I'm having an issue with a Williams System 11 High Speed where the bottom row of back box lamps will only work if they're LEDs. These are the "Game Over", "credit", "ball in play", and "Shoot Again" lamps. If I use regular incandescent bulbs, I get no light, but if I switch them to LEDs, they work. I was thinking even the LEDs looked dim, but over some hours of play they started to brighten up. Does anyone have any ideas where I can start to troubleshoot this? I'm assuming the voltage at these lights is MPU controlled and it's not getting enough voltage to the socket, but it's not even dim. Also, nothing else in the back box or playfield lighting looks dim.

Thanks!

The incandescent lights don’t glow at all (that i can see) but the LEDs do flicker faintly when they’re supposed to be “off”. Doing a DC voltage test had proven difficult because the lamp test flashes the light on and off so the voltage rises and falls faster than my DMM will read on the other working bulbs. I was able to get just over 5V on the Game Over socket because that’s easy enough to get the game so light up

Should I just replace Q66 or is there another way to test it?

So you're saying I can test voltage from the transistor tab to ground while the light test is running and compare it to other transistors-ground voltage for known "working" controlled lamps?

Based on that and these clips from the manual, I'm inferring that Q66 should switch voltage and Q65 should switch ground, and that means row transistors is where I can test for voltage?

Capture (resized).JPG

Capture2 (resized).JPG

So I can see the same voltage between 3.6V and 12.5V ish strobing up and down as the lights blink on and off on both the column and row transistors. Does that mean the issue is on the grounding side (Q65?)? Should I just replace both Q66 and Q65?

The transistors are ordered and the resistor measures good. I guess I’ll just replace it too while the board is out. I get the same voltage reading out of J7-1 (3.5v - 4.5v, then sometimes spiking to 8-12v). I even tested about 12v-14v on either side of the resistor.