The daemon Mac IIci: Can it be fixed?

[stephen_usher]

Found a bad connection on the VIA A9 pin.

I even got the machine once booting to the question mark screen once!

Got it into diagnostic mode and connected a serial port but because the the zero documentation I have no idea what the error code values mean, e.g.

IMG_3156.jpeg (302.9 KiB) Viewed 934 times

Is that the address of the memory error, or what it read back? Or what? No idea.

[stephen_usher]

I'm wondering if the problem here is an address line or RAS/CAS line issue when going to the RAM.

The trouble is that without the docs on that debug monitor system I can't test my theory and track down the problem by writing a word and then searching memory to see if/when it repeats.

[rubber_jonnie]

That's certainly an interesting step forward, I just had a look myself for that code and mostly got results where you were asking!

I also looked for a service manual, which I'm sure you've already done, and whilst I found one, boy was it poor. Can't help but think there are bad traces somewhere, reminds me of the leaky cap fix on my Amstrad Alt-386 and similar repairs to the A5000.

Can't recall if you said, are there any useful schematics?

[stephen_usher]

Only the rubbish Bomarc schematics unfortunately.

I may try to remove the SIMM sockets nearest the battery damage to see if the damage extends underneath them and possibly bridges two lines or something.

[rubber_jonnie]

Only the rubbish Bomarc schematics unfortunately.

I may try to remove the SIMM sockets nearest the battery damage to see if the damage extends underneath them and possibly bridges two lines or something.

Probably not a bad idea TBH. The battery chemicals are insidious and can get into all sorts of places without much trouble, and the little aluminium dustbin caps when they leak. I know my A5000 had extensive trace damage in the battery corner, but the caps leaked as well and caused damage across the board. The battery also killed ICs, though not obviously. so you may have to go on a cross board hunt to see if you can see any dull solder as evidence of other issues.

I suspect you may end up with a lot of bodge wiring to do.

[stephen_usher]

A lot of bodge wiring has already been done by previous people!

[rubber_jonnie]

A lot of bodge wiring has already been done by previous people!

I bet it has!!!

[stephen_usher]

Checked all the physical connections. Checked that there's no address bus shorts.

Given that whichever bank the memory is put the error is the same it can't be the RAS and CAS lines as they're all separate. The only commonality is the MDU chip itself and the 74F245s, which could be doing spurious things I s'pose.

I'll wait for the ROM SIMM to arrive and see if that changes anything. That should be here in the next couple of days if the Royal Mail manages to find Swindon. (It was misdirected to Croydon MC last night.) If it doesn't then I'll order a set of 74F245s from Mouser. (RS & Farnell don't stock them any more.)

The one really surprising thing about the Mac IIci design is the almost complete lack of pull-ups. Only the 68030 control lines have them, everything else just seems to float.

[rubber_jonnie]

Checked all the physical connections. Checked that there's no address bus shorts.

Given that whichever bank the memory is put the error is the same it can't be the RAS and CAS lines as they're all separate. The only commonality is the MDU chip itself and the 74F245s, which could be doing spurious things I s'pose.

I'll wait for the ROM SIMM to arrive and see if that changes anything. That should be here in the next couple of days if the Royal Mail manages to find Swindon. (It was misdirected to Croydon MC last night.) If it doesn't then I'll order a set of 74F245s from Mouser. (RS & Farnell don't stock them any more.)

The one really surprising thing about the Mac IIci design is the almost complete lack of pull-ups. Only the 68030 control lines have them, everything else just seems to float.

I wouldn't be surprised if it was the 74F245s, I had to replace the 245s on my A5000, even though they looked just fine.

[stephen_usher]

Yeah. But we'll see.

On the ROM SIMM front, it seems that Royal Mail can't find Swindon as the ROM has now travelled to Gatwick MC. It's doing a tour of southern Britain.