I finally got around to working on the atari 800xl that I mentioned back in October. This machine has been a nightmare.
To start, it would power up to a slightly reddish black screen. I couldn’t find any good info online about this particular ailment, so I decided to just start swapping chips between the atari 800xl and the 600xl that I have.
I’m not sure if the 600xl is in working order. I don’t have anything with an RF input that I can test the 600xl with, but luckily all of the chips in the 600xl were socketed. Unfortunately, that was not the case with the 800xl.
I swapped the few chips that were socketed, like the ANTIC, GTIA, Sally, and Pokey, but no luck.
At this point, I figured that the ram was a likely problem, so I desoldered all of the ram and installed sockets and new ram chips. This board had a weird extra resistor bodged in for each ram chip. You can see the broken trace on the bottom of the board in the image below followed by the resistor on the top of the board. I left it as is when I installed the sockets, but later on, I bridged the resistor with a jumper wire.
New ram didn’t fix it, so I desoldered all of the rest of the chips on the board and installed sockets. I then swapped each chip individually and tested.
At some point during the testing, I went from a black screen to a screen with stripes, and full white screen, and variations between. The screen changed every few times that I powered it up, but there were no other signs of life.
I had no real way of testing the chips in either machine, but I was going under the assumption that some combination between the two machines should work, but it seemed that everything that I tried left me with the same result.
I was running out of ideas, so I just started replacing other active components like the transistors, common logic chips, etc. I touched up all of the suspicious looking solder joints, and tested continuity on all of the buses.
At some point, I started thinking that there may be an issue with the rom chips. I pulled each one from both machines and dumped the roms and bingo. The basic roms looked okay, but both OS roms seemed to be bad. Or did they…
Since they both read bad, was I just reading them wrong?
I guess an easy way to check is to just replace it with a known working rom, but I didn’t have a pin compatible rom chip. I did have an AT28C256 that was close.
I wrote an OS rom binary to the AT28C256 and I bent pins 1 and 27 up 90 degrees. I bridged pin 27 and 28 together to keep WE high and I soldered a jumper wire from pin 1 (A14) and connected it to the socket’s pin 27.
I installed my makeshift rom and powered up.
Voila! It worked!
At this point, I was tired of this machine, so I quickly re-assembled it for a final test.
Still good. It still has a few sticky keys on the keyboard and could use some new capacitors, but for now, I’m going to put this one aside and work on some of my other recent aquisitions.