OK... my first pin and restore is up now and running, just need to dial in a few tweaks and minor bugs. Hopefully this one will fall into the "minor" category:

Problem: I have an issue with Lamp Matrix Row #8. This row runs (in order): multiball; left lock; right lock; Bonus holdover; two of the backlass thrusters; special; bonus grid 7; bonus grid 60.

History & Diagnosis:I converted the whole machine to LEDs and they all worked previously. Then I noticed the bonus grid acting weird. I swapped the bonus grid back to brand-new incandescent bulbs. The two on that row location ("7" and "60") burned out instantly.

I did not like the look of the grid with bulbs, so I but everything back to LEDs except those two, so I could easily identify any further bulb blowouts in those locations. At this point, I realized the whole matrix row was out. So I swapped the whole row to incads too (again, so I could easily see any bulb blowouts to trace the problem).

Well it seems I might have had a flaky LED, but the short of it is, with incadescent bulbs in the row, all indicators DO illuminate again. But they seem to be on all the time: while the rest of the bonus grid lights flash, the two in question (7, 60) just stay on. So do the others. One of the 3 thruster lights remains flickery too, but I haven't yet identified if it's in this row or not. Bonus grid still does not spell out letters (I am hoping it's related).

Questions:So... any ideas here? Is this due to having incads in this row, while the rest are LEDs? Since the row is illuminating I'm hoping my board components are good... or is a transistor stuck "ON" (and how would I check that)?

The TIP122 for that row previously tested out of tolerance when I had the MPU out for other things, so I proactively replaced it. It still tests good. However, the small transistor (as you later taught me to replace in kind) is a 5060, which I did not have on hand... it tests "OK" besides. I would like to get some but GPE is out of stock.

I feared something worse when they were all OUT and those two bulbs burned so quickly. Since they are back on (after wiggling / changing bulbs) I am wondering if there's something else to look at, or if it's truly down to a transistor issue.

Just trying to get some ideas before I get back home. Thank you!

OK, bulbs burned again so yes, there is definitely a stick-on somewhere. Looks like another GPE order on tap, oops. I have a couple 102's left but need the 5060/64's.

I'll probably order the chips while at it; Just found your testing advice here, will do that in a bit: https://pinside.com/pinball/forum/topic/lamp-row-stuck-on

Man, I never would have guessed pinball machines would "teach" me more about SS electronics repair than a 20+ year career of computer hardware and such... oops!

OK, now I am stumped.

I tested the 7406 chips in-board, and all the relevant pin pairs seem to test OK, and similar to each other (~.52x thru .53x)

I read that 2N5060 SCR's can't be reliably tested unless under load... but I was afraid to do this. For what it's worth, in-board with no load, none of them tested close to each other. So I replaced them ALL (as we surmised might be needed above)... and the replacements all test the same. So at least there is now consistency and I can rule one variable out.

BUT the problem still remains! Row 8 is locked on, and other rows (seems like 3,4,5) are more "visible" under game play. Attract mode does not look normal; the spell-out in the bonus light grid is gibberish.

All the TIP122's test the same and good (I replaced the one for Row 8 already, with a TIP102).

I checked all the associated diodes and resistors for each row, all are good and in spec. The SIP resistor tests in spec on all pins. The capacitors for each row test the same.

What should I do next? Replace the 7406's anyway? I have the parts so I might do that while I wait for a response, but I am afraid of that not solving the problem. I REALLY thought the new SCR 5064's would change the symptoms if not solve the whole problem.

OK, I replaced the both 7406 chips. Since I had troublesome rows handled by each, this made sense.

After doing this, predrivers (5064s) for Rows 7 and 5 fried. BAD. During the lamp test, when they would flash I could see an arc/spark on the board terminals for both S5 and S7.

This did not happen before I replaced the chips: I had the machine on for a few minutes, checking bulbs and such.

When I had the board out to replace the chips, everything was fine.

After I replaced the chips it was immediately apparent something was wrong but it took a few moments to trace and observe. Machine was probably on only less than a minute but I now have scorch damage and who knows what else to deal with...

After all my hard work and being so careful, this is heartbreaking. I don't know what to do next. That lingering ozone smell and such just rubs salt in the wound.

The fuses were the very first thing I inspected and replaced if necessary when I got the machine. They are all the proper rating (to their credit, the PO's had mostly the right ones except a 3A in the solenoid block - which was the only area I had trouble with due to the faulty left pop bumper circuit - I have since fixed it and am using the proper 2.5A there).

OK, pics you requested: 2400px wide in full-res...!

Sys9RowBurnFront.jpg

^ Note the scorching on the Anode pin 3, on both S5 and S7. During the diagnostic "all flash" controlled lamp test, when the lamps would flash I would see a similar flash on those pads. S7 was MUCH worse. One of those "am I really seeing this?" things I kinda stared at for several seconds. I do not know if anything was amiss before I started the test.

Note I replaced the previous direct-mount 7406 chips, with sockets and new chips. Light glare seems to indicate I don't have them both level at identical Nth degrees, but they are solid.

At this magnification it would "almost" seem like the top anode pad on S5 is touching the trace next to it... but when I check this, there is no continuity.

You may notice the discoloration near U54: I accidentally began desoldering that row, but caught my mistake after a few pins and resoldered them. Post screw-up, it tests identically to U53, and FWIW I did not notice a "column" issue (granted, I probably didn't have time to make sure).

I did run the test long enough to see that Row 8 was STILL locked on.

Sys9RowBurnBack.jpg

I labeled the areas with pencil (afterward) since this back view is inverted.

I'm not a pro but I think my solder work is OK? I have checked all traces for continuity / suspected possible shorts, and haven't noticed anything.

Postmortem Diagnosis

Here is what I know post incident:

- Q38 tests bad now - whether cause or effect I have no idea... at any rate it is not logically related to S7/S5 though it is physically very close.

- SIP S19, 6.7 Kohm, pin 8 (from the left) now tests low, ~5.6 Kohms. (this is also for that blasted Row 8... BUT I know this was good before since I tested everything I could, including this specifically).

- All n4001 diodes indicate OK (on board). All discrete resistors are OK.

- Very strange: 7406 #U59, pin 4 only tests at ~.2xx when in the socket. Outside the socket, it is fine. Both chips test the same off board; same behavior with U59 socket4 if I swap them. I can't discern why this would happen or what could be wrong with the socket. I do realize this is schematically related to S7... but is it cause or effect or WTF...?

- The 3W power resistors are all testing .8 ohms... not the rated .4 ohms. I am second-guessing myself on these components now. I know I tested them all originally, and they were all the same. So IF they were this out of spec previously I didn't catch it (since most of the rows were OK and I was looking for obvious outliers given too many numbers in my head and on paper and.... sigh). How serious is this aspect? I would surmise the higher resistance would be a "better" failure than going open (might just dim some rows, right?), and since they are all the same is it really a "failure" or just age/wear... again, most rows are OK...? Replacements seem pretty hard to find....

My tests are all with a DMM in the appropriate mode. I don't have a logic probe (never had a conceivable need for one until... now....?)

To help anyone for reference, here is how the components break out from the Manual Schematic. Conveniently, the Q and S component numbers correspond to the rows! The 7406 pinouts are a little more "random":

Row....Q#.....S#....U#.....Upins
------------------------------------
1........31......1......58......3-4
2........32......2......58......1-2

3........33......3......59......12-13
4........34......4......59......1-2

5........35......5......58......10-11
6........36......6......58......12-13

7........37......7......59......3-4
8........38......8......59......10-11

Thank you to anyone who can help. I'll be licking my stupidity scars from this one for a while...

Morning bump, since I clarified some parts of my last post.

Seeking advice from anyone who has the time to help a newb and likes an electrical challenge... maybe give poor Vid a break, ha ha...

I understand... it is a lot of "symptoms". I provided all in hope someone with random/arcane knowledge or experience might recognize something. Thank you for looking.

I realized that no matter what, the scorched parts had to come off either way. I did so and:

1) I discovered that Q38's center pin (not easily seen) was also scorched badly. Almost as if it shorted. That seems to be an important clue...

2) After removing S5, S7 (and Q38).... the funky readings I noted on U29 and SR19, have gone away. That's a hopeful sign...? (As in, the fried parts in-circuit, throwing off measurements... makes sense)

So my (albeit optimistic) working theory is that the TIP102 Q38 was defective / locked on all along. But there must have been a problem with the original U58 7406 that never really tried to operate it "full on" in the first place? When I replaced the 7406, the signal was sent through, revealing the issue with Q38...? I presume S7 fried because of close proximity.

(As to that issue... defective part or a bad install, who knows... I've replaced several other TIPs without issue but I will always second guess myself. For what it's worth, Q38 had been there for quite some time. It never went boom until I replaced the 7406 IC (after replacing its 5064 pre-driver)).

IMG_7213.JPG

I hope that seems plausible.

I'm going to try and fix it one more time. If it doesn't work, no less than Mr. Cheung has agreed to help me out, so at least I have a Plan B.