250425 Board repair/One Flash on dead test

Disk drives, Monitors, SuperCPU etc.
DistantStar001
Member
Member
Posts: 9
Joined: Thu Jun 18, 2020 9:53 pm
Contact:

250425 Board repair/One Flash on dead test

Post by DistantStar001 »

As the title says, I'm trying to repair a 250425 board that's giving me a single flash on the dead test.

I've conducted extensive repairs to the board, removing and eventually socketing almost every single IC on the board (and at this point, all the ICs that went in them). The only exceptions are the Kernel and BASIC ROMs as I ran out of sockets. However, all the traces are intact and connected to what they're supposed to be.

I've checked and rechecked the sockets. As far as I can tell, everything that should be connected is connected, and there are no unexpected shorts on the board. Even so, I'm getting a single flash RAM error.

Sadly, I don't have a working oscilloscope, but my logic probe reads activity on the data lines of U9, U21, and U22. The others seem to be held high. However, my logic probe isn't that good (on is off, red is high).

I then tried removing RAM to check the errors on the dead test. Below is what I got:

U9: 7 7 √
U10: 5 5 √
U11: 3 3 √
U12: 1 1 √
U21: 8 8 √
U22: 6 6 √
U23: 1 4 X
U24: 2 2 √

In short, with U23 removed, I still get 1 flash when I should get 4.

As I understand it, that could be an issue with data bits 0, 1, 2, or 7. However, since removing U9 and U21 results in the expected flashes, that would seem to indicate that 0 and 1 are fine? I can't rule out 7, but 2 seem to be the most likely suspect.

I can also eliminate the CIAs, SID, and ROMs as likely suspects as they're not installed. I've tried two CPUs (6510 and 8500) with the same results. The PLA is brand new (PLAnkton) and the VIC II, as well as the 8701 have been tested good in another machine. Also, as mentioned, all the support logic is brand new as well.

Anyway, what am I missing here? Any help would be appreciated.


levoman
Member
Member
Posts: 20
Joined: Wed Oct 06, 2021 9:36 am
Location: Milton keynes
Contact:

Re: 250425 Board repair/One Flash on dead test

Post by levoman »

I am assuming that you have swapped the ram chips around to see if the error follows either u12 or u23 and have checked for data bus shorts to ground with a multimeter?

The results you are getting would indicate to me either a ground short on D4 ( u23 data ) or more likely a short between D1 and D4 . The PLA only handles address decode and does not touch the data and should not be the problem. Can you test for a short between D4 and D1 with the ram in and out (U23 and U12 ).

Could be that the problem is with a ROM .

Are U26 and U16 socketed?
levoman
Member
Member
Posts: 20
Joined: Wed Oct 06, 2021 9:36 am
Location: Milton keynes
Contact:

Re: 250425 Board repair/One Flash on dead test

Post by levoman »

According to this site D1 is U9 not U12??

https://myoldcomputer.nl/technical-info ... 5-01-c64b/
DistantStar001
Member
Member
Posts: 9
Joined: Thu Jun 18, 2020 9:53 pm
Contact:

Re: 250425 Board repair/One Flash on dead test

Post by DistantStar001 »

levoman wrote: Tue May 03, 2022 9:54 am I am assuming that you have swapped the ram chips around to see if the error follows either u12 or u23 and have checked for data bus shorts to ground with a multimeter?

The results you are getting would indicate to me either a ground short on D4 ( u23 data ) or more likely a short between D1 and D4 . The PLA only handles address decode and does not touch the data and should not be the problem. Can you test for a short between D4 and D1 with the ram in and out (U23 and U12 ).

Could be that the problem is with a ROM .

Are U26 and U16 socketed?
No dead shorts, but I'm getting 1kΩ of resistance between all the data lines and to ground. I compared this to three other boards (two of which were admittedly nonfunctional 2504070s) and they have no connections to the other data lines at all. The 2504070 boards had no connection to ground, but my other functional 250425 showed about 0.8 to 0.9kΩ.

In any case, I rechecked the back of the board and found that the insulation on one of the patches had come off and there was some conact with an adjasent pin (ont on U23 though). I adhusted the wire and now there's no longer any contact between the data lines, however, the dead test won't boot at all now.

Iguess I'm going to hace to trace all the data lines back to the CPU now (to say nothing of the CIAs, ROMs, and VIC II).

Also, I know it's not the ROMs since they aren't installed at the moment. My understanding is that the dead test can boot without them. Certianly it was if it could run a RAM test.
DistantStar001
Member
Member
Posts: 9
Joined: Thu Jun 18, 2020 9:53 pm
Contact:

Re: 250425 Board repair/One Flash on dead test

Post by DistantStar001 »

levoman wrote: Tue May 03, 2022 10:11 am According to this site D1 is U9 not U12??

https://myoldcomputer.nl/technical-info ... 5-01-c64b/
It is. But the dead test can present certain RAM failures as a fault in data bit 7 when the real fault is in either bits 0 or 1.
DistantStar001
Member
Member
Posts: 9
Joined: Thu Jun 18, 2020 9:53 pm
Contact:

Re: 250425 Board repair/One Flash on dead test

Post by DistantStar001 »

I just moved that exposed wire back in contact with the adjacent pin and am getting the single flash again. I MISSED A TRACE!!! The patch is on the 74LS257 at U13 (a RAM multiplexer I think). I'm going to have to follow all the traces again to figure out where that goes.
DistantStar001
Member
Member
Posts: 9
Joined: Thu Jun 18, 2020 9:53 pm
Contact:

Re: 250425 Board repair/One Flash on dead test

Post by DistantStar001 »

Success!!! I missed a floating pin (pin13) on U13. Hey, I just realized! Unlucky 13!!! But it's working now! :-D
levoman
Member
Member
Posts: 20
Joined: Wed Oct 06, 2021 9:36 am
Location: Milton keynes
Contact:

Re: 250425 Board repair/One Flash on dead test

Post by levoman »

Glad to hear it!!
Post Reply Previous topicNext topic

Who is online

Users browsing this forum: No registered users and 0 guests