Williams WPC - random mid-game blackout

[good_buddy83]

Recently did an NVRAM install on a Williams Getaway Machine. It was playing playing great, still plays great, but recently one little nagging issue has popped up. The game will randomly completely lock-up mid game. No reset, DMD goes black, sound stops, lights stop, everything stops, but game is switched on and there is power. I was on a game, ready to kick it into 6th gear, and then instant nothingness.

Thinking next step is to play with the back box open, and check out the LED lights on the CPU when it happens. I suspect this is a CPU lock-up mid game, given that it did not reset. Guessing it could be related to a recent NVRAM install, maybe cracked solder pad where it is intermittent? Plan was to recheck continuity and reflow solder to the NVRAM socket and reseat the ROM and RAM. Anything else this could be? Any chance a Rom chip fails like this? It is the original Rom that came with the machine; revision L-2 if that helps. Curious to get your thoughts, this was a weird one for me.

 

[ToMMy]

Post a picture of the NVRAM you installed. There are varying qualities depending on the type of NVRAM you have.
The cheaper ones from offshore will act as you describe. Repairing boards I have seen that scenario a hundred times.

 

[good_buddy83]

This one was a Dave Astill NVRAM. I assume top of the line, never had an issue with any of his stuff. I had another person do the socket soldering (because it was so small) and they created problems on more than of these CPU boards. I had to do extensive work to repair, clean, meter and install jumpers. The Getaway is actually the only one that worked without needing repairs, but maybe solder resin could cause this? I'm not sure if cracked solder pads apply as much to fine solder sockets? Let me know. I'll grab a pictures tonight. I was more wondering if original EPROMs can fail overtime. Just weird that it plays so well and hard crashes. Solder pads don't make as much sense to me in this scenario.
I know there are traces that go in between the solder pins on the NVRAM socket, so maybe something heats up and becomes conductive? Just seems odd, but I will find it. Thought I would ask about EPROMS because I didn't want to chase a ghost if the original EPROM was a suspected culprit.

 

[ToMMy]

It won't be EPROM's. It will either be an inferior NVRAM (Dave sells both the good green ones and the cheap black ones with a white sticker depending on how much you paid), damaged traces from poor socket work, or something totally unrelated like the power connectors.