Eaton 40 amp Double Pole GFCI Tripping

[BigTicket]

I have a 50 amp DP Eaton GFCI that has been in use for almost a year and a half. I have been using my welder on it. I recently mounted a 220V garage heater and it uses 31.25 amps. I swapped out the 50A breaker with a 40A Eaton BR GFCI because that is sufficient for my welder and will accommodate the heater.

The 40A breaker will trip sporadically. When you reset the breaker, it runs it self test and blinks a white light 4 times which is a high-level ground fault ( more than 30mA typically) or a electrical leakage to ground according to Eaton's info.

I have a NEMA 6p-50R receptacle. Two hots and a ground. I never once had this issue with the 50A breaker.

The only thing different is I moved the 40A up to the top of my panel to make room because I swapped out a single pole 30A 120v breaker for a 30A DP to accommodate a 220v tire machine.

I have double checked the wiring on the plug and the breaker. Everything is tight. Main panel so neutral and grounds are together.

I've read a few posts about nuisance tripping but thought I'd ask here.

Chris

 

[mike93lx]

Is the GFCI tripping or the breaker? They serve different purposes and have different reasons for tripping

 

[mm08822]

The heater should be on a 50A circuit with wiring that's rated for 50A. What is the wiring used.

What are the heater specs?

 

[BigTicket]

Is the GFCI tripping or the breaker? They serve different purposes and have different reasons for tripping

GFCI. It will trip with nothing plugged in.

 

[mike93lx]

GFCI. It will trip with nothing plugged in.

Then it is either bad or there is a problem between the receptacle and panel.
I'd start by removing the receptacle, wire nut the wires and inpsect the box. Turn the breaker on and see if it trips. Work back from there.

 

[BigTicket]

The heater should be on a 50A circuit with wiring that's rated for 50A. What is the wiring used.

What are the heater specs?

6Ga wire

 

[BigTicket]

Then it is either bad or there is a problem between the receptacle and panel.
I'd start by removing the receptacle, wire nut the wires and inpsect the box. Turn the breaker on and see if it trips. Work back from there.

I'll do that. Thanks.

 

[mm08822]

So it's 31.25 amps not the 37.1A you first reported. In that case, the 31.25 load could be on a 40A circuit minimum.

Exactly when is it tripping? what's plugged in and running when trips? Nm,, I saw post 4.

 

[BurtEggley]

is there a GCFI at the receptacle too, or in the welder plug? Two GCFI in the same circuit can trip each other even with nothing plugged in. Been there.

 

[BigTicket]

So it's 31.25 amps not the 37.1A you first reported. In that case, the 31.25 load could be on a 40A circuit minimum.

Exactly when is it tripping? what's plugged in and running when tr

is there a GCFI at the receptacle too, or in the welder plug? Two GCFI in the same circuit can trip each other even with nothing plugged in. Been there.

No. Just a 6-50. Thanks.

 

[wyliesdiesels]

Is the heater on this same circuit that is tripping? if not, then what is the point in mentioning the heater?

If all that is on that circuit that is tripping is the receptacle, then you will need to pull the receptacle and check the wiring. also try swapping with another breaker and see if it trips

 

[wyliesdiesels]

is there a GCFI at the receptacle too, or in the welder plug? Two GCFI in the same circuit can trip each other even with nothing plugged in. Been there.

This is incorrect. How could the 2 GFCIs even know they are connected to each other? there is no communication. All they care about is a hot to neutral load imbalance of more than 4-6ma....

The GFCIs themselves dont create a load imbalance so how on earth could they trip each other (they cant and this has been debunked before).

 

[dave*99]

I have a 50 amp DP Eaton GFCI that has been in use for almost a year and a half. I have been using my welder on it. I recently mounted a 220V garage heater and it uses 31.25 amps. I swapped out the 50A breaker with a 40A Eaton BR GFCI because that is sufficient for my welder and will accommodate the heater.

The 40A breaker will trip sporadically. When you reset the breaker, it runs it self test and blinks a white light 4 times which is a high-level ground fault ( more than 30mA typically) or a electrical leakage to ground according to Eaton's info.

Is the heater on this same circuit that is tripping? if not, then what is the point in mentioning the heater?

No. Just a 6-50. Thanks.

I'm hung up on the same question. OP - please confirm if the welder receptacle and the heater are on the same 40A GFCI breaker.

If I read this thread right, they are and the GFCI trips with nothing plugged into the 6-50R. If true, either the heater or your wiring is at fault. How did you tap/split connections to the 6 ga wire? Something nicked?

 

[BigTicket]

Is the heater on this same circuit that is tripping? if not, then what is the point in mentioning the heater?

If all that is on that circuit that is tripping is the receptacle, then you will need to pull the receptacle and check the wiring. also try swapping with another breaker and see if it trips

Is the heater on this same circuit that is tripping? if not, then what is the point in mentioning the heater?

If all that is on that circuit that is tripping is the receptacle, then you will need to pull the receptacle and check the wiring. also try swapping with another breaker and see if it trips

I suppose the mention of the heater was a little background info. It's the reason I swapped with the 50A breaker which never gave an issue.

 

[BigTicket]

I'm hung up on the same question. OP - please confirm if the welder receptacle and the heater are on the same 40A GFCI breaker.

If I read this thread right, they are and the GFCI trips with nothing plugged into the 6-50R. If true, either the heater or your wiring is at fault. How did you tap/split connections to the 6 ga wire? Something nicked?

It is a single circuit. I either plug the welder in or I plug the heater in. It's either or. I'm think the receptacle has an issue. Will test as soon as I can. Breaker is off until I figure it out.

 

[BurtEggley]

This is incorrect. How could the 2 GFCIs even know they are connected to each other? there is no communication. All they care about is a hot to neutral load imbalance of more than 4-6ma....

The GFCIs themselves dont create a load imbalance so how on earth could they trip each other (they cant and this has been debunked before).

because it happened to me and I have read it does happen when two GFCIs compete to protect the same circuit. Sometimes a GFCI will confuse another GFCI if they are on the same circuit. Specifically what happened to us is that we have a GFCI in the master bathroom. That circuit then goes 30' outside to a built in protected receptacle beside the back door. Common path in 1979 on homes being built here. I added a GFCI outside because I did not realize it was already protected by a GFCI 30' away in a bathroom. Randomly the GFCI outside the door would just trip itself with nothing plugged into it. I replaced it three times and checked the wiring. Randomly, the one in the bathroom started tripping itself about the time I added the one outside. Replaced it also several times and checked its wiring. When I made a spreadsheet of every receptacle, light, hard wired appliance etc., in the house after I retired and time for that sort of thing, I realized they were on the same circuit and removed the one outside since it was downstream of the bathroom. Since then, 8 years ago, nothing has tripped where one of them used to trip about every 45 - 60 days before. There are electrical articles out there explaining it. I came across them while researching the random tripping but it never occurred to me that I had two on the same circuit. Back in the 1990's I also put one in a bedroom because it fed a koi pool outside. I also put one in another bedroom on the same circuit to protect everything else. Both those would randomly trip. I removed the one to the Koi pond and had only the one. It stopped the random tripping. So I know it does happen from first hand experience, and the articles say it does happen.

 

[wyliesdiesels]

It is a single circuit. I either plug the welder in or I plug the heater in. It's either or. I'm think the receptacle has an issue. Will test as soon as I can. Breaker if off until I figure it out.

so the heater is cord and plug connected?

 

[wyliesdiesels]

because it happened to me and I have read it does happen when two GFCIs compete to protect the same circuit. Sometimes a GFCI will confuse another GFCI if they are on the same circuit.

compete? in what way? how could they confuse each other? there is no communication between GFCIs so that doesnt make any logical sense.

people have made this claim before and it has been debunked.

now what you may have experienced is an older GFCI tripping because it lost power as the other GFCI tripped. Some older GFCIs tripped upon loss of power. But that is not the same as what you claimed.

Specifically what happened to us is that we have a GFCI in the master bathroom. That circuit then goes 30' outside to a built in protected receptacle beside the back door. Common path in 1979 on homes being built here. I added a GFCI outside because I did not realize it was already protected by a GFCI 30' away in a bathroom. Randomly the GFCI outside the door would just trip itself with nothing plugged into it. I replaced it three times and checked the wiring.

how did you check the wiring without a meter? If you had no way to check for ground fault then you have no way of knowing if it was a ground fault causing the issue. youre just making assumptions here

Randomly, the one in the bathroom started tripping itself about the time I added the one outside. Replaced it also several times and checked its wiring. When I made a spreadsheet of every receptacle, light, hard wired appliance etc., in the house after I retired and time for that sort of thing, I realized they were on the same circuit and removed the one outside since it was downstream of the bathroom. Since then, 8 years ago, nothing has tripped where one of them used to trip about every 45 - 60 days before. There are electrical articles out there explaining it. I came across them while researching the random tripping but it never occurred to me that I had two on the same circuit.

cite the articles

Back in the 1990's I also put one in a bedroom because it fed a koi pool outside. I also put one in another bedroom on the same circuit to protect everything else. Both those would randomly trip. I removed the one to the Koi pond and had only the one. It stopped the random tripping. So I know it does happen from first hand experience, and the articles say it does happen.

again that doesnt mean the receptacle was causing it. your process of elimination and troubleshooting is poor. you didnt check for an actual ground fault so you have no clue what was causing it. youre making an assumption that somehow a GFCI would cause a load imbalance between hot and neutral causing the upstream GFCI to trip.

you do realize that if a GFCI caused a line imbalance that could cause the potential for shock, it would not be listed by an NRTL?

 

[Cruzan80]

because it happened to me and I have read it does happen

Sorry, but no.

 

[PCustoms]

Not to detract from the OP (which was clear to me, he swapped a breaker, had issues, even with everything unplugged), but I've personally seen GFCI do some weird ****. It's annoying when people jump and and swear that it's not possible.

Some of you are very competent electricians that Garner lots of respect here, but I suspect you aren't electrical engineers that can explain the specific internal mechanisms and possible interference within a GFCI.

 

[dave*99]

I asked ChatGPT to find articles about stacking GFCIs. It reported it’s not recommended. It said they don’t actually interfere with each other but it makes troubleshooting difficult. Slight differences in sensitivity cause one to trip before the other, etc.

I asked for an article on the topic from a GFCI manufacturer. The Leviton guidance said the same stuff about difficult troubleshooting. It said nothing about interference.

Here is the funny part: ChatGPT cited Garage Journal as the place to look for claims of interference. Go figure.

YMMV.

 

[BigTicket]

so the heater is cord and plug connected?

Correct

 

[wyliesdiesels]

Not to detract from the OP (which was clear to me, he swapped a breaker, had issues, even with everything unplugged), but I've personally seen GFCI do some weird ****. It's annoying when people jump and and swear that it's not possible.

Some of you are very competent electricians that Garner lots of respect here, but I suspect you aren't electrical engineers that can explain the specific internal mechanisms and possible interference within a GFCI.

I dont need to be an EE to know that there is no communication or radio waves or harmonics generated by a GFCI to cause interference with another GFCI on the line.

there is no such thing with a GFCI. A GFCI is actually really simple.

but people think its a magical black box that has some kind of magic voodoo in it

 

[BurtEggley]

Sorry, but no.

no trolling allowed. You are calling another member a liar hoping to start a fight.

Wylie - you can think whatever you want but I am telling you what happened. Hook some up in your house that way and report back in a year. Make sure there are a couple lights on the same circuit, preferably something with a transformer in it like a florescent. Run something like a hair dryer that has a small motor on it occasionally on the circuit.

 

[wyliesdiesels]

no trolling allowed. You are calling another member a liar hoping to start a fight.

Wylie - you can think whatever you want but I am telling you what happened. Hook some up in your house that way and report back in a year. Make sure there are a couple lights on the same circuit, preferably something with a transformer in it like a florescent. Run something like a hair dryer that has a small motor on it occasionally on the circuit.

what you dont know is i have lived in a house with that setup and never had an issue.... i know how GFCIs work and I know what circuitry is in them. there is no black magic voodoo inside them that would allow them to communicate with each other....

 

[Cruzan80]

no trolling allowed. You are calling another member a liar hoping to start a fight.

Wylie - you can think whatever you want but I am telling you what happened. Hook some up in your house that way and report back in a year. Make sure there are a couple lights on the same circuit, preferably something with a transformer in it like a florescent. Run something like a hair dryer that has a small motor on it occasionally on the circuit.

Not trolling (was late and tired, meant to edit to a fuller post), but rather pointing out that simply because you read something, especially on the internet, doesnt mean it is true. As has been stated, there is no way for them to "interfere", especially without a load on them. Anecdotal experience does not indicate proof.

While not an EE specifically, I do teach HS Engineering, including circuit design and basic residential electrical.

 

[Cruzan80]

If you are convinced that there is an issue, you can always run a pigtail inside the box at the first GFCI, so that you only utilize the LINE side for that GFCI. Then install the second GFCI normally. Since there electrical path doesnt go thru the first one (nothing on LOAD side), so there isless/no chance of a (theoretical) interference. Keep in mind, if you set it up like this, any other outlets on the circuit are not protected unless on the LOAD side of a GFCI (either one).

 

[mm08822]

Not to detract from the OP (which was clear to me, he swapped a breaker, had issues, even with everything unplugged), but I've personally seen GFCI do some weird ****. It's annoying when people jump and and swear that it's not possible.

Some of you are very competent electricians that Garner lots of respect here, but I suspect you aren't electrical engineers that can explain the specific internal mechanisms and possible interference within a GFCI.

I am a EE but I can tell you no mfr would release their board level circuits, you can forget that idea.
That info is not needed anyway.

• Every production lot of gfci cb's and recepts has lot #'s. Within each of those lots, there are sub-lots of components, where each sub-lot has variation within the lot and from lot to lot.
• Next comes age of the device and the temp and humidity the device is subjected to.
• 4-6mA is stated b/c no consumer could afford the cost of precision components yielding a device tripping at exactly 6ma. The 4-6 mA window is there to accommodate the overall component tolerance impact within the final product.
• The 4mA limit is for practical purposes. All protected wiring downstream of the gfci device is subject to leakage current. This will vary based on insulation quality and humidity. A mfr can't predict how much wire will be hooked up to it. The longer the length, more leakage possibility. (I remember 1 mfr(maybe more) stated a do not exceed wire length, but I haven't read directions in 30 yrs.) There is also various humidity exposure levels that can change throughout the year affecting leakage.

So, taking all of that into account, it should be easy to understand how two devices could race each other into trip conditions. Usually no one is thinking of all the variables between devices, performance levels shifting over time, field wiring, etc. These trip windows and more specifically, the trip value move relative to each other.
This creates 3 scenarios for unpredictable tripping between GFCI Recept A & GFCI Recept B (B downstream of A & protected by A):
When a fault occurs downstream of B:

• A Trip point > B Trip point.........> B trips
• A Trip point = B Trip point.........> Both could trip (but highly unlikely to be the same)
• A Trip point < B Trip point.........> A trips

So this is where people get pissy, b/c it is "random"/unpredictable.

Pressing the test button on GFCI Recept B could cause GFCI Recept A to trip before B did, both trip, or only B trips.

GFCI devices are dumb. They only sense an imbalance within their 4 respective terminals. They don't know or care about what the external devices are connected to it.

Having only 1 gfci providing protection eliminates the race condition and we go back to a simpler set of expectations.....the gfci tripped.

 

[dave*99]

I found this article interesting. Especially the section on sensing neutral to ground current.

Teardown: Leviton ground fault circuit interrupter - Power Electronic Tips

www.powerelectronictips.com

"There is a second toroid used as a current sensor in the GFCI. Judging by comments we’ve seen on forums, there is some confusion about its purpose and how it works.

The point of the second toroid is to sense current flowing from ground to neutral, rather than a difference in the line and neutral current. One point of confusion among some online commenters is how current can flow in the ground lead without causing a difference in the hot/neutral current that the first toroid would sense. The answer: One can envision a scenario in which current flows from hot to load, then to neutral, and then to ground. There would potentially be no difference in the line and neutral current. So the GFCI circuit must sense this fault current with the operating assumption that the hot and neutral currents are the same."

 

[mm08822]

I found this article interesting. Especially the section on sensing neutral to ground current.

Teardown: Leviton ground fault circuit interrupter - Power Electronic Tips

www.powerelectronictips.com

I didn't read it yet, but the circuit shows a toroid coil with only a hot and neutral through it to drive the imbalance sensing. GFCI's don't need a grounding conductor to trip, typical situation for old wiring.

 

[dave*99]

I didn't read it yet, but the circuit shows a toroid coil with only a hot and neutral through it to drive the imbalance sensing. GFCI's don't need a grounding conductor to trip, typical situation for old wiring.

Which circuit are you referring to? The article describes a second sense coil and it's pictured in the tear down.

 

[mm08822]

Which circuit are you referring to? The article describes a second sense coil and it's pictured in the tear down.

You're correct, 2 coils, but neither measure an equipment grounding path current. They both are only measuring the same hot and neutral currents. 2 terminals in / 2 terminals out. (Neutral = grounded conductor.)

 

[dave*99]

You're correct, 2 coils, but neither measure an equipment grounding path current. They both are only measuring the same hot and neutral currents. 2 terminals in / 2 terminals out. (Neutral = grounded conductor.)

I agree with your assessment of the coil locations in the circuit. Still, I encourage you to read the article as it describes the approach to sense neutral to ground current.

 

[mm08822]

I agree with your assessment of the coil locations in the circuit. Still, I encourage you to read the article as it describes the approach to sense neutral to ground current.

I will.

Here is the Fairchild IC cutsheet, if interested. A quick scan shows some of your article provided is directly from this cut sheet. (Which is good!)

 

[dave*99]

I will.

Here is the Fairchild IC cutsheet, if interested. A quick scan shows some of your article provided is directly from this cut sheet. (Which is good!)

That's where I started. Manufacturers of the IC's often provide reference designs and app notes etc. Back in the day I had a shelf full of those books in my lab at work.

 

[mm08822]

741's were the ubiquitous choice for beginners. So many configs. Really cool back in the day!

 

[Cruzan80]

Two GCFI in the same circuit can trip each other even with nothing plugged in.

This creates 3 scenarios for unpredictable tripping between GFCI Recept A & GFCI Recept B (B downstream of A & protected by A):
When a fault occurs downstream of B:

• A Trip point > B Trip point.........> B trips
• A Trip point = B Trip point.........> Both could trip (but highly unlikely to be the same)
• A Trip point < B Trip point.........> A trips

So this is where people get pissy, b/c it is "random"/unpredictable.

Bolding of quotes was added for my emphasis.

The part I was responding to was a fourth scenario. With nothing plugged in, the "phantom" tripping isn't generally ascribed to an actual fault, but rather "interference" between GFCI's (especially with nothing plugged in). Without an imbalanced load, nothing should trip, regardless of the number of GFCI on the circuit.

 

[mm08822]

As far as the gfci design is concerned, leakage is a fault........leakage in conductors and connected load(s). Values are not necessarily constant throughout the year. Every load does not have to be running to provide a leakage path. Multiple loads can be plugged in downstream.

 

[Cruzan80]

Agreed, but would you also agree that if a circuit has only one GFCI and it isn't tripping normally, that if the only thing that changes is swapping a downstream GFCI into a regular outlet, they shouldn't start "magically interfering" and tripping each other? Same loads, same items plugged in (in the "hypothetical" quoted, none) etc. That is what my main point was about.

I agree that "if" there is a fault scenario, the lack of repeatability on which GFCI trips would be frustrating (the part I quoted from you above).

 

[mm08822]

Agreed, but would you also agree that if a circuit has only one GFCI and it isn't tripping normally, that if the only thing that changes is swapping a downstream GFCI into a regular outlet, they shouldn't start "magically interfering" and tripping each other? Same loads, same items plugged in (in the "hypothetical" quoted, none) etc. That is what my main point was about.

I agree that "if" there is a fault scenario, the lack of repeatability on which GFCI trips would be frustrating (the part I quoted from you above).

IF there were any "interaction" between the 2 gfci recepts, only B could trip A. A would never trip B. (B downstream of A.)

Leakage current (the invisible rascal) is additive at any given time, although not necessarily constant throughout the year.