Fixing my C128D

[PhilC]

74LS244's are prone to dying. It's not a MOS part is it? MOS TTL chips are the worst possible for reliability. I'd preemptively replace any on the board with new TI versions.

It's not a MOS part fortunately and once I got it out and tested it it actually passed the TL866 test. I'll probably put a fresh one in it anyway. I have learnt something the input side of the chip on pin 13 does respond to a reset but the out side pin 7 doesn't, so i'l off to find out what else is on pin 10 and has died :lol:

[stephen_usher]

Oh, what fun.

It's whack-a-mole time on the address bus!

The pull-up resistor's OK isn't it? Just a random thought. Then again it would be floating if not pulled up and not driven low.

With the machine off what's the resistance between A10 and ground?

[rubber_jonnie]

So doing a bit of fault finding it looks like one of the buffers is potentially not working, so I'm going to remove that and put it in the chip tester.

Of course it could be something else on the output side of the buffer causing the issue but everything else on that buffer chips seems to work.

Wish me luck!!!

Screenshot 2023-09-12 at 14.15.39.png

p.s. its pin 7 for those that are interested. we have a pulsing waveform on pin 13 but other on 7.

Nice find :)

Be good to see how it tests in the chip tester.

[PhilC]

@stephen_usher actually A10 is 3.5 ohm to gnd whereas the rest are about 9.5K

I was just socketing the Z80, a couple of legs fell off of it, no joke, so its been replaced but still don't work :lol:

Well I'll get my fault. finding moneys worth with this no doubt

[stephen_usher]

Start removing one chip at a time which is connected to A10 until the short goes (if it does). Unfortunately that's the only way unless you have a very accurate multimeter which can measure very small resistance changes. If you do then you can try to find the chip[ with the lowest resistance to ground and that will be the region where the short is on the board.

[PhilC]

Start removing one chip at a time which is connected to A10 until the short goes (if it does). Unfortunately that's the only way unless you have a very accurate multimeter which can measure very small resistance changes. If you do then you can try to find the chip[ with the lowest resistance to ground and that will be the region where the short is on the board.

Thanks Stephen, I realised that's what you were going to suggest, so tried all the socketed chips on A10 and it stayed the same, resistance lowered as I got near to the 8502, so I've desoldered it and measuring between A10 and GND on the chip gives me the short to gnd. Well, good job I've ordered one, even if it was bloody expensive!

[stephen_usher]

Oh dear! Well, at least you found the culprit.

What happens if you try to boot to CP/M with the 8502 removed?

[PhilC]

Oh dear! Well, at least you found the culprit.

What happens if you try to boot to CP/M with the 8502 removed?

I'll have to reattach everything and fine the CPM disk first :P. but will try in a mo

EDIT: Well I get a tiny bit of disk activity and then nothing, LOL, so will try again when the 8502 arrives.

[rubber_jonnie]

Good bit of detective work there :)

I once had a weird short on a BBC B and it was under an IC in a socket, put 5v straight into the video signal. I had to cut the track either side of the short and bodge it. Funny old thing, but I got video after that :)

Not sure if you watch Adrians Digital Basement, but he was working on an Apple ][ clone that had shorts underneath some sockets. He had a bit of a job finding them, not to mention one under an expansion slot.

[PhilC]

@rubber_jonnie haven't found any under sockets on this board so far but have found a few soldered pins that were left too long and just about touching the one next to it, that and the case of the rust monster leaving his foot prints everywhere.

Once I’ve got this board booting again I’m sealing the whole thing with some pcb lacquer.

Hopefully the 8502 comes tomorrow.