Service A8 computers

[russg]

I'm re-posting a conversation about servicing A8s from Sept 14 that seems to be well down the list.

I'm trying get help fixing my A8s.

 

• 
• 1,212 posts

• Location:Cleveland, Ohio

Posted Yesterday, 2:50 PM

ricortes, on 10 Oct 2014 - 10:57 AM, said:

 

I haven't seen the Sams photofacts for the computer series, I just had the disk 1050 drive one. Sams had a lot of wave forms/oscilloscope pics and methods for the 1050 anyway. IMHO, they are better in detail then the other sources like the Technical Ref. They don't go into details with the why and howto like an instructional manual, just info on diagnosis and fix.

I have Sams for the 800xl. It has wave forms on the schematics. The thing is, I have a working 800XL right next to the busted one.

I can compare waveforms on the working one with the busted one. I also use a logic probe.

For instance, I'm pretty sure this is significant, but I don't know what it means. Pins two of three of my drams are low (the first, second and fifth drams from the bottom).

All the other pin two drams are high in the busted one. On my good 800xl all pins two are pulse. I get high pins all over the place that are supposed to be pulse.

But I don't know why. I've tested the clock waveforms, they seem to be OK.

Edited October 11, 2014 by russg

[Rybags]

800XL uses 1-bit Rams and all 8 are involved in any Ram access. As such, activity from each should be reasonably synced to the others.

 

If you have a good 800XL too then you could test the chips on it - if all the Ram turns out good then it's probably a problem in the memory select logic somewhere. Probably start with the MMU then work along the chain. Schematics come in handy there.

[ClausB]

Pin 2 of each DRAM connects to the data bus and it should be pulsing. If all the pins 2 are stuck high or low, it is unlikely a RAM problem, since many other chips connect to the data bus, such as the ROMs and all 40-pin chips. Replace them one at a time from your good XL. If they are not socketed, well, then good luck with that.

[russg]

Pin 2 of each DRAM connects to the data bus and it should be pulsing. If all the pins 2 are stuck high or low, it is unlikely a RAM problem, since many other chips connect to the data bus, such as the ROMs and all 40-pin chips. Replace them one at a time from your good XL. If they are not socketed, well, then good luck with that.

I have four 800XL motherboards. But right now, only one works and the drams aren't socketed on that one. I'm trying to bring up at least one fully socketed one.

[russg]

I have four 800XL motherboards. But right now, only one works and the drams aren't socketed on that one. I'm trying to bring up at least one fully socketed one.

I got a socketed 800XL, removed a Newell 256k and put back the 64k. I tested my bad board and found two 64k drams bad and a LS158 (U27) bad.

I've now swapped all the chips except the keyboard 4151s, including the GTIA-40 pin chips. I get a good boot now, it can load a APE disk drive

and makes the pbblllt sound of BASIC READY. But... the display is corrupted. While booting up it shows a black screen with GRAPHICS 2 big letters

scattered on the screen, then when bootup is finished it goes to a blue and white screen of gibberish. It will boot PACMAN, but the screen isn't

correct. I can play PACMAN, gets the correct sounds. I can see pacman and the ghosts, but not the maze. The background is flashing

red and blue and there's a flashing yellow line. Two white graphics 0 size lines at the top. A corrupted display, like the display list is garbage.

I guess player-missle graphics work, pacman and the ghosts are visible and the correct dying sound when pacman hits a ghost.

Edited October 13, 2014 by russg

[kogden]

ANTIC issue maybe?

[russg]

ANTIC issue maybe?

I swapped Antic. I will do a logic probe of all the pins, from Sam's.

I have found LOTS of bad drams on different motherboards today.

I have a working mb and just test them one at a time. I was pulling drams from a bad mb.

Anyway, I found five bad drams. Is there something wrong with my LS158 or otherwise that can eat drams?

(I just logic probed the Antic, all pins give good results.)

Edited October 13, 2014 by russg