Wednesday, July 29, 2009


This was the error message I got monday evening on my Attack From Mars pinball (AFM) machine. The left flipper didn't stay in the up position, so I had soldered in a new flipper coil. The old coil tested bad with my dmm, probably the small wire had broken somewhere internally.

Anyway I played a testgame now the flippers worked again.
Suddenly the game got confused, strobe multiball suddenly started when it shouldn't, and the game kept firing out pinballs.
Something definitely was wrong here.

I turned off the pinball machine and when I switched it on again I got this error on the dmd: CHECK FUSES F101 AND F109, J127 AND OPTO 12V SUPPLY

Ouch. That's bad. The weirdest thing was that I couldn't get past this error, the test buttons at the inside of the coin door just didn't work at all ?!
That's worse.. because I couldn't get into the diagnostics and selftests and see what else the game said was wrong or what was working.

I opened the backbox, all leds were fine though. To be sure I checked all fuses. All ok.
Measured voltage on the test points, 12v regulated and unregulated were fine. So this 12v circuit was working.
J127 was also still connected like it should be, hadn't become loose or anything. Also checked all other connectors (if I'm not mistaken J140 also is needed for the 12v circuit).

Looking at the underside of the playfield I noticed the led on the opto board was still lit. All connectors were still in their place.

Further troubleshooting. The error message means the games computer thinks the 12v power supply is broken because it suddenly can't see any opto switch anymore. These switches are more than the 12v voltage itself. There's also an opto board
at the bottom of the playfield involved, and the switch matrix itself.

I disconnected the wires on the opto board, tested again, plugged in the connectors, tested again..

I don't know if the not working testswitches were related and default behavior ? These are outside of the switch matrix
and have their own connector on the cpu board, direct input switches.

Started to think the cpu board itself was broken.

Disconnected the switch matrix connectors, then the error message changed into 'check fuses 101 and 106'.
Testbuttons still didn't work though.

Anyway as it was getting too late I stopped troubleshooting.
Read Clays guides to refresh my memory about how the switch matrix and 12v opto board worked (hoping to find
a detail I missed that would solve the problem), searched for the specific error message..
Seems not a lot of people have gotten it, there's not a lot of information to find about it and only a few people
who had it but not how they solved it..

Yesterday evening then I continued my search.

Checked the opto board, disconnected all connectors, no difference, still the error.
Plugged them back in, problem wasn't solved.

Played around with several combinations, unplugging direct switches, unplugging switch matrix on the cpu board, .. put pinballs in trough, left them out, ..
Sometimes I got the F101, 109, J127 error, sometimes I got 'Check fuses F101 and F106'. Sometimes the game even went right into attract mode but I couldn't start a game and always still couldn't use the test switches.

As the 12v was still on the board and connectors were plugged in, I focused on the cpu board and the direct input switches. I even considered if the 'enter' testswitch had suddenly become bad and didn't make contact anymore ?

Replaced chip U20 on the cpu as this IC is socketed and easy to change. Still no difference.
I first had taken an IC of my spare parts.. but maybe there was a short in the switch matrix and this new IC would be immediately broken again ?
So I removed it from AFM and put it into my Medieval Madness pinball machine which has the same type of cpu board.

MM still worked fine.. so this U20 IC was ok.

Next step: swap cpu boards.
I removed the cpu board out of AFM and put it in MM. I did unplug the connectors that power the coils, as the AFM roms would operate incorrect coils. Didn't want to risk causing damage to my MM too :-)

Much to my surprise MM booted fine with the AFM board installed. The machine started, went into attract mode, all lamps went on, .. (it gave some errors about stuck switches which is normal as MM has some optos where AFM has regular
switches so it thinks they are stuck).
But I could use the testswitches so the direct input circuit on the AFM cpu board was ok and switch matrix was working too..

Weird. So the AFM cpu board itself was fine and the wiring should be fine too ? Only thing I could think of was a short somewhere underneath the playfield of AFM (although then fuses would have been blown ?), or a problem with the testswitches or their wiring ?

So the next things I did:
- checked the connectors at the interconnect board (but they were ok)
- checked every fuse again, removed it from the holder, measured with my dmm, put back in fuse holder
- put the cpu board back in, but I did remove the batteries so it lost its memory

Switched the game back on.. and it worked fine ?
Gave the expected 'Factory Settings Reset' error because the remote battery holder had been disconnected, but the game booted fine, testswitches worked fine, and I could play.

So this problem suddenly went away ? I have no idea what exactly solved it.

Was it a bad fuse that gave intermittent contact ? Connector that was a bit loose ?
I'm not sure. I guess the cpu somehow got into a corrupt state and removing the battery fixed it (in the past I also have had a few times a weird error that spontaniously went away after a factory reset).

I don't really like this kind of problem. If something is broken, I want to find the source and solve it.
It problems suddenly solve themselves, you're not sure what caused it, and what's more important: you're not really sure it is completely solved and will not come back in the future..

I'm glad it's solved so fast.. still spending too much time troubleshooting problems on my Black Hole pinball machine :-(

Wednesday, July 1, 2009

Flooded Fire

We bought another game this week.. this time it's a special project.

It's a Williams Fire. Problem with it is that it's been in a flooded basement.
The water came just up to the bottom of the backbox.
The playfield is covered with mud, some assemblies are rusted, ..
It doesn't look very good, but the electronics are still fine, so probably this game will still power up without any problem.. The playfield itself is totally covered with mylar so the paint and artwork doesn't look to have had any water damage. Most damage is to the assemblies at the bottom of the playfield which need a very thorough cleaning. But as some people take coils and wire looms into a shower to clean them, these parts can be cleaned without any bad long-term effects.

We're not sure yet if we'll clean and restore the game or just part it out..
it'll be a lot of work..

Pictures are posted on facebook, search for the page of
You really have to see the pictures to get an idea how bad it looks..