Yes, most likely.
Once you disconnect it you’ll know.
Epilogue:
It appears to be fixed, though I don’t quite know why.
This afternoon I:
• Removed green 1 ohm SMD resistor (as meter showed short between both sides of resistor and one side of the LED) - fault still there.
• Replaced with 100 ohm smd resistor (the lowest resistor I had) - fault still there.
• Removed switch pcb completely leaving loose wires - fault fixed.
• Made test rig of two loose 3mm LEDs and two switches - fault fixed.
• Unwired test rig and refitted switch pcb - fault fixed.
I have no idea what it was. The only short I could detect was between one side of the green LED and both sides of the 1 ohm resistor, but that might be my meter deciding that 1 ohm is too low a value so counts as a short (though bizarrely, in resistance check mode, my meter also correctly measured that 1 ohm resistance). But it can’t have been the cause as even with that resistor removed completely (and the LED completely dead obviously) the fault remained. All the joints were clean and tidy, no stray strands, and no specks of solder anywhere, and wire jackets were sound and cut right up to the joint. I examined every wire with the most powerful magnifier I have and found nothing contaminating any of them. The only bridge was the deliberate one between the two LED + pads.
Anyway will keep soak testing, but for now it seems to be sorted. I just don’t like not knowing why. 
I’ve had things like this happen a few times. It’s frustrating, but at least it works now, right?
1 Like
Indeed. The silver lining is that as time goes on, I’m working my way through more and more bizarre and unusual problems which, with yours and others’ generous help in resolving them, will surely make me a better installer in the long run. 
Epilogue to the epilogue: just when we thought it was safe to fire up the saber…
But there’s an interesting outcome…
I just went to add some fonts to this saber at the customer’s request, when suddenly the problem came back, albeit much less severe, but this time from the aux switch. So I measured the voltage coming out of both switch pads: main was rock solid at 3.27 volts, but aux was fluttering between 3.29 and 3.1 over perhaps a three or four second cycle.
You’ll remember that I replaced the smd resistor on one of the LEDs on this pcb in the course of tracing the first switch fault, though it should have made no difference and I wasn’t convinced even afterwards that that was what fixed the first switch problem. But with nothing to lose, I now did the same on the second LED. And blow me down, as soon as I did it, the aux pad gave out a solid 3.27 volts.
What is that about?
There is definitely something funny about these switch pcbs, as the LED lines and the switch lines should all be completely separate from each other, but they’re clearly not. And how can swapping one smd resistor for another realistically affect anything other than LED protection/brightness? The values in this case were a 68 ohm fitted at the factory which I replaced with a 100 ohm (which was all I had).
Anyway it is sorted now, hopefully for good.
