GFCI Breaker on deep well submersible

[2stupid2fixit]

I installed a hallmark 2 wire to replace my dead goulds/frankin, and replaced almost every part of the system, like tank, pressure switch, wire that goes from the pump up to the ground, safety rope etc etc.

One day I got the bright idea to install a squareD 20 amp 2 pole GFCI breaker, ya know, to be safe. The system is already grounded and bonded as well as any could be, but I decided "why not add a helmet protector protector protector". Whole thing worked well until today, adding power instantly trips the GFI.
So I took the feed off of the breaker and rang out the resistance on the feed wires going to the pump, and the reading is around the same if not lower than when I installed it. I expected to see something sticking out like a sore thumb but I didnt get that. So, maybe something internally changed within the hallmark pump. Beats me. I took out the gfi breaker and put the conventional breaker back and it works just fine.

Most of the pump guys say no way to GFI breaker on deep well pumps... So does code require this component that doesnt protect against much as installed? Maybe I read the NEC wrong. I already have more good ground points for this setup than I do for any other appliance.

 

[Bannerman]

One saying is: 'If it ain't broken, don't fix it'

Another: 'Fix it till it's broken'

It could be some water leaked into one of the seals surrounding the submerged external wiring connections which join the long drop cable to the pump wiring, resulting a slight ground fault condition that may not have been an issue or detected without a GFI.

Have you attempted to utilize the GFI to supply power to anything else to see if it will continue to trip? Perhaps the GFI is the problem?

 

[Sponsor]

 

[2stupid2fixit]

One saying is: 'If it ain't broken, don't fix it'

Another: 'Fix it till it's broken'

Government said the second one.

 

[Reach4]

So I took the feed off of the breaker and rang out the resistance on the feed wires going to the pump, and the reading is around the same if not lower than when I installed it.

The resistance that mattered for this is from either hot to ground.

 

[WorthFlorida]

GFCI is designed to protect people, not appliances. Installing a surge protector on the circuit will help protect the appliance. Is this a 120v or 220v motor?

GFCI monitors the current of the line and neutral sides are equal. Motors need a start and run capacitor and that right there might make the difference and the GFCI will trip.

Last year I installed a Square D surge protector in my main panel.
https://www.lowes.com/pd/Square-D-QO-50-KA-Indoor-Surge-Protective-Device/1000642897

https://terrylove.com/forums/index....and-why-you-might-want-one.94985/#post-683646

 

[2stupid2fixit]

GFCI is designed to protect people, not appliances. Installing a surge protector on the circuit will help protect the appliance. Is this a 120v or 220v motor?

It's 220. It was a 2 pole 20 amp GFCI breaker, and the white piggly tailed wire connected to the bus bar in the panel that all the other white wires are connected. Other forums tell me not to call the white "neutral" but thats another discussion. I am not concerned with protecting the appliance from surges. I was just trying to follow code since I had replaced everything else. Maybe I misread or misunderstood the code. The goulds pump that was there from the 1970s had 2 wires, 10 gauge going right from a 2 pole breaker to the pump. The connections on the long drop cable were soldered and wrapped with lots of plain old black electrical tape. No ground. 2 wires, that's all. It lived and worked for about 2o years that way. Yes, I know, something that works doesn't mean it is in any way safe.

The hallmark pump i used for replacement is also a 2 wire, but there is a ground coming from the pump. It is impractical but I guess not impossible for me to dig up the wire going back to the house and replace it with a 2 wire plus ground cable. So I just screwed the ground that came off the pump motor to the well casing, which is metal and goes at least 80 feet down before it stays metal or turns into onlygodknows. I figured that makes a pretty decent ground because the well case is essentially a ground rod. Right?

When I made my connections from the pump motor to drop wire, I used those hot-melt clear shrink tubes and heated them until I saw the plastic melt and seal the connections. Then I used 3m self-fusing tape, stretched it really nice when I was wrapping it. IF my connections allowed water in, I would be VERY surprised. However, if the under $200 Hallmark pump let water in somewhere it should not have, I would NOT be surprised at ALL.

I guess herein lies my real question: I got everything working by putting the conventional 2 pole breaker back in, instead of the 2 pole GFCI breaker.

Let's suppose the pump lets water into the electrical connections but still works. Without the GFCI that I recognize is there to protect people, what happens next? I'm not an electrician. Does the 20 amps of 240 volts sent down that well energize the water all the way back to the house and
kill someone in the shower? Or since it is 200 feet down in the water and 100 feet away from the house, does the killing potency of the electricity find somewhere else to go before making it back to the shower?

I doubt there is anything at all wrong with my GFCI breaker or how I installed it. I believe it worked properly and tripped because something wasn't quite right, It did its job. The GFCI breaker was installed for over 4 months without incident, and on Saturday, it was an instant trip as soon as power was applied. I played the role of human guinea pig and tested the install myself when I swapped out the GFCI breaker for the conventional and the pump worked as expected.
I took a shower and was not killed, but is it possible that water infiltration to where it shouldn't be was minuscule, but could become much more severe and then become lethal in the house?

I know the correct answer in the perfect world is to get the pump to run properly when connected to the GFCI breaker.

What do you think?

 

[Reach4]

The hallmark pump i used for replacement is also a 2 wire, but there is a ground coming from the pump. It is impractical but I guess not impossible for me to dig up the wire going back to the house and replace it with a 2 wire plus ground cable. So I just screwed the ground that came off the pump motor to the well casing, which is metal and goes at least 80 feet down before it stays metal or turns into onlygodknows. I figured that makes a pretty decent ground because the well case is essentially a ground rod. Right?

What you describe should not have caused the GFCI breaker to trip.

Water infiltrating into a splice could trip the gfci.

I took a shower and was not killed, but is it possible that water infiltration to where it shouldn't be was minuscule, but could become much more severe and then become lethal in the house?

I don't see how.

 

[Bannerman]

While the metal well casing will act as an effective grounding rod, it may provide a better or worse path to ground compared to the grounding rod the main electrical panel is tied to.

Because there is no grounding conductor between the pump and the electrical panel, the ground for the pump is not utilizing the same grounding reference as the panel-mounted GFI. Electrical codes commonly require all grounds to be tied back to the main electrical panel so that all circuits and subpanels will be grounded using an identical reference point.

The GFI may be sensing a ground fault condition because the pump ground is not utilizing the same grounding reference.

 

[2stupid2fixit]

the GFCI breaker detected something because it worked and there were no issues for almost 6 months (I just checked my records). Suddenly on Saturday the GFI tripped and would not stay reset the second the pressure switch kicked on.

 

[Reach4]

If you are curious, check ohms to ground after isolating the wires from power. I expect you have less than 25 kΩ, and normal standard of a well person would be greater 500 kΩ measured with a megger (high voltage ohmmeter). I would not worry about 100 kΩ. In your case, I would probably keep using the non-GFCI breaker and only have a minor worry that things could be deteriorating.

I would not worry about getting shocked in the shower etc due to leakage.

 

[WorthFlorida]

I would not worry about getting shocked in the shower etc due to leakage.

"Your honor, I told the wife not to take a shower!"

 

[2stupid2fixit]

"Your honor, I told the wife not to take a shower!"

Funny you say that. I did have the bright idea to install a convenience outlet (20 amp conventional, of course) and small ledge of a shelf in the shower to hold a pop up toaster so that my now-ex would save time and have the added convenience of making toast in the shower. I never did get around to that project. Don't worry, she is still alive and well, and ruining someone else's life instead of mine. There were no freak bathroom toaster accidents.

 

[wwhitney]

How long is the circuit from breaker to the pump? GFCIs impose a limit on the circuit length, because of the capacitive coupling of the circuit conductors. Of course, this couldn't explain why the behavior changed recently.

Connecting the pump down cable EGC to the well casing probably isn't bad, but it doesn't do much by itself. While the EGC is called a ground and is connected to earth, its primary purpose is to be connected back to the service neutral conductor at the service entrance (the only place neutral and ground get connected). That way if you have a fault from hot to EGC, the circuit is completed and enough current flows to trip the breaker. At 240V, when earth is part of the current path, the resistance will be too high to trip a breaker on overload (but it can be enough to trip a GFCI).

Running a separate appropriately large grounding electrode conductor (GEC) from the service to the metal well casing would improve your earthing, and then it could serve as an EGC for the pump circuit (since for existing ungrounded circuits, one of the allowable methods for adding an EGC to the circuit is to connect to any GEC.)

Cheers, Wayne

 

[2stupid2fixit]

If you are curious, check ohms to ground after isolating the wires from power. I expect you have less than 25 kΩ, and normal standard of a well person would be greater 500 kΩ measured with a megger (high voltage ohmmeter). I would not worry about 100 kΩ. In your case, I would probably keep using the non-GFCI breaker and only have a minor worry that things could be deteriorating.

I would not worry about getting shocked in the shower etc due to leakage.

One leg:

 

[2stupid2fixit]

One leg:
View attachment 75163

The other leg:

 

[wwhitney]

So how about that length question? What brand and model of breaker?

Cheers, Wayne

 

[2stupid2fixit]

So how about that length question? What brand and model of breaker?

Cheers, Wayne

That was measured 218 feet from the pump motor on #10 stranded wire.
It's Square D HOM220GFI.
Connected to underground cable back to the house, that adds another 175 feet of #10 solid.

 

[wwhitney]

From the Homeline GFCI Instruction Sheet

"Do not connect circuit breaker to more than 250 ft. (76 m) of load conductor for the total one-way run."

So if you really want your pump GFCI protected, you'd need to install the GFCI breaker at the well head. Then if it tripped, it's arguably a defective pump, although the manufacturer might disagree with you on that. I'm not sure what the allowable ground leakage standard is for submersible pumps.

Cheers, Wayne

 

[Valveman]

I have never had any luck getting a well pump to work with a GFCI. And is all my years I have never heard of anyone getting shocked by a shorted well pump.

 

[2stupid2fixit]

And is all my years I have never heard of anyone getting shocked by a shorted well pump.

Then that settles it. The setup had a conventional breaker for 40 years and no one was lit up, so that's how it will stay.