Strange problem with a Pac Man board and its solution
I've not seen this documented yet so I thought I'd post this info. Some of the problems with this board are common but the ultimate problem, I've never seen before.
Another KLOV member sends me a PacMan PCB for repair. At the very least it needs software as all the ROMs are missing.
OK, burn the ROMs, install them on the board, nada. Ugh.
Check the CPU clock at pin6, its there. Check the reset at pin26, its stuck (should be low at power up and go high shortly after but it was staying hi). Reset button on the board does nothing. Probe the 74LS161 @ 9C (it drives the reset) and its output is stuck (pin15). OK, replace it. Reset now works but the game is still dead, no activity on address or data bus lines. Replace the CPU IC and the game fires up! YAY!
NOT!!! About 30 minutes later the board starts to lose its mind. Now what? Lets check the RAM. They all test good with my new tester but replacing them fixes this current problem. Let the board sit some more but its still going ape shit. This time ONLY when PacMan comes on the screen. Thats right, attract screen works great until our little yellow buddy shows up. Same thing happens if I start a game right after power up. AS SOON as PacMan shows up the game resets.
UGH....
I go back to the RAM. I've never seen ALL 6 be bad before. I figured it had to be something tied to the RAM... The 245 @4H, nope. Board worked better but was still not right. Next in line was the 139 @5L. THAT DID IT!
So original the 139 at 5L was contributing to all the original RAM being reported as bad by the board and only after it had been on for a while. It was also contributing to the reset when PacMan showed up on the screen.
Very odd one indeed.
BTW, freeze spray did nothing to isolate this problem. However, when the original 139 was cold it tested fine in my TTL tester. If I heated the IC up it would fail. Seems strange that freeze spray didnt help me in this situation.
One more thing. This board was VERY sensitive to +5 when the bad IC was on the board. Even cold the +5 had to be well above, like 5.3, for the image to look right. After replacing the IC the board works fine even all the way down to 4.4.
Oh, I also found a shorted (253 ohms worth) glass cap on this board. I've NEVER seen one of those go bad.
I've not seen this documented yet so I thought I'd post this info. Some of the problems with this board are common but the ultimate problem, I've never seen before.
Another KLOV member sends me a PacMan PCB for repair. At the very least it needs software as all the ROMs are missing.
OK, burn the ROMs, install them on the board, nada. Ugh.
Check the CPU clock at pin6, its there. Check the reset at pin26, its stuck (should be low at power up and go high shortly after but it was staying hi). Reset button on the board does nothing. Probe the 74LS161 @ 9C (it drives the reset) and its output is stuck (pin15). OK, replace it. Reset now works but the game is still dead, no activity on address or data bus lines. Replace the CPU IC and the game fires up! YAY!
NOT!!! About 30 minutes later the board starts to lose its mind. Now what? Lets check the RAM. They all test good with my new tester but replacing them fixes this current problem. Let the board sit some more but its still going ape shit. This time ONLY when PacMan comes on the screen. Thats right, attract screen works great until our little yellow buddy shows up. Same thing happens if I start a game right after power up. AS SOON as PacMan shows up the game resets.
UGH....
I go back to the RAM. I've never seen ALL 6 be bad before. I figured it had to be something tied to the RAM... The 245 @4H, nope. Board worked better but was still not right. Next in line was the 139 @5L. THAT DID IT!
So original the 139 at 5L was contributing to all the original RAM being reported as bad by the board and only after it had been on for a while. It was also contributing to the reset when PacMan showed up on the screen.
Very odd one indeed.
BTW, freeze spray did nothing to isolate this problem. However, when the original 139 was cold it tested fine in my TTL tester. If I heated the IC up it would fail. Seems strange that freeze spray didnt help me in this situation.
One more thing. This board was VERY sensitive to +5 when the bad IC was on the board. Even cold the +5 had to be well above, like 5.3, for the image to look right. After replacing the IC the board works fine even all the way down to 4.4.
Oh, I also found a shorted (253 ohms worth) glass cap on this board. I've NEVER seen one of those go bad.