I have a Williams Disco Fever. New pin to me so was changing out burned out bulbs and ran into some issues.

So first, a few of the lamp sockets on the light strips were dead. Replaced the lamps and nothing. So looked inside and the grommet that rivets the spring down was broken. Was able to bypass the broken grommet by soldering a wire to the spring by passing it through the hole, so that's all fixed.

Now my problem is with one light. I show power to the socket, so the wires are OK, but as soon as the power passes through the diode the power disappears. I replaced the diode and it still has the problem. So I thought maybe it was a socket issue so I removed the power lead from the socket and installed a new diode and tested it directly without the socket. Again, power to the striped side of the diode, but none through the diode.

What could be the problem? No other issues with lights on that column or row that intersects this lamp.

Mystified?

Update. Found one more lamp circuit with the same problem. They are both on the same row, but everything else on that row seems fine. Also the lamp does light when in test lamp mode, but doesn't work when in game mode. I'm super stumped.

Please help. As I continued to troubleshoot more lights died. Now all the lights have died on rows 2 and 4. Everything else works fine. Checked and have output at the connector 2J7 for row 4, but row 2 is dead at the connector. But all lights in these rows are dead, but show voltage before the diode, but not after. What am I missing?

Well I'm guessing its a transistor issue. I pulled the driver board and tested the transistors. Found what I believe to be row 2 has a transistor that isn't shorted, but tests way out of range. Row 4 doesn't even have a transistor in place. Now funny thing is I found a transistor for it laying in the bottom of the cabinet. It has all it legs, so I'll bet it broke off, and someone had this transistor ready to install and never did. So row 4 has been out all along, but since its lights are at the four corners of the playfield I never noticed it.

I ordered another TIP41 for row 2 (actually IRFZ34N) so will replace both Row 2 and 4 with those.

Fingers crossed.

pasted_image (resized).png

So I don't think row 4 TIP41 has been missing all along. the transistor that I found in the bottom of the cabinet has been installed before, But it tests good. Row 2 TIP41 is bad for sure, shows .56 on one leg and 1.23 on the other.

So the question is, how and why did the transistor come out of the board. Could it really have overheated enough to come unsoldered? And if so why would it still test ok? I know I accidently soldered the center pin of three lamps in that row to the wrong side of the diode, but could that really have caused these issues?

Inquiring minds want to know.

Fixed the issues and pin is working properly. I'll post the resolution in case someone else runs across this and needs the resolution. I hate when I search and fine my exact problem and there's either no responses to it or no feedback on the resolution, so I vowed to never be "that guy"

So I surmised that the power I was getting before the diode was the blanking signal, but the ON signal wasn't on the other side of the diode which is why it appear4ed dead since we are dealing with AC current which can flow back and forth.

Since my rows died, albeit not all at once, is the reason I had trouble tracing the original problem.

So I tested the transistors and found that row 2 was passing .5v on one leg and 1.3v on the other, should be between .4-.6 on both legs.

And the row 4 transistor had "unsoldered" itself. Holy cow, don't know what happened there.

Checked all other transistors and pre-drivers and everything was within specs.

Had some TIP42 transistors but that didn't help as the Rows use TIP41. Wow are these expensive and hard to find.

Ended up sourcing IRFZ34N MOFSETS NPN or N-channel to replace the TIP41. For reference you can replace the TIP42 with IRF9Z34N which is the P-channel or PNP.

+++ as a side note. You don't seem to be able to test the IRFZ34N the same as a standard TIP transistor as it tests as dead or bad, but in fact works fine. Need more research on that I guess.

After double checking my work powered it up, and BINGO!!! Works like a charm.

Hope this may help someone else.