I've just finished installing a new main loom to my car (1989 City). Other work I've carried out at the same time is to a) replace the short headlight looms b) fit relays into the main beam and dipped beam circuits, wired in the usual manner c) re-wired the aftermarket alarm which had been done quite shoddily by the guy who sold it me and is hopefully safer now. Dim-dip relay is bypassed at the holder under the dash if that's relevant, but it's been like that for years.
I may have come close to narrowing down the source of the problem but still can't understand the logic behind it.
Everything seems to work perfect except for one strange thing: there is a current leakage somewhere between the sidelight circuit and the hazards/indicators.
I know this because when I turn the sidelights or headlights on, the two indicator bulbs on the dashboard glow dimly and there is no gauge illumination. There is, however, no detectable glow to the indicator bulbs themselves. The sidelights and headlights themselves work perfectly, all the time.
On the other hand, when I turn the hazards on, I hear the headlamp relays click, the headlamps flash once and once only, and then the hazards carry on working as normal! Very strange!
The problem is intermittent - about 50% of the time these two strange things happen, and the rest of the time everything works as it should.
I think I may have narrowed down the problem to the multiconnector which goes into the PCB at the back of the instruments. The multiconnector which came with the new loom is very poor quality (the rest of the loom is excellent). It feels as though it's slightly too loose to make reliable connections. Sometimes I can get trigger these malfunctions by jiggling the connector. However I'm really not looking forward to chopping this off and soldering the original one back on whilst the loom is in situ, so I'd like to be certain first that this is where the problem lies.
Exactly how could bad connections at this connector cause the two malfunctions I'm seeing?
If the black earth connection to the PCB is dodgy, could this somehow cause the sidelights to find an earth via the indicators? Looking at the Haynes circuit diagram I really can't see how. Is it doing this on the front wing or on the PCB itself?
Or could a bad red/green connection to the PCB cause this behaviour?
Or bad indicator connections to the PCB?
Note: strangely, the main-beam tell-tale light works without fail, which I wouldn't have expected if there's a bad earth on the multiconnector. Also the ignition light and oil light work without fail but I see from the circuit diagram that these don't rely on the black earth wire.
The problem can't be with the PCB as I've tried no less than THREE different instrument binnacles, the same occurs with each. Also, I bench tested one before fitting and everything worked fine.
I don't think the problem can be in the new relay circuits I've fitted, as these have nothing to do with any indicator wiring.
Similarly, I don't think the alarm wiring's at fault because even though it's connected to the indicators, it goes nowhere near the headlamp circuits and nowhere near terminals 7-8 on the fusebox.
How would I go about diagnosing this with a multimeter, test lamp or Sealey probe?
[Apologies for the long post, conciseness isn't one of my talents]