REALLY hoping someone can help...

I have had my Dr Who for several years, and it was working fine yesterday.

I bought a set of Cliffys and in the process removed the MPF in order to fit them - not a big deal; I have previously replaced the motor and found Faz's guide to be absolutely excellent (thanks, Faz!)

Upon reassembly, everything worked as expected with the exception of the 5 MPF Opto buttons. They don't register in self-test (Switch Edges, or individual switch)

I double-checked everything, and noticed that on the connector that leads to the MPF optos (12-way) there was a break in one of the wires - it was attached to the molex but frayed / broken near the connector. The wire was black in color, located top right on the connector array if looking at it face on with the two notches at the top see here:

The 12-way with the frayed wire repaired

I assume black is a ground wire? Anyway, I resoldered it back together and.. nothing changed - weird...

Looking at the other side of the connector (attached to the loom that heads upstream to the backbox) I noticed that on the corresponding plug on the Molex there were two black wires leading out of it, but both had obviously been snipped - a nice clean break, fairly close to the molex.

Looking further up the loom, I could see a black wire had been separated from the loom, and twisted together with another black wire, which terminated in a nine-way Molex, with a "null" key diagonally opposite the black one:

The 9-way that was bridged from the loom

I should also state that I ran the self-tests, and the optos didn't register, tried the digicam trick and could see that there was no illumination on the optos themselves.

So, I then ran a wire from one of the pair of "snipped" blacks leads to the "bridging" pair - essentially restoring (I think) an earth path to the 12-way Molex. I checked for continuity with my DMM: "twist"->upstream connector->downstream connector->optos - this appeared to work somewhat (they "lit", and would register during the MPF self-test.

Thinking I had managed to solve the problem, I reassembled everything (boy, do I *hate* that MPF )

Where I am now - the optos are not registering again.

It seemed to me that by wrapping / unwrapping the twisted set of three black wires I could sometimes get the optos to register in self-test, but upon reloading the balls and actually trying to play a game it would (almost without fail) stop working again.

At this exact moment I can measure electrical continuity on the black wire from the "twisted trio" to the "downstream connector" but the optos do not register in self-test.

So, a theory and a couple of questions

Theory: The frayed / broken black wire in the 12-pin connector that I originally fixed wasn't needed, because the previous owner had bridged the black wire directly to the second connector.

a) Is the earth wire supposed to daisy chain between those two connectors?
b) Assuming so, how does the Opto board in the MPF get an earth path, if there was no path from the 12-way molex? In other words, how could that board ever have worked with no earth path from the 12-way?

And of course, the biggest question - what did I break when I removed, and reinstalled, the MPF? I should stress that apart from the Opto board, everything appears to work as expected (all flashers, solenoids, motors etc in the MPF behave as usual)

Is there anything that you can think of in this setup (with or without my extra bridge wire) that would cause the optos to work occasionally?

Thanks in advance for any help, this is as I say, a real brain-scrambler.

So, I somewhat answered a bit of my own question - the intermittent behavior seemed to be a dry solder joint / poor connection at the 12-way connector, by wobbling it a bit I could get it to stay connected reliably -> 4 out of the 5 optos are working. (probably have to get into the MPF again and see about the other one)

Which leaves me stuck on the original head-scratcher - if a previous owner bypassed the ground circuit entirely (which I still think that they did) - how could this ever have worked? Is there another path to ground for the optos?