800XL Sudden Death

[flashjazzcat]

I just turned on my stock 800XL with the missing MMU chip: booted up with a blank red screen (hardly surprising). Fitted new MMU chip. Booted up and worked perfectly. Socketed OS ROM. Boots up to blank red screen again. I've been over all the solder joints three times. Has this thing just spontaneously died? Seems odd that the symptop is the same as having no MMU. I've no reason to suppose the IC is bad, since it worked perfectly when fitted and hasn't been touched since (apart from checking the soldering).

Edited January 28, 2010 by flashjazzcat

[+Roydea6]

I just turned on my stock 800XL with the missing MMU chip: booted up with a blank red screen (hardly surprising). Fitted new MMU chip. Booted up and worked perfectly. Socketed OS ROM. Boots up to blank red screen again. I've been over all the solder joints three times. Has this thing just spontaneously died? Seems odd that the symptop is the same as having no MMU. I've no reason to suppose the IC is bad, since it worked perfectly when fitted and hasn't been touched since (apart from checking the soldering).

 

Now that the OS is socketed you can try a different OS rom chip there and see If you still get the red screen.

[Rybags]

No MMU should mean logic not there to select ROM when appropriate, therefore no startup occurs - GTIA and Antic remain in powerup state.

 

Not present or bad OS ROM should mean exact same symptoms, I'd reckon.

[flashjazzcat]

Funnily enough, I tried THREE different OS ROMs and the MyIDE adapter: same result with each, so I assume the OS can be ruled out. Next suspect would be the socket, but as I say, that's been checked three times.

 

 

 

[flashjazzcat]

Most weird: since the stock 800XL's dead, I decided to socket the MMU in the 1MB 800XL. I tried the GAL MMU Lotharek kindly sent me, and I got exactly the same result as with the stock machine: red screen on boot. Fearing I had two dead machines, I tried the (last) spare real MMU I have (pulled from my damaged 65XE), and it works fine. I appear to be in the very odd position of receiving - as donations - two faulty MMUs: one "real", the other on a GAL. To confirm this, I'll try an original MMU in the stock 800XL (odd, though, since the one Spookt sent me worked in that one for a few minutes at least).

 

I'm running short of MMUs yet again (the one I just removed before fitting the socket on the 1MB 800XL lost a leg on exit: it was astonishgly hard to shift). I'm now down to three working machines out of five!

Edited January 28, 2010 by flashjazzcat

[Rybags]

I wonder if an MMU could be tested easily outside a machine... ie mount in a proto board then use PORTA I/O to verify the logic.

 

Possibly not enough I/O bits available but it might be able to be done in stages, rewiring between each one.

[flashjazzcat]

I wonder if an MMU could be tested easily outside a machine... ie mount in a proto board then use PORTA I/O to verify the logic.

 

Possibly not enough I/O bits available but it might be able to be done in stages, rewiring between each one.

Interesting. I don't doubt the provenance of the new chips, but something's not right. I'll ask Lotharek if he can verify the GAL somehow, of perhaps flash another. I'll order another couple of 20 pin sockets and put one in the 800XL. Both MMUs will be easier to test if I can swap them out without desoldering. The MMU seems to be the epicentre of many useful mods, and it's been useful to have it socketed in both "main" machines.

Edited January 29, 2010 by flashjazzcat

[flashjazzcat]

Socketed MMU this morning and tried the GAL chip in there: still a red screen. Since the "good" machine didn't work with the GAL either (but it did when I put an original MMU in there), I have to conclude that both MMUs are bad. 800XL is dead on the scene until I can source another chip.

 

 

 

[flashjazzcat]

Update on this: even though a continuity test of the OS ROM socket revealed no problems, I went over the solder joins yet again and the system suddenly booted. This is with the MMU that spookt very kindly sent me a few weeks back, which I inexplicably couldn't get working in any of the machines I initially tried. Apologies to spookt for inferring that the chip didn't work: clearly it does, and I'd like to thank him again for donating it.

[spookt]

Well I wouldn't have been surprised if it didn't work, it's been sitting in my spares box for goodness knows how long. It was new and not a pull but I had never tested it. It's good news that *one* of your machines is working now

[flashjazzcat]

Yeah. At least the day (well, most of the week, actually) hasn't been a complete disaster. I think I'll leave MyIDE alone for a while: it's just not working for me and it's consumed a huge amount of time I could be spending on other things. It's actually worked its way through four of my five machines since I got it, and it hasn't been functional in any of them. I'm going to pack the board away before it makes a dash for my 1MB VBXE machine!

Edited February 6, 2010 by flashjazzcat