Mr. and Mrs. Pacman light/display problem after mpu corrosion rebuild

TheShanMan

Well-known member

Donor 2011, 2024
Joined
Sep 9, 2008
Messages
5,657
Reaction score
725
Location
Hillsboro, Oregon
Mr. and Mrs. Pacman light/display problem after mpu corrosion rebuild

It's almost like there is a crossed signal or something because at times some of the lights will look correct (particularly the pac maze) and other times they won't - when some lights turn on or off it seems to affect other lights, maze, and score displays. None of this problem existed before so it would seemingly be an MPU board issue.

I've checked a number of chips, particularly near the battery area, to make sure each pin traces properly to wherever it's supposed to go, and that I don't have solder splashes from one pin to the next.

This is the only pin I've ever owned so I'm not that good yet at diagnosing this kind of stuff. Anyone have any ideas on what to check? I've got a logic probe and an o-scope at my disposal if that matters.

It would seem to me that focusing in on what's going on with the score displays would be easiest, but what do I know.

Here's a video that focuses on test mode, showing both the lamp test and the display test.



Here's a video that shows attract mode.

 
Most likely a RAM or ROM issue (IMO).

EDIT: Ideally I would swap in another MPU to be sure it's an MPU issue but I'm guessing that's not an option?
 
Last edited:
Meaning a bad ram or rom (as opposed to a bad connection as my mindset was due to the rework)?

Not necessarily. It could be a RAM or ROM issue related to your work. Like 2 solder pads connected together that shouldn't be or a pin not connected at all that should be.
 
Yes. I have looked for solder splashes and disconnected pins. I also tried a new 5101. I'll give the 5101 another another look.
 
Could also be an issue with the 6810 RAM. Did you test the PIAs and RAM with Leon's test ROM? If not that's probably what I would do next.
 
With a corrosion repair -- first thing I would look for is a broken trace. Can be anything from a hairline crack (not visible) to a bad plated thru hole. Only way to find these is by buzzing each address, data and control line from start to finish. Long and tedious, though.

Ed
 
Could also be an issue with the 6810 RAM. Did you test the PIAs and RAM with Leon's test ROM? If not that's probably what I would do next.

Not yet. I need to restrap the board because it's got 2532's in there now. Gotta figure out exactly how I do that, and I do plan to do so.

With a corrosion repair -- first thing I would look for is a broken trace. Can be anything from a hairline crack (not visible) to a bad plated thru hole. Only way to find these is by buzzing each address, data and control line from start to finish. Long and tedious, though.

That's exactly the kind of problem I've been trying to locate. I haven't systematically gone through every trace in the corrosion section to that degree yet. I have checked all the vias in that section though. I have also gone over several chips to make sure each leg makes contact to a distant component, but so far that hasn't revealed anything.

post detailed high def pictures of the board? maybe someone will see something?

Absolutely. Here they are! Note that the klov forums are sizing these down, so to view the larger images, click these links:

http://img.photobucket.com/albums/v...011-02-23MrandMrsPacmanPinballMPUBoard001.jpg

http://img.photobucket.com/albums/v...011-02-23MrandMrsPacmanPinballMPUBoard002.jpg

I appreciate the feedback guys!

2011-02-23MrandMrsPacmanPinballMPUBoard001.jpg


2011-02-23MrandMrsPacmanPinballMPUBoard002.jpg
 
Last edited:
you linked the same picture twice. instead of the front and back as you inserted any ways i see several things i dont like, not sure they are your problems but i would re-solder them whats 20 minutes of soldering?

i will edit your pictures and post them back maybe its just something getting lost in the picture
 
Fixed the link. Yeah, I think pics don't always perfectly illustrate what's there. I wouldn't say I love all of the work either, but at the same time it's tough to make it look pretty when you have traces completely worn through and whatnot. I look forward to your input though. Thank you!
 
i love stuff like that :) its always fun to have a challenge and when you fix it, it makes you feel good haha that or maybe all the lead fumes :)
 
Yeah, I don't really see anything specific with what you pointed out. It certainly doesn't look pretty and I think that's what's catching your attention, but like I said there's not too much I can do to make it look beautiful. Many cases of having to jumper something where the copper was worn through. The best I can do is make it functional. :) I thought maybe the feedback was going to be based on intimate familiarity with this board to say something like "this trace doesn't seem to be making contact from point A to point B like it is supposed to". I do appreciate the helpful attempt though.
 
sometimes that corrosion from the acid can cause the connection to not be good. there were several that appeared to be cold solder joints or broken to me, if you re-flow them with a solder iron and hit them with just a tad of new solder they should look 1000 times better as well possibly fix whats not right

http://forums.xbox-scene.com/index.php?showtopic=684774 will come in handy if you decide to fix the traces ive used that many many times
 
on pic 002 from the top left corner looks like maybe 3 inches to the right and 1 1/2 inchs down looks like whatever that is with the long trace thats going down isnt making any contact at all looks like maybe the R16 area?
 
Last edited:
Ugly Acid Damage

Ugh. My Cheetah MPU was worse but not sure by much! Also a good not is to check/replace IC sockets. I have had leakage under the sockets that cannot be seen cause all kinds of trouble. Likely these are the dreaded "scanbe" sockets as well. Acid trouble is not fun to repair. Got me an all in one and still havent managed to get the original fixed (2 more pins came by me hehe).
 
Back
Top Bottom