Ground Fault tripping on house related to pool pump turn-on - how to fix?

Status
Not open for further replies.
Update

Update

It has been a few months now and I wanted to update members who provided thoughts on this mystery above (greatly appreciated!).

We had a licensed electrician come out and (a) swap out virtually all of the "dual function" (AFCI/GFCI/Overload) breakers; and, (b) reposition the whole-house surge protector at the top / where current enters the panel (from near the end/bottom of the main panel). The pool company sent a tech to ohm out the motor and controller to ensure no overt anomalies were happening with the pool pump. They suspected the main pool breaker at 60A was seeing too much steady-state current draw at ~50A and changed it to 70A (leaving all pool equipment breakers at the panel on the opposite side of the house as is). The electrician did not measure too much current, about 25A when he was present (for steady-state operation of heat pump and pool pump combined).

The original issue of spontaneous, random main circuit breaker trips of the dual-function breakers on the main panel when the pool heat pump plus pool water pump turn on in the mornings did not disappear. I think it might have become a little less frequent, but that might be an optimistic view. Breaker #27 was most often the one tripping, so in desperation (having been taught how to change these, and noting that the particular circuit did not really need AFCI or GFCI), I went to home depot and got one of the simpler, Schneider HOM overload-only breakers with the same rating and changed it out. Rock solid, this has never tripped since. (As a reminder, the tripping has always been ground fault related for any of the breakers.)

However, once in awhile we have seen other breakers still tripping - not frequent, maybe once or twice a week. (We have also had lightning storm related surges that have caused few breaker trips too, but that must be a different issue though still speaks to the hair trigger behavior of these newer breakers that frankly don't seem ready for prime time... a real problem and worry for us if say, we are out of the house on vacation travel or whatever and circuits start turning off in the house!)

So, we have been noting the circuits remaining at issue (all 15A) and I plan to similarly change these out for regular (old-fashioned) breakers, backing in GFCI at the outlets if needed to compensate for removing at the breaker. This as recommended by the pool electrician, who cannot do it for me as this is technically not current code, but he was sympathetic.

Only other thing we can think of was to try pulling bigger wire (currently 60A rated) but go to the next level (even though it should never draw more than 60A) to try to smooth out anomalous ground fluctuations/start-up transients. A more recent idea here is instead, to put the heat pump on a new home run circuit apart from the pool pump. Given the costs and the uncertainty as to whether this would be a sufficient fix in the end, is driving us to simply go with the tried and true old-fashioned breakers for the ones that still trip (~3-5).

So that's where we are. No smoking gun found yet and everyone seems mystified by this behavior, but we're likely going to do the simple breaker changes. I'll update folks when this happens to let you know if this works.
 
It seems that you may have missed the points brought up in the first go around. This isn't about the breakers, it's about the variable speed pump. That pump has the equivalent of a Variable Frequency Drive, a technology well known to cause havoc in GFCI breakers of almost all stripes. the pump mfrs seem to not want to address it in their design, where it COULD be addressed, because it will involve increasing their costs. So instead, they PRETEND to have never heard of this before, knowing full well that when talking to the HOMEOWNER, they are talking to someone who will not likely have ever installed a VFD in their lives. But I guarantee you, THEY have heard it all before, many many many times, as we have in this forum and several other forums I have seen. Their preferred solution is to say that if you use a Siemens breaker, it won't trip. That SEEMS to be anecdotally true, but I have also seen people say they USED a Siemens breaker, and Hayward told them to use a Sq. D.!

The basic issue is that VFD technology (PWM firing of transistors from a DC bus in patterns that make the motor ACT AS IF it is getting AC current), causes a significant amount of Common Mode Noise (electrical noise) that operates at high frequencies and is traveling on all conductors trying to return to its source, the DC bus in the drive. It will end up being expressed on the AC incoming lines, and that's where it is seen by the GFCI circuitry as if it is superfluous current flow, which the circuit assumes means there is a ground fault.

The only things I have heard of that work are to INCREASE the size of your ground wire, because in many cases people use an undersized ground wire at the motor because the NEC says you can, but PHYSICS demands that you no do that. That may not be the only problem however. The other thing I have confirmed works is to add a shielded isolation transformer ahead of the pump / VFD. Same voltage in and out, but the isolation stops the CM noise from getting through. As was mentioned months ago by Fionna Zuppa, there has been talk of adding "ferrite beads" to the wires feeding the pump/VFD, but nobody has posted back if that has worked. I suspect it might, because ferrite beads (chokes) suppress high frequency noise on wires, which is ultimately the source of the problem. You can buy ferrite beads at electronics stores or on-line. If you try it, let us know if it worked!
http://palomar-engineers.com/rfi-kits/acdc-power-line-chokes
 
I think you guys are looking into this way to much. I think it's bad unground wires from 60amp non gfci breaker and control panel. If wires are bad that would not trip breakers in control panel. And a non Gfci 60 amp would not trip. I would kegger the wire just out of curiosity. Seems to happen only under a huge load. Just a dumb electrician though.

Sent from my SM-G930V using Tapatalk
 
Jraef said:
The other thing I have confirmed works is to add a shielded isolation transformer ahead of the pump / VFD. Same voltage in and out, but the isolation stops the CM noise from getting through.

+1

On the load side of the transformer install GFCI protection ahead of the pump/VFD

I am not going to reread all this thread.....
If the pump/VFD line voltage is 120V I would probably use a 240V to 120V step down isolation transformer. Just going from memory the feeder from the main panel to outside pool equipment is fed from a 2 pole 60 amp breaker.
 
Status
Not open for further replies.
Top