Are they all in the same row or column?

This is pretty common. And for whatever reason before LEDs you won't notice it (even though the problem exists). Most likely cause is a diode shorted or a socket miswired do that the diode is either not present or not doing its job. What you can do is work in single lamp test and make a chart. Like when 11 is on so is 21, 31, 81.
When 12 is on so 22, 32, 82.

If you can make that chart then it gives some hints at which sockets to look at. If you can paste those results here we can help you further if needed.

Yeah, stray voltage may not be enough to lite incandescents but probably enough for LEDs. I think you'l have to check that row of wires to find a short.

Ok. So start at all the lights on col 1: 11, 12, 13, etc and look at each socket. It should have a wire and a diode. The first ones I would suspect are any that are 44 twist bulbs. Often the diode gets bent and shorts to the casing of the socket causing this. The actual wpc light boards are less likely to have issues. But you can unplug them in mass to see of your problem goes away. It'll be caused by a diode either being bent, shorted, or worn out allowing current to pass back.

Interesting, what you're describing sounds to me like an open circuit. Meaning, the transistor that drives column #1 (Q98 on my WH2O - should be the same for all WPC titles) is shorted and open all the time. However, it that was the case you would see the same problem across every row not just rows 6 through 8. For example, in row #1 lamps (21, 31, 41, 51, 61, 71, and 81) would all trigger lamp 11 as well. This is the same pattern you're describing, but only on rows 6 through 8. Are you sure that this isn't happening for every other row as well? Again, if that is the case then you need to look at the TIP 102 transistor (Q98) or it's pre-transistor used to drive column #1 of the lamp matrix. Let me know what you find.

Go to the test single lamp mode of one of the lamps that make the extra lamps light.

One by one, go down the column and take out each bulb. When you get to the one that is the problem only the light that is supposed to be lit will be. After that, figure out what the problem is... bad diode, wired to the wrong lug, circuit board problem, etc.

I had this problem with my TAF and it drove me nuts for the longest time. In the end it turned out that on one lamp they had soldered to the wrong lug, completely bypassing the diode (easy to spot and fix), the other was a phantom trace on one of my lighting boards (much more difficult to figure out). I actually had to scratch down into the board to sever a trace that wasn't supposed to be there.

Good luck!

Chris

Ok, well that makes more since. Also, you're right about the TIP107. TIP102s are used for the rows while 107s are used for the columns (had to look in a WPC manual, sorry about that). Anyway, to test transistor Q98 you need to do the following with the machine powered down: (1) Use a multimeter in diode test and put the black lead on the center post of Q98. The transistor has three posts that connect it to the power driver board. Or you could put the black lead on the metal tab on top of the transistor, same thing as putting it on the center post. (2) Now put the red meter lead on the left post and you should read between .4 and .6 on your meter. (3) Now repeat step two, but put the red lead on the right post. Again, you should read between .4 and .6 on the meter.

Let me know what you find. If you read zero for either (2) or (3) then the transistor is bad and needs to be replaced. If the transistor is good we can take a look at individual diodes for lamps 16, 17, and 18.

Did you go through the steps I gave you already? My bet is in diode / wiring. I've seen this issue tons of times. I also believe problem with transistor would show up prior to LEDs.

Second that

People had me doing the same thing... Check all the transistors, use the meter, etc etc. Not saying that isn't the problem, but if you do the bulb check, one at a time, it eliminates the easy things first.

If you take each bulb in the column out, one at a time, and all of the lights stay lit the entire time THEN check the other things, but do the easy check first.

You are looking for things to be normal. For example... you have 11 lit and 21, 31, 41, 51 are all lit too. If you pull out 51 and the only one that stays lit is 11 you know that the problem is somewhere with 51.

You pull, 11 (replace), 21(replace), 31(replace), 41(replace)... nothing changes, but pull 51 and the only one that stays lit is 11, you found your problem.

Here are the threads from my problem:

First: http://pinside.com/pinball/forum/topic/help-me-understand-lighting-matrix

Second: http://pinside.com/pinball/forum/topic/fixed-lamp-problem-on-taf-almost-more-help-needed

Third: http://pinside.com/pinball/forum/topic/taf-lamps-work-lamps-work-lamps-work-lamps-work

I also fail to see how that will test anything.

Look at what the problem has in common. The problem is that the light in column 1 lights whenever there is another light on in another row. There are only 2 things that will cause this.

#1 would be a diode missing or miswired on column 1. However this is still not the likely problem. as this would cause the opposite problem. It should cause multiple lights to light in rows 6, 7 and 8.

#2 Bad TIP transistor at Q98.

Anytime any light is on in row 6 then lamp 61 will be lit, anytime a light is on in row 7 then 71 will be lit and anytime a light is on in row 8 then lamp 81 will be lit.

Now the op stated that this may have been occurring before the LED swap, but that is not neccesarily true. This transistor could have been damaged if the lamp socket was shorted to something while installing the LED's (saw someone do this recently to both a switch and a lamp transistor while installing LED's in a Dracula and it was near the coffin lock)
OR the LED's draw much less current and the transistor could already have been leaking current, enough current to cause an LED to light but not enough to cause an incandescent to light.

I recently saw the opposite of this with a system 11 PIA output. With an LED display in the game all the display segments lit, with a plasma display one of the segments was out. This was a weak output on the PIA, the output was strong enough to run the newer more effiecient buffer chips in the new display but not strong enough to run the buffers in the older display

If you read through those old posts, you will see that I was told exactly the same thing.

Not saying that this will absolutely solve his problem, but it is a whole lot easier to try this first BEFORE going in and just replacing board parts at random.

Worked for me TWICE on the same machine, so obviously this is not completely in left field.

If it doesn't work, meh.. Wasted all of about 15 minutes. Replacing parts on the board - days/weeks waiting for parts, then hoping it really is the problem. Worst case scenario... making the problem worse because the boards were not the problem in the first place.

I am far from an electronics expert, I just used logic to figure out what was wrong with my machine. As they say... "You're mileage may vary."

Agreed.

The reason I recommended starting with Q98 is because you mentioned that 16, 17, and 18 had a “bright flicker” in your original post. This is typically something you will see with a defective transistor opposed to a bad diode in a light string. I just didn’t understand why you were only seeing the problem in rows 6 through 8 until you mentioned rows 1 through 5 weren’t used.

FYI, I did the exact same thing on my TOTAN, shorted out a socket when installing LEDs. Word of advice; don’t tighten your flipper cranks with the machine powered up