Ran into an interesting one today. Working on a Williams System 7 game. MPU and driver are clean. Power supply section is crazy hacked up. I'll have to get some pictures, but trust me when I say there's some weird stuff going on there.

When you power up the game, it doesn't boot. There's a flash on the LED of the MPU showing a zero for a second before it goes out. Here's the weird part: the flippers, slings, and pop bumpers all work as soon as you power the game up. It's not in attract mode and it's not booted. It's as if someone direct wired to the non-controlled solenoids so they could sort of play the game even though it's not booting. I've never seen that before. Anybody else?

Yeah, interconnect is one of the first things I look at after reviewing the basics. The only reason I stopped right away at the power supply was because of how crazy it looked. I'm talking about the power supply board itself. Hacks and solder blobs, looks like someone started rebuilding it, but didn't have axial capacitors, so they jumpered a radial. No big deal, a lot of us have done that. Some other stuff. I'll take pictures and add them to the thread as soon as I can.

I have not gotten inside the cabinet yet to look at the transformer or anything down in there. So far I've just had a glance inside the head.

So you're saying a bad interconnect can cause non-controlled solenoids to all work, including flippers, even when the game has only been powered on with no boot cycle? I didn't know that if that's true.

That's really interesting, I've never run into that when working on repairing a game before.

I'll set the game up in my workshop and start doing some tests on the voltages and everything. I may have a spare system 7 power supply somewhere. No idea if it works, but it's got to be better than this mess.

snyper2099 thanks for chiming in so far

Okay, so some updates. Set it all up and tested voltages on the Power Supply (with just incoming power hooked up). Everything came out good except for the -100V for the displays. It was Zero. A closer look so a cracked/broken disc capacitor up in that section.

I dug in my pile of boards and found another System 7 power supply. Hooked it up, and all the voltages tested good. I plugged everything back in, powered on the game. (Note: Yes, you'll see in the pics that the GI is not connected to the power supply I swapped in. It was not connectorized on my replacement board. Didn't need it for testing purposes right now)

The displays flashed all 0's quickly, then went to Game ID on the P1 display, 04 00 in the Credit Display. Pretty standard. I had the coin door open, and tried the ON/OFF trick, no luck. I seem to be stuck at the Game ID display. But hey, the good news is all the displays looked good, and the game is at least trying to boot more than I thought it was when they were dead.

Same issue though, flippers/pops/slings all work, but that's it.

Feels like it's time to move on to pulling the MPU/Driver and doing some inspections. One thing I noted, someone soldered a NVRAM module directly to the board, rather than socketing it. Ugh. I'll be looking at that first thing. I need to snag some more interconnect stuff, I used all my stock up recently on some System 3-6 games.

Based on some other similar threads (including some on this same game), guessing the RAM (or NVRAM possible fail install) is a likely culprit. Looking there first.

LQ_1 - Copy (resized).jpgLQ_2 - Copy (resized).jpgLQ_3 - Copy (resized).jpg

Pressing the diagnostic switch on the MPU results in an 8 on the LED display. If I am reading correctly that means IC19 CMOS RAM or memory protect circuitry fail.

snyper2099 I actually did the above (holding the forward coin door button) and when it went through the settings and was supposed to go to attract, it clicked and went back to Game ID / Audit. I tried it with and without batteries, same results.

I removed the board to take a closer look. The NVRAM that was installed (and soldered direct instead of socketed) was done poorly. I just looked, and 3 solder pads were lifted from the board, under the NVRAM. Going to take a closer look now and do some testing, but I'm guessing I don't have good connection on several of those points.

And yep, the previous power supply was likely doing the same thing, just no displays due to -100V missing on it.

Well, I fixed all the issues with that socket, verified it, installed new RAM. Still the same result. Stuck in audit mode.

Stepping through the audits back to reset doesn't fix it, the on/off with the door open doesn't fix it. Verified the coin door switch is registering open and closed as it should at the coin door end.

Interconnect? Memory protect issues on the board?

TP3 shows 0.1 V when the coin door is open, and 3.8V when closed.

Pin 22 of the 5101 shows 4.84V

Interesting. I will give it a try in the morning, thanks

I just remembered, this MPU doesn't have the DIP switches on it. I know some system sevens do, but this game does not. The position where they would be is blank as you can see from the pictures. So I don't think I can perform that procedure.

Sidenote: I realized if I pressed the diagnostic switch with the door closed, it changes from displaying "8" to "9" (which is the same error, except it's letting you know the coin door is closed). This verifies that The board knows whether the coin door is open or closed.

snyper2099 Correct, I socketed the position. I tried booting both with and without RAM installed, same result.

Schwaggs I went through all the traces, they appear to be correct. The ones that were damaged at the solder pad, and the ones that looked okay. I can always check again though. I have tried (3) 5101's I had on hand, including from a working game. I could order another NVRAM module, but I'm starting to think perhaps something failed elsewhere in the Memory Protect circuitry.

I could take a trip over to Allan Davidson I guess, I've done it before and he sorted things quickly with his test rig when I had a weird problem on a System 6.

Looks like I have a reason to swing by Allan's place anyway, as I'm picking up some stuff from another Pinsider not far from there as well. I think I'll just go ahead and haul the boards down to see if he can sort the issue quickly (he usually does). We exchanged some emails and I think that's the best bet. Plus, I always learn something hanging around his shop.

Either way, I'll report back what I find out. Going to go swap another driver board on there, just for kicks, since I have two known good ones.

No change swapping in either of the extra working driver boards I have, so if it's the interconnect, it's going to be the headers on the MPU side I'm guessing. I'll check those out and reflow them perhaps.

Still leaning towards memory protect issues, but we'll see.

Update: The issue is fixed, and the game is booting happily now!

Issue: It turned out to be the 5101 socket/traces. Despite everything seeming to be okay continuity-wise, it turned out that pin 12 had a broken trace that ran under the chip. It was hard to see, and even after having Allan re-socket the chip just in case, it was still missed. Some oscilloscope poking around found that the expected result on pin 11/12 looked off, which brought forth a third inspection, finally noting the almost invisibly broken trace. A nice clean repair to that trace solved the issue, finally.

Sometimes you need a second set of eyes, and a solid triple-check of some things to find the answer, even after you think you've checked it out.

Thanks to all for their help! Happened to get another System 7 board up and running while there, so at least I've got a backup for the next project.

Time to focus on making sure everything works, now that the board is booting up. Going to mark this thread Solved.