Repair Log: Joust MPU board back from the dead

YellowDog

In Memoriam

Donor 2011, 2013
Joined
Apr 12, 2008
Messages
9,851
Reaction score
158
Location
Humble, Texas
I thought I would post this as a help to others who want to try ressurecting dead Williams boards.

I recently received a Joust boardset from a fellow KLOVer for repair. It had been sold to his as working, but as he found out, it was not. The first clue was that the CPU was inserted upside down in the socket. He obtained a replacement CPU chip and tried it with no results. So he sent it in.

When I first examined the board, I could tell that the batteries had exploded at some point in time. There was some corrosion on the chips next to the battery area. I wiped it down with alcohol and set it aside to dry.

Later I plugged it into my bench setup and it was dead. The CPU was not getting a clock pulse, the reset was pulled low regardless of the reset button being pressed.

So I stared at the beginning. The crystal was pulsing. But the signal was not getting past the first set of inverters (7J). I replaced the chip (after socketing it). Still no clock. I checked the first flip-flop (7K) and it was not toggling even though it was getting the clock pulses, so I replaced it.

I checke the CPU and I was not getting clock Q or E. Those traced back to another flip-flop (7F), so I replaced it. I still didn't have Q & E at the CPU. After a lot of head scratching I did something stupid, I replaced the CPU. This time when I powered it up, I got something on the screen! Wahooo! Progress. It was just a yellow screen, but it was the first visible sign of progress.

When I started probing again, I found that the watchdog was toggling the RESET pin and we had clocks Q & E, but no diagnostics on the ROM's LED. [fast forward through a lot of boring logic probing] I found that there was no activity on the data bus D0-D3. That sounded suspicious, so I replaced the CMOS chip (1C). When I powered it up, I got nothing. After some poking around I found that the 6MHz clock was not being generated so chip 4D (another flip-flop) was replaced.

This time, I got the 1-3-1 error on the ROM card. And the Reset switch was toggling the RESET line correctly and the watchdog was working correctly. So on to what's up with the RAM.

After checking the RAM, the CAS and RAS signals were not present. THis was tracked down to chips 6K and 4K. Once those were replaced, it booted up to the RAM test, sort of. It was painting it in strips across the screen. So close.....

Based on the schematics, the culprit had to be one or more of the RAM addresing chips. I broke out the logic comparater. popped in a 74153 chip and on the second try, found that chip 3D was bad. On quick socket replacement and a perfect rug test! Well, near perfect, I got a 1-3-7 error, so I swapped chip 3Q with 3R, powered up to a 1-3-8 error. Replaced the RAM chip at 3R, powered up and I got a 1-1-7 error. :mad: Swapped chip 1Q with 1R, powered up and got a 1-1-8 error. Swapped chip 1R with a new chip and ...... [wait for it] .... the RAM test passed. Everything else passed and there it was a 0 on the ROM card.

picture.php


What a beautiful sight! So far it has been 2 hours and it is running the attract mode like a champ.

This is probably the most damaged board I have brought back, so far.

Hopefully if you are ever faced with a totally dead board, this will give you some things to look at to help you bring it back from the dead.

ken
 
Nice write up Ken! Thanks for sharing. I have a stack of boards I will be going through (doing Defenders right now), and I'm sure I'll run across a dead board.
 
I thought I would post this as a help to others who want to try ressurecting dead Williams boards.

I recently received a Joust boardset from a fellow KLOVer for repair. It had been sold to his as working, but as he found out, it was not. The first clue was that the CPU was inserted upside down in the socket. He obtained a replacement CPU chip and tried it with no results. So he sent it in.

When I first examined the board, I could tell that the batteries had exploded at some point in time. There was some corrosion on the chips next to the battery area. I wiped it down with alcohol and set it aside to dry.

Later I plugged it into my bench setup and it was dead. The CPU was not getting a clock pulse, the reset was pulled low regardless of the reset button being pressed.

So I stared at the beginning. The crystal was pulsing. But the signal was not getting past the first set of inverters (7J). I replaced the chip (after socketing it). Still no clock. I checked the first flip-flop (7K) and it was not toggling even though it was getting the clock pulses, so I replaced it.

I checke the CPU and I was not getting clock Q or E. Those traced back to another flip-flop (7F), so I replaced it. I still didn't have Q & E at the CPU. After a lot of head scratching I did something stupid, I replaced the CPU. This time when I powered it up, I got something on the screen! Wahooo! Progress. It was just a yellow screen, but it was the first visible sign of progress.

When I started probing again, I found that the watchdog was toggling the RESET pin and we had clocks Q & E, but no diagnostics on the ROM's LED. [fast forward through a lot of boring logic probing] I found that there was no activity on the data bus D0-D3. That sounded suspicious, so I replaced the CMOS chip (1C). When I powered it up, I got nothing. After some poking around I found that the 6MHz clock was not being generated so chip 4D (another flip-flop) was replaced.

This time, I got the 1-3-1 error on the ROM card. And the Reset switch was toggling the RESET line correctly and the watchdog was working correctly. So on to what's up with the RAM.

After checking the RAM, the CAS and RAS signals were not present. THis was tracked down to chips 6K and 4K. Once those were replaced, it booted up to the RAM test, sort of. It was painting it in strips across the screen. So close.....

Based on the schematics, the culprit had to be one or more of the RAM addresing chips. I broke out the logic comparater. popped in a 74153 chip and on the second try, found that chip 3D was bad. On quick socket replacement and a perfect rug test! Well, near perfect, I got a 1-3-7 error, so I swapped chip 3Q with 3R, powered up to a 1-3-8 error. Replaced the RAM chip at 3R, powered up and I got a 1-1-7 error. :mad: Swapped chip 1Q with 1R, powered up and got a 1-1-8 error. Swapped chip 1R with a new chip and ...... [wait for it] .... the RAM test passed. Everything else passed and there it was a 0 on the ROM card.

picture.php


What a beautiful sight! So far it has been 2 hours and it is running the attract mode like a champ.

This is probably the most damaged board I have brought back, so far.

Hopefully if you are ever faced with a totally dead board, this will give you some things to look at to help you bring it back from the dead.

ken

Ken,

What a lovely sight that "0" is !!!!

That was amazing work! I thought that my board was completely toasted for sure! I still can't believe that it is once again alive - and that there were so many problems with it.

The guy that I bought this Joust from said that it was working and just quit one day (um-humm) and that he noticed that the batterys were leaking. When he tried to replace the batteries, he broke the battery holder - he did tell me that. What he didn't tell me was that he was messing with it and installed the CPU in backwards in his efforts to get it back up and running!! I had bought a new CPU at Jameco and installed it correctly - but as you discovered, the NEW CPU was DOA as well!

Thanks again for all the hard work! I'll be looking forward to finishing my Joust restore now!

...also, thanks for sharing this Resurrection story with the KLOV community!

Kudos and +1 (Bows before the Repair GOD)

-Gino
 
Last edited:
Back
Top Bottom