Have a bally light driver that I'm trying to debug. The board is testing in a fully functional Flash Gordon with a good board set and harnesses. With my boards I can verify everything as good. When I put in the board to test, in lamp test mode, all lamps cycle as expected. But when in attract mode, the lamps are not functioning correctly. Some stay on, most flicker incorrectly... basically it's spastic and the attract light show isn't correct. Again, swap the board and everything is fine.

Tracing which lamps are misbehaving it's not unique to a decoder... the behavior spans all the decoders. I've redone the J4 connector as it was a bit of a mess. I've checked all the resistors on the strobe line and data/address lines in... and didn't see anything out of line on the 20k and 2.2M resistors. The boards are not hacked up... and everything looks fine when the test mode lights everything on/off in unison.

Any idea what could cause this flicker and bad light show? Is it just change all four decoders and hope for the best? It has to be something shared.... strobe, data or address I figure but can't find anything out of line so far.

Ideas?

I know the signal into j4 is good

I know it's getting signal to the decoders because the test mode can cycle all the lights

I'm trying to figure out why it can cycle the 'all' pattern but not the changing pattern well. Which leads me to think something with the address or data lines. Maybe the truth table will be telling...

No, I'm just swapping the lamp board. The game is fully functional, I'm just testing and trying to repair a second light board.

The MPU is an alltek and running fine with multiple other light boards in the same game.

I was suspecting the strobe but I can't nail it down.

The logic probe and voltmeter both say the strobe is pulsing... as is the I input (inhibit) to the decoder. I have a simple BK dp-21 logic probe, and its output I interpret as logic pulse with low duty cycle and the Interupt is a pulse with a high duty cycle. That all makes sense to me from the schematic. In attract mode the voltmeter isn't fast enough to show the change on strobe effectively, but in test mode you can clearly see it change between about .3 and 1.3V on the strobe input.

In display test mode, the lamps are all off as they should be.

I'm starting to think I need to hunt for maybe a shorted data line.

Yes, as I said, I replaced J4 entirely (mainly because the old pins were so mangled). Fairly confident of my work there. Plus, the symptom precedes that change

Again... the entire game operates fine except for this one board so no reason to suspect the mcu strobe output or aux board. And the attract mode symptoms map back to the decoders .

Why wouldn't strobe cycle? I need to check the mpu diagram but I don't expect strobe to just be held down

I don't know why it's different... but I also know a simple voltmeter is not great for this, so I'm not too stressed over it. I have multiple light boards working in this same machine, so I'm hesitant to blame stuff upstream of the light board in question.

This wasn't consistent with what I saw. Even in sound test mode, the Address lines were pulsing away, as were the Inhibit lines. The system seemed to be sending the truth table to output 15, which is N/C, but I could see the associated pin pulsing away while all the other output lines were low. I saw this on both the good and question boards.

The board is basically virgin and no signs of any shorts or even work besides the header pins being kind of nasty with old flux. I did swap C1 with another known good so there is signs of work there, and on J4 which I redid. I also redid Q60 I think it was which had been replaced with the classic 'solder the transistor without removing the board' trick.

I toned out all the data and address lines on the logic side, and there were no shorts. I also checked strobe and Inhibit again, and everything just connects as it should to the resistors and onto the J4 connector.

I can't think of anything else to do beyond possibly just removing J4 entirely, wiring in +5vdc and GND and trying to manually trigger strobe and a data line.. and plan 2: trying to remove decoders as if one might be messing up the others.

IMG_3816.JPG
IMG_3819.JPG
IMG_3820.JPG
IMG_3821.JPG

IMG_3815.JPG

IMG_3817.JPG

IMG_3818.JPG

Possible interop issue with the alltek and these chips? Unforunately all the bally games in my collection have alltek's in them currently... would have to jump around a bit to get a mpu I can trust. But maybe that would be worthy before depop'in all the decoders.