My Space Invaders sat awhile un-played and the other day I fired it up and it had some very odd problems after starting a game. I verified that the problem is with the MPU by switching ROMs and putting the board in my Paragon, it does the same thing, this:

1. Boots with 7 flashes, all looks normal but displays have whacky numbers (might be due to bad settings in 5101 from game switch)
2. Start a game, ball is kicked into shooter lane but machine is flashing lamps.
3. Game plays, scores points, sounds work but lamps still flashing.
3. Pops and slings work for a bit, then go dead.

I've swapped out CPU, both PIAs, RAM and 5101, no change in behavior.
Voltage at TP1 = 5.2v, TP2 = 14v, TP3 = 22v, all nominal

Someone please save me a month of debugging and give me clues on where to look, Thanks in advance!

I think it's safe to say it's the MPU, I've already moved it to a different machine. Looks like maybe the lamp address/data lines might be stuck on - the output of U10 PIA, but I don't see anything on the MPU that is common to those lines, so looking now at the inputs from the CPU to the U10 PIA.

Bump, help a brother out....

Stared at the schematics for an hour, no buffer chips on the address lines, tending to think it might be U14 or U18 but I really do not see what on the MPU would cause all lamps to strobe.

Bump - one more cry for help, any input appreciated.

Thanks. I'll run further tests before taking a hammer to the MPU.

This is probably not what I thought initially, I ran downstairs and fired it up again, the game is starting, but the playfield lamps are being strobed as if the machine is still in attract mode. The game scores, bumpers work (for a while), and it plays normally except for the lamps, then at some point dies.

Would it make sense that an address line to the ROMs might be tweaked? Like perhaps it's fetching and running the attract mode instead of game code? Not sure if that is the case if scoring would work.

Thanks, I looked at the signal, it ran from 120 to 127Hz, looked good, I replaced U14, R16 and R113 just for kicks but it's still off the rails. Not sure where to go from here. I didn't replace R17, might just go do that for lack of other ideas.

Thanks for the input.

Based on Internet searches, heading towards U15 and U16 to see if the clock circuit is hosed. Any advice on how to debug that circuit? Would it cause slightly random behavior?