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

[dns718]

[FONT=Open Sans, sans-serif]Hi, I’ve seen quite a few posts about tripping breakers that are associated directly with the pool pump. In my case, I have a new house + new pool with, varying by day, 1 or 2 so-called “3-way” (Schneider Electric Square-D brand overload/arc/ground fault detect) breakers that are tripping - on the main house panel. After some detective work and data collection, we see a very high correlation between the time of the pool pump turns on and begins priming and when the other house breakers trip. The pool pump is hooked into its own junction box (breaker stable), that runs to a 60A breaker (also stable) on the house main panel. We have a Hayward SP3400VSP EcoStar Variable Speed Pump. It was set by the pool company Aquatech to prime for 2 minutes at the maximum speed of 3450 rpm, which appears to cause the trips in other part of the house panel. We altered the turn-on time of the pool pump, and indeed the time of the trips are following this time exactly, so we have high confidence at this point in the cause-and-effect of the correlation seen. After some time invested with the nice support engineering staff at Schneider Electric in North Carolina, have learned further that these trips are uniformly due to ground fault interrupts (vs. overload or arc conditions – these breakers have a way to discover type of last trip). We have as of today started an experiment to reduce the maximum prime/turn-on speed from 3450 rpm to 3000 rpm in an effort to reduce the inrush current demand/surge to see if this reduction helps in the ground fault, will update later on if this helps. I should add that the run from the main house panel to the pool pump junction box is about 60′ but may have up to 80′ of wire, and runs through an under-slab conduit put in prior to house foundation (wire pulled later of course and was sized to 60A / 220v – not sure as to exact gauge; pool pump is about 11A supposedly; trench is about 18"). The pool pump is grounded along with the control electronics on the side of the house. In my pre-retired life in engineering of micro-circuits, ground shifts in PCBs were a constant challenge requiring extensive capacitive filtering, so I am wondering if anyone has heard of whether this kind of problem could require some added high-voltage capacitor / filter at the pool pump itself; Schneider Electric thought this might interfere with the correct performance of the circuit breakers. After I’ve collected some more data I will go back to my builder or another independent electrician but wondering if anyone else has seen this? My pool company/electrician has not seen it before and I’m not finding this case on the web as yet… Appreciate any thoughts or insights! Seeing some submissions on surge protector possibly contributing to nuisance ground faults, I'll add that there is a whole-house surge suppressor installed by the local electrician (brand "SyCom") on the main panel, with dual-breaker 35A installed (in parallel most likely to other panel circuits given location). Thanks[/FONT]

 

[Dennis Alwon]

I can only tell you that Hayward pumps have has lots of issues with GFCI breakers but this is the first I heard of it tripping afci breakers in the house. I think the first thing I would do is get the pool company to try a different brand of pump.

 

[Jamesco]

Hi, I’ve seen quite a few posts about tripping breakers that are associated directly with the pool pump. In my case, I have a new house + new pool with, varying by day, 1 or 2 so-called “3-way” (Schneider Electric Square-D brand overload/arc/ground fault detect) breakers that are tripping - on the main house panel. After some detective work and data collection, we see a very high correlation between the time of the pool pump turns on and begins priming and when the other house breakers trip. The pool pump is hooked into its own junction box (breaker stable), that runs to a 60A breaker (also stable) on the house main panel. We have a Hayward SP3400VSP EcoStar Variable Speed Pump. It was set by the pool company Aquatech to prime for 2 minutes at the maximum speed of 3450 rpm, which appears to cause the trips in other part of the house panel. We altered the turn-on time of the pool pump, and indeed the time of the trips are following this time exactly, so we have high confidence at this point in the cause-and-effect of the correlation seen. After some time invested with the nice support engineering staff at Schneider Electric in North Carolina, have learned further that these trips are uniformly due to ground fault interrupts (vs. overload or arc conditions – these breakers have a way to discover type of last trip). We have as of today started an experiment to reduce the maximum prime/turn-on speed from 3450 rpm to 3000 rpm in an effort to reduce the inrush current demand/surge to see if this reduction helps in the ground fault, will update later on if this helps. I should add that the run from the main house panel to the pool pump junction box is about 60′ but may have up to 80′ of wire, and runs through an under-slab conduit put in prior to house foundation (wire pulled later of course and was sized to 60A / 220v – not sure as to exact gauge; pool pump is about 11A supposedly; trench is about 18"). The pool pump is grounded along with the control electronics on the side of the house. In my pre-retired life in engineering of micro-circuits, ground shifts in PCBs were a constant challenge requiring extensive capacitive filtering, so I am wondering if anyone has heard of whether this kind of problem could require some added high-voltage capacitor / filter at the pool pump itself; Schneider Electric thought this might interfere with the correct performance of the circuit breakers. After I’ve collected some more data I will go back to my builder or another independent electrician but wondering if anyone else has seen this? My pool company/electrician has not seen it before and I’m not finding this case on the web as yet… Appreciate any thoughts or insights! Seeing some submissions on surge protector possibly contributing to nuisance ground faults, I'll add that there is a whole-house surge suppressor installed by the local electrician (brand "SyCom") on the main panel, with dual-breaker 35A installed (in parallel most likely to other panel circuits given location). Thanks

I should add that the run from the main house panel to the pool pump junction box is about 60′ but may have up to 80′ of wire, and runs through an under-slab conduit put in prior to house foundation (wire pulled later of course and was sized to 60A / 220v – not sure as to exact gauge; pool pump is about 11A supposedly; trench is about 18"). The pool pump is grounded along with the control electronics on the side of the house.

GFCI breaker for this motor branch circuit is installed in the house electrical panel, correct? JMHO, it's the length of the branch circuit wiring that is causing the problem, coupled with there is a good chance the branch circuit ungrounded conductors could be laying in water inside the underground conduit.

I would suggest you hire an electrician and look at installing the needed GFCI protection closer to the pump.

 

[retirede]

GFCI breaker for this motor branch circuit is installed in the house electrical panel, correct? JMHO, it's the length of the branch circuit wiring that is causing the problem, coupled with there is a good chance the branch circuit ungrounded conductors could be laying in water inside the underground conduit.

I would suggest you hire an electrician and look at installing the needed GFCI protection closer to the pump.
[/COLOR][/FONT]

They way I read the original post, the breakers that are tripping are combos that are not even on the pump circuit.

 

[stephenweyant]

hello dns

can you verify a few things regarding your issue?

in the main panel you have a GFCI breaker feeding power to a disconnect 60' or so away located at the pool pump motor? and when the pool pump motor turns on via programmable interface, you have breakers unrelated to the pool pump located in the main panel in the house that trip?

is there any kind of control cabinet located at the pump like a Hayward Omni logic or Prologic?

 

[stephenweyant]

you mention this pump, Hayward Ecostar, is being fed by a 60 amp GFCI breaker? the equivalent HP of this pump is 3hpn(3.45 total HP). is there anything else connected to the 60 amp branch circuit? does it feed a control cabinet at the pump that also operates a heater, salt generator, solenoid valves, etc?

 

[ptonsparky]

They way I read the original post, the breakers that are tripping are combos that are not even on the pump circuit.

That's the way I read it.

 

[Jamesco]

They way I read the original post, the breakers that are tripping are combos that are not even on the pump circuit.

OP said:
In my case, I have a new house + new pool with, varying by day, 1 or 2 so-called “3-way” (Schneider Electric Square-D brand overload/arc/ground fault detect) breakers that are tripping - on the main house panel. After some detective work and data collection, we see a very high correlation between the time of the pool pump turns on and begins priming and when the other house breakers trip.

OP said:
We altered the turn-on time of the pool pump, and indeed the time of the trips are following this time exactly, so we have high confidence at this point in the cause-and-effect of the correlation seen. After some time invested with the nice support engineering staff at Schneider Electric in North Carolina, have learned further that these trips are uniformly due to ground fault interrupts (vs. overload or arc conditions – these breakers have a way to discover type of last trip).

The other breakers in the panel that trip when, I assume the 60 amp GFCI breaker trips for the pool pump, and possibly other related pool equipment ???, are Combination GFCI/AFCI breakers.

staff at Schneider Electric in North Carolina, have learned further that these trips are uniformly due to ground fault interrupts

IF I understand the OP's original posted message correctly.

 

[stephenweyant]

I think I would change the 60 amp gfci breaker to a standard breaker.
at the pump control panel I would install a Pentair PA220GF breaker (20 amp gfci siemens breaker) for the Ecostar pump.
the 60 amp GFCI, if feeding a control panel, sees a lot of "noise" from multiple pieces of equipment.
this could fix your issues in the main panel

if you are feeding the ECOstar pump only with a 60 amp breaker, its sized too big. a 3 hp pump will have a FLA of 14.4 amps +/-

 

[Jamesco]

I think I would change the 60 amp gfci breaker to a standard breaker.
at the pump control panel I would install a Pentair PA220GF breaker (20 amp gfci siemens breaker) for the Ecostar pump.
the 60 amp GFCI, if feeding a control panel, sees a lot of "noise" from multiple pieces of equipment.
this could fix your issues in the main panel

if you are feeding the ECOstar pump only with a 60 amp breaker, its sized too big. a 3 hp pump will have a FLA of 14.4 amps +/-

dns718 Said:
I should add that the run from the main house panel to the pool pump junction box is about 60′ but may have up to 80′ of wire, and runs through an under-slab conduit put in prior to house foundation (wire pulled later of course and was sized to 60A / 220v – not sure as to exact gauge; pool pump is about 11A supposedly; trench is about 18"). The pool pump is grounded along with the control electronics on the side of the house.

We have a Hayward SP3400VSP EcoStar Variable Speed Pump. It was set by the pool company Aquatech to prime for 2 minutes at the maximum speed of 3450 rpm, which appears to cause the trips in other part of the house panel.

Is that Freq Drive? Some of those, cheaper built ones, can really spew harmonics back out on the AC mains.

 

[kwired]

The other breakers in the panel that trip when, I assume the 60 amp GFCI breaker trips for the pool pump, and possibly other related pool equipment ???, are Combination GFCI/AFCI breakers.

staff at Schneider Electric in North Carolina, have learned further that these trips are uniformly due to ground fault interrupts

IF I understand the OP's original posted message correctly.

I believe he further clarified that by using the diagnostic function that Square D has on the latest generation AFCI's:

press and hold test button while resetting breaker, if it trips in less then one second that means it has been tripping on GFCI function, if it trips in two seconds it means it has been tripping on AFCI function, If it trips in five seconds there is no fault, or it has been tripping on short circuit or overload functions.

I believe all he was saying was after talking to Schneider, they determined the breakers in question are tripping on GFCI function. Problem is they are breakers not associated with the circuit that is causing the trips, tells me there is probably some sort of interference going on causing this trip condition, like maybe RF interference. RF from non associated circuit has been something that gets brought up on this forum from time to time causing tripping GFCI's and AFCI's.

 

[ptonsparky]

I believe he further clarified that by using the diagnostic function that Square D has on the latest generation AFCI's:

....Problem is they are breakers not associated with the circuit that is causing the trips, tells me there is probably some sort of interference going on causing this trip condition, like maybe RF interference. RF from non associated circuit has been something that gets brought up on this forum from time to time causing tripping GFCI's and AFCI's.

This happens enough that it’s beginnig to be not news at all. Just different circumstances.

We also know that while, according to SQ D, there is nothing wrong with older models the newer ones are less apt to randomly trip.

 

[Jamesco]

I believe he further clarified that by using the diagnostic function that Square D has on the latest generation AFCI's:

press and hold test button while resetting breaker, if it trips in less then one second that means it has been tripping on GFCI function, if it trips in two seconds it means it has been tripping on AFCI function, If it trips in five seconds there is no fault, or it has been tripping on short circuit or overload functions.

I believe all he was saying was after talking to Schneider, they determined the breakers in question are tripping on GFCI function. Problem is they are breakers not associated with the circuit that is causing the trips, tells me there is probably some sort of interference going on causing this trip condition, like maybe RF interference. RF from non associated circuit has been something that gets brought up on this forum from time to time causing tripping GFCI's and AFCI's.

Agree. +1

Problem is they are breakers not associated with the circuit that is causing the trips, tells me there is probably some sort of interference going on causing this trip condition, like maybe RF interference. RF from non associated circuit has been something that gets brought up on this forum from time to time causing tripping GFCI's and AFCI's.

Could it be the freq drive controller for the pump motor causing the problem?

I have reread the OPs original posted message over at least a couple more times and I can't find where he definitively says the 60 amp GFCI breaker, for "variable speed pump", is tripping out on ground fault.

OP said:

The pool pump is hooked into its own junction box (breaker stable), that runs to a 60A breaker (also stable) on the house main panel. We have a Hayward SP3400VSP EcoStar Variable Speed Pump. It was set by the pool company Aquatech to prime for 2 minutes at the maximum speed of 3450 rpm, which appears to cause the trips in other part of the house panel.

The problem could be a defective freq drive or just a poorly designed freq drive.

 

[kwired]

Agree. +1




Could it be the freq drive controller for the pump motor causing the problem?

I have reread the OPs original posted message over at least a couple more times and I can't find where he definitively says the 60 amp GFCI breaker, for "variable speed pump", is tripping out on ground fault.

OP said:

The problem could be a defective freq drive or just a poorly designed freq drive.

I agree, likely a defective or poorly designed drive or application of said drive.

RF interference from drives is nothing new, we just haven't been seeing them outside of industrial applications all that much over the years and are now seeing more of them - even in/around dwellings at times.

 

[dns718]

GFCI breaker for this motor branch circuit is installed in the house electrical panel, correct? JMHO, it's the length of the branch circuit wiring that is causing the problem, coupled with there is a good chance the branch circuit ungrounded conductors could be laying in water inside the underground conduit.

I would suggest you hire an electrician and look at installing the needed GFCI protection closer to the pump.
[/COLOR][/FONT]

Thanks - there is a GFCI breaker at the pool pump junction box (20A) that ties back to the main house panel non-GFCI breaker (60A, sized to include pool heat pump as well). I believe the junction box (branch circuit) is grounded.

Today's single data point with the Hayward pump prime period (first 2 minutes of operation) maxed at 3000 rpm (vs. former default 3450 rpm) resulted in no house ground trips. But the nature of the problem is marginal and sporadic, so will continue to collect data in coming days to see if this made a sufficient difference.

 

[dns718]

I can only tell you that Hayward pumps have has lots of issues with GFCI breakers but this is the first I heard of it tripping afci breakers in the house. I think the first thing I would do is get the pool company to try a different brand of pump.

Appreciate the reply - yes it seems from posts *and* from Hayward's own admissions that they see GFCI breakers often nuisance tripping. Apparently it is an odd effect in our case that we are seeing our 3-way breakers in the actual house main panel (vs. pool junction box panel) tripping. Just to clarify, these are consistently tripping as ground faults (vs. AFCI trip), according to the Schneider Electric Square D test (while in "off" position, press the "test" button, then turn on and count number of seconds to trip to indicate prior trip condition, with immediate=ground fault, 2 seconds = arc fault, and 5 seconds = overload [or no prior trip]).

So after reading lots of posts here, I'm wondering now if we might have two events interacting: a) strong in-rush current from Hayward pump causing voltage drop followed by spike as it "settles", followed by b) SyCom whole house surge suppressor clamping the spike with result that c) ground shift interpreted by various main panel 3-way breakers interpreted (erroneously) as ground fault. I think my case of the latest 3-way breakers + pool pump + whole house surge protector might be a new and possible rare combination...

My next experiment, following reduction of initial 2 minute prime pump rpms (from 3450 to 3000 rpm) will be to close/turn off the 35A SyCom breakers. Today's first data point with reduced rpms had *no* trips - but will need additional data to be checked in coming days.

If I continue to have issues, I'll consider using all the data collected to go back to the pool company and ask for a different pump as you suggest.

 

[dns718]

They way I read the original post, the breakers that are tripping are combos that are not even on the pump circuit.

That's right! Strange thing...

 

[dns718]

hello dns

can you verify a few things regarding your issue?

in the main panel you have a GFCI breaker feeding power to a disconnect 60' or so away located at the pool pump motor? and when the pool pump motor turns on via programmable interface, you have breakers unrelated to the pool pump located in the main panel in the house that trip?

is there any kind of control cabinet located at the pump like a Hayward Omni logic or Prologic?

Yes, that is right - house main panel 3-way breakers 60' away are the ones popping (while local pool junction box 20A GFCI breaker is rock-solid stable). The pump control panel ("Digital Control Interface") is mounted on the pump itself (integrated - no remote control installed). There is a salt Chlorinator control box but no OmniLogic or ProLogic control per se... Couple of other items near the pump like the pool light, Hayward heat pump and Rainbird / Berkeley irrigation pump/control - all on different breakers in the pool junction box. The irrigation is timed not to run concurrently with any pool items; the pool heat pump does turn on about 45 seconds into the priming period, which might be contributing to the intense current draw. As mentioned elsewhere, today's experiment where I notched down the max prime (first two minutes) of variable Hayward pump operation from 3450 rpm to 3000 rpm resulted in no house panel trips; not conclusive yet but will continue to see if this is enough to make a difference or help with the dataset overall.

 

[dns718]

The other breakers in the panel that trip when, I assume the 60 amp GFCI breaker trips for the pool pump, and possibly other related pool equipment ???, are Combination GFCI/AFCI breakers.


IF I understand the OP's original posted message correctly.

The 60A pool breaker at the main panel does not trip (also is not GFCI); this 60A circuit goes to the pool pump pad junction box area (on side of house, ~60' away); the Hayward SP3450VSP pump is tied to 20A (GFCI) breaker. Neither the 20A GFCI nor 60A "main" breaker have ever tripped - only the "3-way" breakers for other house circuits(!).

 

[dns718]

I think I would change the 60 amp gfci breaker to a standard breaker.
at the pump control panel I would install a Pentair PA220GF breaker (20 amp gfci siemens breaker) for the Ecostar pump.
the 60 amp GFCI, if feeding a control panel, sees a lot of "noise" from multiple pieces of equipment.
this could fix your issues in the main panel

if you are feeding the ECOstar pump only with a 60 amp breaker, its sized too big. a 3 hp pump will have a FLA of 14.4 amps +/-

Thanks for the breaker suggestion... to clarify, I've not seen any trips either at the pump 20A GFCI breaker (believe this to be Square D as everything else is, but strangely no identifying marks as with all other breakers and box) or the 60 non-GFCI main house panel breaker. But are you suggesting that the Siemens is better at filtering out noise that might affect other main panel 3-way breakers? Hayward did tend to like the Siemens QF220 as tested across a wide range of frequencies, though Schneider Electric (Square D) also claimed testing well into GHz range.

Other clarification is that the 60A 220v circuit is sized not just for 2 HP (~11A) Hayward pump, but also the heat pump and miscellaneous other things (Chlorinator, light, irrigation). (Several independent breakers at the pump panel).