UPDATE 7/18: please look at this post to see the current status of this machine.

https://pinside.com/pinball/forum/topic/firepower-wont-boot-after-bulletproofing#post-2573434

I'll leave this other stuff up for posterity.

----------------------------------------------------
I have a Firepower that I've bulletproofed, according to Vid's guide, and now it won't boot. I'm sure I did something wrong but I figured I'd document my troubleshooting process here in case anyone has any advice.

The Issue

When I turn the machine on, the GI comes on, but nothing else happens. Sometimes, the sound/speech board will say FIREPOWER and go "pew pew pew pew" and sometimes I can hear a relay engaging and/or disengaging, but that's it. The two red LEDs sometimes stay on, sometimes one of them goes out but the other stays one, and sometimes they both blink on and then off and stay off. Neither of the two push-button CPU testers do anything.

As I mentioned, I did a bunch of component replacements, so I've been meticulously going over my work to make sure that I didn't screw anything up. First up, is the CPU and Driver boards.

CPU/Driver Boards

I replaced all six of the IC sockets in the lower left as well as the sockets for IC13 and IC1. They were all Scanbe sockets. I've gone through the CPU logic diagram on http://firepowerpinball.com and tested each of the connections at the pins for each chip (as opposed to starting at the underside trace). All of the pins on these sockets chips go where they're supposed to, and I don't see any evidence of shorting from pad to pad that a sloppy soldering job might leave behind.

One question I have, though, is for pins 9 and 10 on IC13 and IC16, the two RAM chips. I don't see anything in the diagram about this, but pins 9 and 10 on these two chips all have connectivity between them. Pin 9 on both of these is supposed to be D7 on the data busk, and pin 10 is supposed to go to Bus 02 and TP06. However, Pin 9 also goes to BUS 02/TP06, and Pin 10 is a part of the D7 data bus channel(?). If there's anyone reading this who has a chance, can you test out pins 9 and 10 on IC13 and IC16 to see if yours is doing the same thing? It's weird.

On the driver board, I replaced the eight lamp driver transistors with MOSFETs and replaced the big resistors attached to them with jumper wires. I haven't had a chance to fully test out the driver board yet, though.

Power Supply

I've replaced the capacitors and the high holtage section's resistors and diodes. I haven't gotten around to testing it out. Maybe something I did here messed up something in the 5 (or 12?) volt logic circuit. I'll have to do some digging here.

Sound Board

I've replaced all of the sockets for the speech board as well as a couple of sockets on the sound board. I've also replaced the OP-AMP chips on the speech board, and put them into sockets. I've also replaced all of the capacitors. All of this was because the sound on this game was pretty flaky, so I figured I'd take the shotgun approach and replace all the old stuff. The sound board isn't part of the equation right now, since I understand that System 6 sound is independent to the core function of the game. As such, I've removed all connections to it. It may as well not even exist.

What now?
So there we have it. I have a Firepower that I've done a ton of stuff to in a single shot and now I have about 30 million different things that could be wrong. I really should have done this a board at a time (and even then, a section at a time) but the guy's machine was an hour drive away, and I didn't have room in my garage at the time to bring it back to my place. That'll teach me, I guess.

If anyone has any advice for me, then please feel free to chime in. otherwise, I'll use this thread to document my investigation so that other people can learn from my mistakes.

Well, I wouldn't go that far. I was able to make some room in my garage and set it up. I don't think anything needs to be sent off. I'm fairly certain this is a problem I can tackle if I just work through things one at a time.

This machine belongs to a guy who lives about an hour away from me, so I can't tell exactly when it stopped working. The game wasn't booting right (with different symptoms) before I started this, which prompted the workover. However, it was working in the time before that. The chips were way loose in their sockets, and one of the capacitors was bulging. It was definitely time.

Agreed. I'll start working on that tonight.

I don't have a ROM Burner handy, but I think I know some people who can get me a set of test ROMs. Or I can order a set online, I'm sure. I'll give that some thought.

Could I use something from from Hans's Siegecraft site?

http://www.siegecraft.us/presta/index.php

I've been looking at picking those up anyway. They seem really handy for on-the-bench testing. How many of the solenoid output testers should I pick up? Three of them?

I did, and everything tests fine there. The header pins have a strong connection to the input side. I don't think there are any issues there.

This lines up with my suspicion. I removed the IC13 socket I put in, and don't see any sign of pins 9 and 10 being shorted together. IC16 is soldered in, and I haven't touched it, so unless something fused inside that chip, I don't know of any other way to confirm my suspicion other than pulling that one out and testing the pads with both chips out. At that point, I'd need to order a new RAM chip and socket, so I want to avoid messing something up that might not already be broken if the problem might be somewhere else down the line.

In case I can't trace this problem to a bridged pair of solder pads somewhere, my finger is hovering over the "Checkout" button at the Siegecraft site for some stuff that I need to test the board with something like Andre's ROM. I figure the minimum is three Solenoid output testers. Is there anything else that would making testing this board way easier? I mean, I'd love to have all of his stuff, but I kind of need to draw the line somewhere.

Excellent news, everyone!

There was, in fact, a bridge between pins 9 and 10 on the female side of the 40-pin interconnect. Cleaning this up allowed the game to boot into attract mode, but the solenoid fuse blew. I can only assume that this is because something locked on when it shouldn't have. I replaced the solenoid fuse, re-seated all of my connections and turned it back on.

Now, the game appears to boot and the GI comes on, but the playfield lamps aren't flashing as if the game were in attract mode. Player 1 is alternating between the high score and all zeroes (as it should be) but on every score display, only half the digits are showing, and some of them are missing their upper-right segment.

I can't remember what the state of the score displays were when the machine was considered "working". So, as mentioned above, I'm going to just tackle this problem as if I've seen it for the first time. I'll start investigating what could cause every other score display digit to not show up, as well as having a consistently missing segment on several of the digits that do show up.

ScoreDisplays.jpg

I've been doing some reading, and Clay says that an every-other-digit problem is caused by the Lamp and/or Solenoid PIAs (IC10 and /or IC5, respectively). Is there a way for me to test this without having to buy new PIAs and hope they work? Can I do some kind of logic probe or voltage test thing somewhere?

I'm not sure what you're getting at.

I went out and bought a new PIA to replace IC10, which is the PIA for the lamp matrix. No change.

So, ignoring where I was before, here's where I'm at now:

The machine boots. From the score displays, it looks like it's in attract mode. However, every other digit is out on all the displays. The upper-right segments on Player 1, Player 2 and the master display is missing, and the segments that do appear on the master display flicker like crazy.

Despite the score displays looking like the game is in attract mode, the game won't run the lamp matrix in either attract mode or test mode. The game also doesn't recognize any switch inputs. This means I can't get it to register credits, nor can I get it into Free Play mode to see if a game will even start. It probably wouldn't start anyway, since it would need to be able to see switch inputs to detect that the balls are all where they should be.

So, this machine is pretty much screwed until I can figure all of this stuff out. I think this is the last time I offer to help someone. I'm turning out to be not very good at this.

At the time he made that comment, I hadn't socketed any of them. As of right now, I only have the lamp matrix PIA socketed, with no resultant change. I'm not able to pull these PIAs out without ruining them, and I'd rather not blindly throw another 45 dollars in parts at this problem without some kind of idea of whether or not it will fix anything.

Well, I have the machine in my garage so it's effectively mine until it's fixed.

I suppose it's time for me to drop some money on a test ROM and some of that Siegecraft test gear.

Is there anyone who can at least point me in the direction of some good reference materials on how the displays and lamp/switch matrix work? By that, I mean the ICs and processors involved, as opposed to the general concept of a matrix, which I understand.

I'm trying to figure out what might be causing the entire lamp and switch matrices to not work. There's got to be some kind of central point of failure that I don't know about, and a lot of stuff out there is written less for a person learning for the first time and instead like a reminder for someone who already knows what they're doing.

OK. I'm back, if anyone's still listening.

I got Leon and Andre's ROMs in the mail and, hearing that Andre's was way better than Leon's, tried that one out first. It reported that PIAs 2 and 3 were bad, so I replaced them. I re-tested, and nothing changed. I should have known it wasn't going to be that easy.

I tried Leon's ROM, and from what I understand, it alternates the outputs of the PIAs once a second. The PIA on the CPU board appears to work fine, but none of the PIAs on the driver board properly do this alternating on the output. So, either one original PIA and two replacements are all bad, or there's some kind of central failure to get them to properly pulse.

I'm trying to work through an old copy of Clay's guide I found out there, but he doesn't really go into detail about this situation.

I'm really at a loss for how to fix this board. I have this sinking feeling that I'm going to end up spending 500 dollars on replacement boards because I offered to help a buy bulletproof his Firepower - an operation I've done twice before with no problems.

I checked the inputs on the PIA on the CPU board as well as the inputs on one of the PIAs on the driver board, and all the voltages on pins 26 through 33 line up. For example, voltage on pin 26 on one PIA is the same as the voltage on the other PIA, which makes sense, since they're all on a shared circuit.

The question is this: How does a PIA know that it should be pulsing its outputs once a second? Why is the PIA on the CPU board able to process the instruction to do this pulsing, yet the PIAs on the driver board don't? There has to be something about the way the test ROM and the CPU work together to tell the PIAs what to do that I'm missing. Is there some other signal that plays a part in telling the PIA what to do?

As a Support Engineer at a tech company, my job is primarily to provide upper-tier support for legacy products that are no longer being actively developed by the company, but continue to make money. This means that I spend a lot of time digging through code that nobody at the company owns, or has any idea what it does. This means I need to rely a lot of pattern recognition and a really vague sense of "the way things ought to be". Something about some of the tests I've been doing have kind of set off that Spidey Sense of mine that something might not be right.

When Leon's ROM is pulsing the PIAs, one of the things I've noticed is that Pin 22 on the CPU board's PIA is pulsing. Not "alternating high and low slowly" but actually pulsing. Pin 22 is CS0 on the PIA (whatever that means) and leads to pin 6 on a 7408 chip (IC12). Pin 6 is the output of pins 4 and 5. Pin 4 is marked as A13 on the address buss and reads high. Pin 5 is marked as VMA (whatever that means) and reads as pulsing. When you AND these two together, you get a pulse on pin 6. Great. Let's assume the same design exists on the driver board as well.

On the driver board, all three PIAs share Pin 22 on the same circuit. This pin leads into Pin 3 of IC7, which is another 7408 chip. On this chip, Pin 1 is A13 on the address buss, Pin 2 is the VMA line, and pin 3 is the output. Pin 1 is high, pin 2 is pulsing, but pin 3 reads nothing.

All of this is with my Elenco Lp560 set to TTL. The readings are different when I switch to CMOS, but they're still inconsistent with each other in the same way. The readings I get from the PIA and the 7408 on the CPU board don't line up with the readings for the PIAs and the 7408 they're all connected to on the driver board.

So, my thinking is that I might have a bad 7408 on this board. I wonder if I messed it up while working on it. Does this sound logical to anyone else? Can anyone reading this try putting a Leon ROM into their System 6 and see if they get something different?

I've also found some weirdness with the two 7402 NOR gates next to the flipper relay on the driver board, IC8 and IC9.

http://datasheet.octopart.com/NTE7402-NTE-Electronics-datasheet-21328372.pdf

Based on the inputs to this chip, none of the outputs line up with what they're supposed to be. For example, Pin 2 and 3 in IC8 have low inputs, but the output on Pin 3 is also low when it should be high. The inputs on pin 8 and 9 are low, but the output on pin 10 is mostly dead with occasional high blips. Am I going crazy? Could both of the 7408s and both of the 7402s be bad on this board?

I feel like I'm running a blog, here. Either the issues I'm having are too in-depth for people to help with, or I'm so incompetent that it's not worth the effort.

None of these PIAs are AMI. All three are Motorola. One is an original, and two are new ones I just bought and socketed.

None of the lines on the interconnect are open. For each pin, I can trace all the way from the component on one side to the component on the other. No problems. Current is flowing freely between all of the pins and their housings.

I don't understand how to get the three PIAs on the driver board working when I have no idea how they work. What is it that tells them to pulse their outputs? This has got to be some kind of "garbage in/garbage out" kind of situation.

Yep. This is where I'm having problems. I'm doing the best I can to test what I think are the inputs on the PIAs, which is what led me to the 7408 I mentioned earlier. Pin 3 on IC7 on the driver board goes right into pin 22 on the three PIAs on that board. PIn 22 for those PIAs is listed as "CS0". Is that an input? Because if that's an input, then a messed-up output on IC7 would most likely cause some kind of problems, no?

Even if not, I'm going to have to replace that chip eventually. A pin that is supposed to be pulsing high, but isn't, is probably going to mess something up down the line.

If Pin 3 on IC7 on the Driver Board is the signal that feeds into pin 22 on the three Driver Board PIAs, and that pin isn't putting out a proper signal, then if I send a pulse into the line that PIn3 is connected to, that could temporarily make up for it, right? What if I jump pin 22 on the CPU board's PIA to the line with the three Driver Board PIA's Pin 22? Does it work that way? Can I just re-route signal like that without breaking anything?

Does floating mean that it's neither high or low? Because that's what PIn 3 on IC7 is doing. Pin 1 is high, pin 2 is pulsing high, and pin 3 is putting out a teeny little tippy tap sound on my Elenco 560. Pins 4, 5 and 6 act weirdly too. I'll replace the whole chip tonight and see if that helps. Fingers crossed.

Success! Replacing the 7408 chip worked the game boots into attract mode successfully. The lamps flash like they're supposed to, and the game recognizes switch inputs. All the regular solenoids work during the game, although the outhole kicker fires once when the game is turned on. I should look into that.

The only thing I have left to deal with is a pop bumper that sticks on when the special solenoids are plugged in (which should be easy enough to troubleshoot) and a display issue where the upper right segment is missing on the master display, as well as player one and two. The master display also has a severely flickering right two digits. The flicker increases greatly the fewer segments are on display.

I'll start a new thread for the display issue, since I know very little about that. I think we can consider this thread closed. Thanks for the help, everyone!