the type of matrix they were being glued to.

As a bit of trivia, it's allso worth noting not every device uses two linear polarizer sheets: a Game Boy Color will use a reflective layer at the bottom, and a circular (λ/4) polarizer up top.