I am experiencing a very strange, intractable ground fault issue. I have a bandsaw that runs on a GCFI/ACFI-protected branch. It has a standard capacitor-start induction motor. Every OTHER time I start the motor, it causes a ground fault, like clockwork (NB: not an arc fault.) I will turn it on, it trips, I reset the breaker, turn it on again, it runs, turn it off and on again, it trips, ad infinitum. It's an old motor, and the wiring is a recent installation. I have another squirrel cage motor on the same branch for a drill press, and it never causes a trip. What on earth could be causing this ground fault on every OTHER motor start?

Edits to address comments: I know it's not an arc fault because the breaker has lights that indicate whether it's an arc or a ground fault or an overload fault that tripped the breaker. It's definitely a ground fault. I haven't tried it on other GFCI branches because it would be a pain in the but to get the saw over there, but I will look into doing so. It does not trip a non-gfci breaker.

The idea that it might be time running, and not the fact of starting it twice is an interesting idea mentioned in the comments. But I don't think that this is the case because I can run the saw for 20 minutes, turn it off, turn it back on again, and have a trip.

  • 1
    Have you checked the band saw for loose wires or worn insulation? It could be that the breaker trip is telling you something... – FreeMan Nov 11 '20 at 12:35
  • How do you know it's not an arc fault trip? – Steve Sether Nov 11 '20 at 14:56
  • I'd also consider that it's not so much every other start, but possibly just time. i.e after running for the first time, it heats up enough, or distributes some internal oil enough to change the conditions. Maybe after it's warm it takes more power, or less power, or the electronics work a bit different? Or maybe it gets warm, something inside expands, and truly does create a ground fault because the hot wire touches the ground. – Steve Sether Nov 11 '20 at 15:04
  • Have you tried it on another circuit, maybe a different AFCI circuit AND a circuit with a normal breaker? – Jimmy Fix-it Nov 11 '20 at 15:13
  • 1
    You don't need to move the saw, just use an extension cord. I won't tell :) For testing I mean. Don't run it that way in daily use. – Harper - Reinstate Monica Nov 11 '20 at 19:28

Well, it's not so much every other time as it is every second time. This suggests that something inside, be it a cap or something else, is storing charge which back-EMFs the second time you start it up. Then the breaker trips, everything dumps, and you start clean. If you can, see if you can :

  1. run the machine, then turn off
  2. use a grounding stick to discharge the cap
  3. turn on again

I know from my days messing with CRT televisions that caps can hold charge for days.

  • I had been thinking along these lines, which is why I put in the details about the motors being capacitor start. But what I couldn't figure out was why the bandsaw motor was tripping but the drill press motor wasn't. The bandsaw motor is 1.5hp and the DP motor is 0.5hp, so maybe the size difference in caps is enough to account for the different behavior? What I don't understand about this potential answer is: if I discharge the capacitors and then start the motor, won't the capacitors recharging also cause a ground fault because the current is being stored in them and not going to neutral? – user278411 Nov 11 '20 at 17:15
  • @user278411 That's not how electrons flow, but I totally understand thinking that and I have trouble explaining why it isn't so. The upshot is that to trip a GFCI without actually faulting to ground, you would have to remove electrons from the wires, and quite a large number of them. The cyclotrons at Fermilab couldn't store that many electrons :) – Harper - Reinstate Monica Nov 11 '20 at 19:33

It might be the AFCI part giving you grief.

A few years ago, I found the previous homeowner had tied some outlets in the master bath into the lighting circuit for half the house (there's an attic light that made that easy). One night my wife plugged her hair dryer up and POW, the lights went out.

I have a subpanel on the wall outside the master bedroom and so I bought a 15A AFCI breaker and added a new circuit to power this and a couple of other electrical additions. One day, I needed to vacuum something up (I have a newer vacuum too), so I plugged up to an outlet on that circuit and accidentally double-bumped the power switch. The vacuum wouldn't work after that, but the upstream GFCI outlet light was out and it wasn't tripped (tried another outlet and the vacuum worked fine). I went outside and, sure enough, the AFCI breaker had tripped (the light told me it was an AFCI trip, and not an amp trip).

AFCI breakers work by "listening" to the circuit to detect the electrical crackles and pops from arc faults. They are supposed to be smart enough to ignore other arcs (like plugging something in and getting a pop), but sometimes powerful motors seem to be able to generate something that the AFCI is mistaking for an arc fault, and thus it trips.

It's entirely possible the capacitor start is doing something (maybe a static charge?) that's tricking the AFCI. It might be you need a better AFCI breaker that won't trip. Either that, or add a dedicated circuit for this one device that uses a regular breaker.


Single phase electric motors (which is what you have unless you have a fancy 3-phase motor, which requires 3-phase power) behave differently on startup than they do after they've started. Essentially they can't start themselves, so there's some tricks that the electronics need to to do get the motor started. This youtube video does a decent job of explaining this.

So that may explain why you sometimes get a trip on startup, and once it actually starts, it behaves without tripping. It's possible something is wrong with the internal electronics that's causing a trip after the device is on. It could be a capacitor that's charged, it could be a solder joint warmed up and is arcing across another joint, etc. If it's as consistent as you describe though, something is creating a memory, and something else is making the device "forget".

In general GFCIs are very simple devices, and a trip (especially one that's consistent as the one you're describing) means that there it's very likely there IS a ground fault happening. All they do is measure the current from the hot, and the neutral, and if they aren't within a small range, it trips the breaker.

To troubleshoot, as others have suggested try the bandsaw on a different circuit. This will isolate the problem from the wiring, to the bandsaw itself. You could also install a GFCI outlet, (or plug it into one) and see if it trips. If the bandsaw trips everything it's plugged into, there's almost certainly some internal problem with the bandsaws electronics.

If it turns out to be the bandsaw, you're now trying to troubleshoot electronics, which is largely beyond the scope of diy. Maybe the electronics people could help you troubleshoot?

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.