Troubleshooting GFCI fault in garage

Question

I recently moved into a home that's about 20 years old and have a problem with the 15 amp outlet circuit in the garage. I was using a 12 amp table saw on the third and last outlet along the wall when the GFCI (the first outlet in the series) tripped. I had powered the saw up several times without any trouble, but one time the GFCI tripped right after I powered up, then would not reset.

Turning off the breaker and checking everything, I found that the GFCI outlet and the two ordinary outlets in the series were good, but the third (end) outlet had wires that continued downstream from it to unknown points. When I disconnected the downstream circuit from the third outlet, the GFCI reset successfully and all outlets worked fine.

As a temporary measure, I terminated the downstream circuit with wire nuts and electrical tape, so whatever it's for, it's now a dead circuit.

Though there are a couple other outlets on the other side of the garage and a garage door opener outlet on the ceiling and a garage door opener button switch all connected to the same GFCI, everything is still working correctly. I cannot find any outlet or device powered by the circuit downstream of the outlet I plugged the tool into.

My questions are 1) Any idea what caused the fault? I suppose there was a startup current spike that damaged something downstream before the GFCI tripped. 2) Any tricks or other ideas how I can find where the circuit goes and what it's for? It goes upward toward the ceiling from the outlet. I don't see any junction box or device powered by the circuit. Directly above the garage at that point is the living room, and every outlet/light is fine up there.

EDIT 1: Following Rab's advice in the accepted answer, I tested for continuity with the multimeter between hot/ground and neutral/ground in the downstream circuit that trips the GFCI when connected. Hot/ground are not connected, but neutral/ground do show connectivity with a resistance of about 40 ohms. So it seems to be a permanent neutral-to-ground fault at some unknown point. I will follow Rab's advice on tracing the circuit but have to buy a new NCV detector as suggested!

Answer 1

1) Any idea what caused the fault?

Of course not being there to diagnose the fault with test equipment, all I can do is try to cover all reasoning.

GFCI faults and motors

Speaking generically, upon how a motor could cause a fault detection in an RCD/GFCI device:

• a motor can indeed cause an inrush of current leading to a transient current inbalance between the line conductors and resultantly trip an RCD/GFCI as stated.
• if the motor causes a parallel arc to ground/earth this could cause an RCD/GFCI trip as an inbalance between conductors is detected (a genuine ground fault/GF occurs). This could be a result of faulty wiring within the motor or just its normal operation. Some motors cannot be fitted to RCD/GFCIs or require a specialised device to be fitted.
• If electromagnetic line interference is caused on the line by the motor it could cause the GFCI to trip due to a current disbalance.
• if an electromagnetic harmonic is fed back from the motor on the line conductor to the GFCI this could in itself cause the GFCI to trip due to harmonic "eddy currents" forming on the GFCI's solenoid.
• If the motor is quite modern and uses a PWM controller to control the speed, the DC of that PWM signal can reflect on the line conductor. Depending on the type of RCD/GFCI this can either cause a false trip or "blind" the device from working at all.
• A fault in the RCD/GFCI device itself. The solenoid windings eventually fail and they may trip incorrectly. It's not beyond all reason that the solenoid could have failed due to either the transient voltage changes or DC reflection occurring due to the motor. But given that it's "working" now, this is maybe a real edge case.

Some of the subjects covered above are better explained by Fluke in their oscilloscope/multiscope literature: https://www.fluke.com/en/learn/blog/motors-drives-pumps-compressors/13-causes-of-motor-failure

A permanent fault in the additional spur

However taking into account the assumption that there's an additional circuit spur upon which a permanent fault has occurred coinciding with the motor starting:

• the motor itself could have caused a fault in a sensitive electronic device (such as a smart light or USB charger module) putting the device in a situation where it now trips the RCD/GFCI. I realise that's a really broad and vague bit of reasoning, but without finding a specific device causing the fault and diagnosing with a multimeter or multifunction tester, it's difficult to not be vague.
• Very speculatively; the motor could have exacerbated an existing wiring fault with the insulation causing the impedance between live/hot to earth/ground, or a grounded conductor, to decrease below an acceptable level, causing the wiring itself to trip the GFCI. This would be picked up on a insulation resistance test, but given those testers are specialised (~$700+), you might want to grab a ~$12 multimeter and just check for direct continuity between the conductors. Basically test if hot and ground are connected together like they shouldn't be.

NB. Do we even know actually this is the case, that the downstream circuit has developed a fault directly correlated with the motor starting? Is it at all possible that this circuit wasn't even making contact (i.e. it was never even electrically connected at all) and when you went to diagnose the fault, you opened the socket, you found a circuit that, when connected in itself causes a GFCI fault entirely discrete to your motor fault? I just wouldn't make any assumptions with such a "voodoo" problem. It could just be a bizarre set of coincidences. Again an electrical continuity/resistance test could show up a wiring fault without seeing it.

2) Any tricks or other ideas how I can find where the circuit goes and what it's for?

Use a non-contact voltage detector in order to trace the cable. Energise the circuit then run a voltage detector along the your wall and it will tell you where the cable goes by detecting the line conductors' electromagnetic field (EMF). Given that the circuit is not energising with the additional spur connected, you could bypass the GFCI temporarily, reconnect and energise to trace the cable. Maybe it's be something you wouldn't think of, like mains-powered smoke detectors.

BTW with regards to NCV testers, I'd recommend getting a little DIY one that's meant to trace cables behind walls. The electrician's pen type ones usually aren't sensitive enough to go through drywall/plaster. https://www.amazon.com/Sperry-Instruments-PD6902-Multi-Scanner-Functionality/dp/B00173CTIA/ref=sr_1_13?crid=30EP8FQM91HOE&keywords=voltage+stud+detector&qid=1662381389&sprefix=voltage+stud+detector%2Caps%2C120&sr=8-13