Before I go hog wild and replace everything in the circuit, I'd love to hear some opinions on where my fault might most likely be. Here's the situation...

All the controlled lights on my Black Knight work except for the numbers 2-9 in the sword. The schematic identifies these lamps as Column 7. Since all other lights work, I am assuming there is no Row issue. All the lamps in the sword share ground connections, so since the 1 and the 10 light up, I can't see ground being an issue. I pulled the Column connector on the Driver board back enough to get a multimeter probe clipped on and I can confirm connectivity all the way to the 9 lamp so I've ruled out wiring too I believe. That leaves the Driver board.

So 2 questions...

1. Although I have all the components to replace the column circuit, I'd like to know if there is any one component more likely to fail than the others, specifically the TIP142, the 7408 or the pre-drive transistor 2N6427? I can check the resistor easily enough. I will eventually do the full upgrade to MOSFETS along with removal of the resistors but I'd like to get the whole board working first before I waste time on it.

2. As a further test to prove to myself its the Driver board, would I harm anything by pulling the column connector and jumping, say, the Column 6 output pin to the Column 7 pin on the playfield side? My assumption here is that this would then cycle the 2-9 lamps, proving the defect to be on the board. Anyone see any issues with that?

Thanks for the input! Back downstairs to swap out the wimpy left flipper!

Dave

Thanks guys!

I need to dig up my DMM manual to see what the diode test was telling me. I was getting around 1100-1200mv between each outer leg and the center tab.

I did try shorting the Column 7 TIP-42 tab to the Column 6 one as Chris suggested and all the 2-9 lamps came on. If I am understanding correctly, this proves out the wiring path all the way down to the bulbs, through the header connector, and also proves out the grounds.

Looking at the schematic, it looks like maybe I could attach a logic probe to the output of the 7408 for Column 7 and watch for it to pulse when in attract mode. Does that make sense? If so, it could rule out having to pull the IC.

Dave

I grabbed the logic probe, which admittedly I have never used though I bought it some time ago for arcade video game use. I hooked the leads up to +5 and Gnd, then poked the probe at pin 8 of IC14 on the Driver board. It showed a steady Low light. Now I thought that might indicate an issue with the 7408 so I also probed the pins associated with the other columns on IC14 - nope, they also showed a steady Low. So I think I have to assume the 7408 is OK. I went a bit further and probed the Driver board's PIA pins associated with IC14. They all also indicated a steady Low state. If I flipped on the Mem button on the probe, the pulse light would come on and stay on when I removed the probe tip from the pin. I've long since lost the manual for the probe so I am not sure what that means other than I guess it saw some kind of signal there.

Anyway, since all the pins on the associated semiconductors look the same, I think I'm ready to assume that it is one or both of the transistors. Any thoughts from anyone before I do that?

Dave

Well, that was interesting and a little confusing. All of the TIP42s seem to be pulsing when I touch their tabs. Now that said, 6 of the 7 seem to favor having the Low light brighter than the High light on the probe, where the Column 7 one seems to favor the High light being the brighter. I'm not sure what that means, if anything.

I guess the thing to do is just pull it and swap out the TIP for a Mosfet and swap the 27 ohm resistor for a jumper, then see what happens.

Thanks Chris. I was looking at the various bulletproofing guides out and about the net and most, if not all, recommend swapping the TIP42 for a MOSFET and removing the 27 ohm resistor and installing a jumper or zero-ohm there. Reportedly this makes the board run cooler, eliminates the crisping that goes on around the resistors and lessens the load on the power supply. I do have some TIP42s, so maybe I will just try the single swap first to see if that's it. I could do the other stuff when I pull the power supply and sound board for re-capping.

Off to the basement to take some pics, label some stuff and pull the driver board! Wish me luck...

Dave

Some progress to report, but more help needed

Replacing the TIP42 has indeed brought lamp column 7 back to life. Now, however, some of that string, specifically 2, 3, 6 and 9, remain on when they should not. Most of them do flicker but at varying rates. The 5, for example, flickers so fast it almost seems to be on steadily. The others in that column seem to work OK, cycling up from 1 in attract mode, increasing with scoring, etc.

Since, as far as I can tell anyway, all the other lights work as they should, at the right times, etc., I am thinking it can't be the row driver circuit. If a row was out, that would manifest itself outside just my 2-9 issue. So I have 2 thoughts - 40 pin connector or PIA. Anyone have any thoughts/comments on those 2 possibilities, or something else even?

Thanks a lot! Almost there...

Dave

Maybe this info will be helpful. It suddenly occurred to me that the manual had mentioned a lamp test. I went down and tried that and it works fine. All controlled lamps blink on and off as they should. The lamps in the 2-9 column that are staying on most of the time did turn off in the lamp test when they should.

Looking at it more closely, I think the problem may start at the 5. The attract mode cycling is fast enough that it's tough to tell but I think the 2-4 may be working right. I'm wondering about the diode on the number 5 lamp now. Any thoughts on that?

Lamp test does work fine, but when in attract or game play move, the 5 lamp is on nearly steadily, and several of the others in that column flicker on and off when they shouldn't. The lamp test on BK is minimalistic - it just flashes all the lights on and off at once. You cannot check each controlled lamp one at a time.

It's definitely broken in some way during attract and game play, though. No doubt about it.

Dave

No LEDs, yet

My plan right now is to swap out the diode on number 5 as a test and see if that helps. If so, I can then do it to the other problematic bulbs.

I uploaded a little 2 minute video to Youtube demonstrating the behavior. Any input on the next most likely cause would be welcomed. I'm leaning toward changing out the diodes, but I'm not sure that's it and I wouldn't mind saving the time if anyone thinks that would be chasing my tail.

And a further bit of troubleshooting info - replacing the diode on Lamp 5 made no difference whatsoever.

I was afraid you were gonna say that

I guess I'll start by swapping out the 7408. I'm thinking the pre-driver transistor is possibly OK since the string works in lamp test mode. I'll try and avoid the PIA if I can. Of course, of the three PIAs on-board, the one I need is the only one not in a socket. Grrrr

Thanks guys.

Chris, would you have a link to that 7408 diode test process? I couldn't seem to find it on Pinwiki.

Dave

I do have a logic probe John. I used it a bit earlier in the process and as best I could tell, in my limited experience with it, the PIA and 7408 looked OK. The pins on the 7408, example, seemed to behave the same for both the bad lamp column and good lamp columns. I will take another look at the 7408 with it tonight though before I decide whether to swap it or not. Sadly, similar to the PIA, one of the 7408s has been replaced in the past and socketed, but not the one for column 7

And don't worry - I'm not touching the PIAs unless it's a last resort!

Armed with 7408 data sheet in hand, I went down and probed the chip. Everything seems to be fine. Each of the pins tied to the Blanking signal is held High steadily. That's one input to the gates. Each of the other input pins is being pulsed. The output pins from each gate, those lamps that are working and those lamps that are glitchy, show a steady Low state. I can't say I understand why they don't pulse on the output either but I checked both 7408s and the probe output is identical for both. This leads me to think that the 7408s are fine, and they seem to be getting driven by the PIA so I am reluctant to swap that out in the absence of anything concrete. The only other thing in the column circuit is the pre-driver 2N6427, the 27 ohm resistor and a couple other small resistors.

Interestingly, when the game first comes on, the 2-9 lamps actually work for a couple cycles, then they start misbehaving. Thermal issue with the pre-driver maybe? The fact that they do work correctly even for a few moments, to me, again suggests that the PIA nor the 7408 are at fault.

OK, pulling all the bulbs and putting just one in at each of the positions 2-9 in turn yielded the same results for each lamp position. The lamp was on nearly steadily, flickering but very rapidly. The lamps on either side behaved normally, the 1 and the 10 through 40.

Keep those ideas coming, because I'm about out of them

I didn't have time for any electrical work tonight but I went down and played a bit. I watched it closely when I switched it on, and the Column 7 string of 2-9 worked perfectly for about 1 minute, then the glitches started to creep in. I'm thinking this has to be thermal somewhere. I've got a half-day off tomorrow so I think I will swap the pre-driver and 7408 just for kicks.

I also went ahead and ordered a replacement modern driver board from Kohout Enterprises (pinballpcb) so once that arrives I can swap that in and see what happens. I'd like to get the original Williams board fixed and working, then static bag it and put it away for safe keeping. I'll use the Kohout board as it's modernized and has a slew of diagnostic LEDs and such.

One step forward...

Swapping out the pre-driver and the 7408 solved the issue for lamp column 7. The numbers 2-9 now work fine. Sadly, now the 10-40 and the 4 Multiplier lamps aren't working, in the same irregular manner. They do work fine in the lamp test. So, I swapped out the pre-driver and driver transistor for Column 8. No luck. So now I am confused - these were working fine before so it's something I did, but I'll be darned if I can see or find it.

Grrrrr.

That was my first thought but I continuity checked every pin on the 7408 to its destination. All were fine. I also did the same with the replaced transistors - also fine. I swapped one new 7408 for another new 7408 - still the same. I did have a couple late-night thoughts that I will look at tomorrow - maybe I screwed up a thru-hole plating and need to solder the top side too. It's also possible, I suppose, that I have a bad transistor even though they are new. I may swap that out again. And I should probably continuity check all of the connector pins back to their sources just to be on the safe side.

The Black Knight shall not defeat me! Well, not in troubleshooting anyway He still kicks my butt in pinball.

All lamp matrix issues have been solved. For those following the thread and for the sake of searches later, the issue was indeed a broken through-hole on IC14, one of the 7408s. Pin 11 was supposed to be getting to Ground through a resistor. When I began continuity checking all the top-side connections, I quickly found it was not. I was able to flow a bit of solder in from the top using a very fine tip on the Hakko 888D. Once I had continuity, I checked everything else and they all looked good. I popped the board back in, hooked her up and powered on to the glorious sight of all bonus lamps sweeping up the length of the sword.

I think we can put this one to bed! Thanks to all for the invaluable advice.

Dave