Jump to content

Tales from Midibobs workshop, Sys80


Recommended Posts

  • Replies 127
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

Came across a stash of NOS PAL's so I'll probably employ those. Didn't realise I had them, it's amazing some of the stuff I've got tucked away for a rainy day. Thinking about it, it was raining.

OK, so I've burnt my first PAL chip, plugged it in, and what do you know it works! Loading the file into my Data I/O had me head scratching for a good while but I worked it out in the end, no thanks t

Been sweating over two CPU boards, the first was a 12k with the usual rot everywhere. New sockets were needed for the CPU, the data buffer and the RAM. These are a real pain to remove, even worse than

Posted Images

Putting my Sys80 hat back on. This one came in as a no boot.

319256291_P1040026(Custom).jpeg.111de2f97adb674124dae852b500e5a7.jpeg

Didn't look too bad on first inspection but it was as stated dead as the proverbial Dodo!

The tester showed there were a couple of Address lines missing and there was no IRQ activity. The quickest way to check out what's working or not working is to use a known good board. Firstly a transplant of the RAM which proved it was knackered. The sockets were contaminated too they were also changed. Next thing was the CPU....

P1040041.thumb.JPG.6f7d67649b379e2f0e2b0e4f2b6c1316.JPG

When this was removed one leg decided it wanted to stay! Another leg was bent underneath and once I picked the chip up two more legs just fell off! With a replacement CPU it still did nothing although all Address lines were now pulsing nicely. Next thing was IC8 (74LS138) as this provides the CE pulse for the RAM but although it's inputs were good there was no pulse coming out. In with a new one and BINGO, it burst into life. IC12 was looking a bit rusty so for good measure that was changed as well even though it was working.

Nice when it all goes to plan for a change.:)

Link to post
Share on other sites
  • 2 weeks later...

I was given some scrap I/O boards recently to mess around with so I thought I'd see what could be done.

I'll refer to this board as Trigger's broom (Only fools and horses) as you'll see what I mean as we go along!

This one started out like this ...

P1030959.thumb.JPG.db7fa1a9a9e6d3af0f96dd73d65d1a65.JPG

Pretty horrific I think you'd agree.

P1030960.thumb.JPG.582d5ce546e1e20dc8a23b4ac467ad19.JPG

First thing was to see if all the muck and grime could be removed. Easiest method for that is shove it in the dishwasher!!
After a light 15 min wash it started to look a bit more like it would be worth working on.

P1030961.thumb.JPG.54d4344339cce16a35f9ee39dd30f7fa.JPG

P1030962.thumb.JPG.7ffef0d70d7e0fd91d1ce7139d3859b0.JPG

All the middle rusty resistors had to come out and a new 20 way socket was needed. Next thing was to power up and see what was or more likely wasn't working. Many hours later, this was the result ...

P1040054.thumb.JPG.a418e2bfaffb98d57c5fbcec4b0f8e85.JPG

Of course it now works a treat but the silicon count was rather high. So you see what I mean about Trigger's broom. :lol:

  • Like 2
Link to post
Share on other sites
7 hours ago, midibob said:

Except if anyone sends me one like that for repair they will need very deep pockets.:lol:

Money well spent if it gets your machine up and running..excellent work as always 😄 

Link to post
Share on other sites
  • 1 year later...

I've been working on 12K CPU which has been driving me nuts. The fault was no boot and I had done it before some time ago. On the test rig it was indeed no boot and during the process of testing the test rig started to play up and wouldn't run properly.

3 days later the test rig would only run the switch test and nothing else, the I/O card looked like it was playing up too so  it became impossible to test the faulty card.
Eventually I found the CPU had gone bad in the test rig, been running fine for ages so I didn't supect it.:headache:

At least now I had a stable platform to test the faulty CPU card. Most of the time it wouldn't boot and just sat there with two address lines stopped? All the buffer chips had been changed from the previous repair so I knew they were good. From there I proved that the RAM and CPU were OK. As it still wouldn't boot all I could do was to take out all the soldered chips and replace them. Once I'd done this it burst into life. Next thing was to put the old ones back in one by one. You're probably way ahead of me here and of course with all the original chips back in it still ran fine.:angry:
Flexing the board made no difference and I did it quite a few times but it stayed running.  Tried it just once more and it died!! A power reset started it up again and it ran fine.

I had previously tap'd out all the connections, data, address and links to the connector and all was good. There must be a corroded via somewhere but finding it will be a nightmare. Sometimes you have to make a judgement whether it's worth chasing these faults as the time involved is horrendous.
I'll see what it's like in the morning after a good nights sleep (me and the board).:lol:

  • Like 2
Link to post
Share on other sites

Thanks Andrew, yes I did, scope'd it and everything just in case and it was blameless.

Well, switched on this morning and .....nothing!! I felt great though.:lol:

So with the fresh head I went for the vias. Got round to the data links between the CPU and the buffer and the one from pin 12 (74LS245) was open circuit. I could swear I'd checked these previously and they were good (meter'd, scope'd and logic tester).

Under the microscope I could see a a corrosion black spot right next to the pin hole. Looking at the solder joints it didn't look like one of mine so I suspect somebody else must have changed the chip before.
A quick tac wire and bingo, solid as a rock. Phew.  

  • Like 2
Link to post
Share on other sites
  • 1 month later...

Been sweating over two CPU boards, the first was a 12k with the usual rot everywhere. New sockets were needed for the CPU, the data buffer and the RAM. These are a real pain to remove, even worse than the chips, especially as the solder has gone crumbly. All the chips needed replacing in the end which isn't unusal on these boards. Unfortunately the CPU ran cold so I knew it was duff too. Shame as these are becoming rarer by the day and that's if you can find a 'real' one. The whole job took just over a day to complete but it's swings and roundabouts in this game.:lol:

Next one on the bench was a 14K. Don't see too many of these and they can be a pain due to the custom PAL chip. Well in this case apart from being rotten the PAL chip had lost a leg (welded in the socket) and another leg was hanging on for dear life. Not so many bad chips on this one thank goodness but what to do with the PAL? I don't have any spares for these so a bit of creative thought was needed.

Thinking back to one of Andrew's earlier achievements of rescuing chips I thought I'd have a go with this one. Using the microscope I had to dig out a small recess to try and find what was left of the old pin. Where it had broken off there was no sign of any metal.:headache:After a while I could see something shiny but we are talking human hair widths here!! After a lot of messing about I managed to tack a very fine wire on to it but any movement and it would fall off. Eventually I managed to secure it to a thicker wire which went into the socket. The chip had to be into it's own socket to prevent any more legs falling off. Anyway, not knowing whether the chip was any good or not I carried on faulting. After locating a few broken tracks and vias I switched on (for the umpteenth time) and blow me it only worked.:clap::whistle:CPU was good too!!

P1040311.thumb.JPG.2ab600dab6a1ae5d35fe8e55c454d096.JPG

P1040312.thumb.JPG.269fea7753efecb5b6665a79ed1cd07a.JPG

I'm pretty sure these PAL chips will have all the security fuses blown so no chance of copying them but I will stick it on my Data I/O and see what it can read just in case, otherwise I'm stuck. Another day's work but very satisfying to see it run.:D

  • Like 3
Link to post
Share on other sites
16 minutes ago, midibob said:

Thanks Andrew, I did look for it but couldn't find anything, must have been looking in the wrong place.

You thanked Alan for the file here 

 

 

I've burned this file to gal using something like 'pal as gal" or pal to gal setting and it runs fine on the 14k card.

 

Nabbed a gal off an impact card. Shame I haven't sussed using the palce chip as I got loads of them.

Edited by Louie Bee
  • Like 2
Link to post
Share on other sites

Blimey you're right and I've got it. I forgot I had dementia!!! :lol:

Thanks so much for that Louie. Just got to work out the process now, I'll need to read up the Data I/O 29B docs as I don't know if it supports GAL?

Link to post
Share on other sites

Came across a stash of NOS PAL's so I'll probably employ those. Didn't realise I had them, it's amazing some of the stuff I've got tucked away for a rainy day. Thinking about it, it was raining.:lol:

  • Like 5
Link to post
Share on other sites

Bob I am amazed you've got life back into those green riddled boards, outstanding work 😲

With all of your threads I read (and haven't the technical know how of allot it), see some of the projects you work on and the state of the boards, and yet you manage to get them up and working again!  Brillant work Bob

 

  • Like 1
Link to post
Share on other sites

OK, so I've burnt my first PAL chip, plugged it in, and what do you know it works!:D Loading the file into my Data I/O had me head scratching for a good while but I worked it out in the end, no thanks to the documentation! It did however come back with a checksum error but I burnt it anyway and from what I can tell on the test rig it appears to run OK.

Flushed with success I thought I'd just see if I could read the original chip and blow me it wasn't protected?? The file off the chip is slightly different in the fuse map so I don't know why that should be. This is all new to me so it's a bit of a steep learning curve.

Anyway, here's the file. I've added the ETX and STX characters to the beginning and end of the file so that may upset the checksum?

Read from IC10.jed

  • Like 3
Link to post
Share on other sites

Announcements


×
×
  • Create New...