GFI tripping on well pump used in a lake.

[itsmoked]

So I've had a question posed to me about a well pump used in a recreational lake. It's a single phase pump with the typical start/run package on shore. The circuit will not run reliably on a GFI but of course runs just fine with a standard breaker. Clearly it has leakage of more than 6mA.

I was asked if an isolation transformer would work. Probably it would work for preventing the GFI from tripping but I don't think that would help the soon to be electrocuted swimmers. That strikes me in this particular case as a tool to isolate only the GFI from reality.

The motor leads show 2Mohms to ground, seems low to me, but I'm told Franklin techs only consider it actionable at 500kohms or less.

The wire run is several hundred feet.

I'm not sure if code would allow a GFI with a higher or adjustable trip point in this app in NEC land.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

Have you considered capacitive leakage? 2 Meg's to ground is only 0.06 ma at 120 Volts. Double it for 240 Volts and at 0.12 ma still short of 6. ma. (I hope i haven't slipped a decimal)
However, "The wire run is several hundred feet."
How about wire with thicker insulation?

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[davidbeach]

Replace the THHN/THWN wire with XHHW wire and you may well solve your problem.

 

[jraef]

GFI tripping of motors is a well known phenomenon. Transformer isolation is the known cure, but is not 100%.

Interestingly, Sqare D Q0 GFI breakers don't nuisance trip. I had a heck of a time on my spa motor using a Siemens breaker, changed it to Sq D and it never nuisance tripped again. I thought it had once, turned out the insulation on the heater coil had failed. So it did its job when it needed to.


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[itsmoked]

Great suggestions!

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Compositepro]

I live near a large recreational lake. The Army Corp. of Engineers has required all the Marinas to post large signs banning swimming near the docks. This was due to a swimmer getting electrocuted on another lake a couple years ago. The boat owners are not happy.

 

[X49]

You may want to check the NEC. Canadian code allows ground fault current up to 10mA for up to 2.7 seconds in this application.

 

[itsmoked]

Thanks X49.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[LionelHutz]

An isolation transformer with a floating secondary does break the ground fault current path.

 

[waross]

The section of the Canadian Electrical Code dealing with isolation transformers in operating theatres requires the transformer to have a grounded shield between the primary and secondary windings. Note; this is not included in Part 1 of the code, it is a supplemental code.
Due diligence may require a similar transformer for life safety.
Better to reduce the capacitive leakage if possible.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[itsmoked]

I agree Bill. It would seem blinding the GFI from leakage won't save the swimmer.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[LionelHutz]

Where is this leakage path?

 

[waross]

With respect.
Yes, I see your point, Lionel and it is well made and I agree with it.
However, please consider the following points.
Thank you.
I would consider it due diligence to evaluate the possible implications of one line becoming grounded.
In the medical application that I mentioned, this possibility was addressed by a monitoring system. The medical systems must be small enough so that capacitive leakage to ground is within a very low acceptable range.
If one line becomes grounded, there will be a possible ground return path, and this path may pass twice as much current as before. The voltage to ground may be 240 Volts rather than the original 120 Volts.
With no ground reference at the transformer I doubt that a GFCI will see this possible ground fault path.
Now to the cost of the transformer we may need to add the cost of ground detection and tripping equipment. Also, for life safety, CYA may suggest a PE stamp on the drawings. This may or may not be a significant added cost.
I suggest a cost comparison between changing the cables and adding a transformer and detection and switching.
I would favour changing cables to reduce capacitive leakage.
As a reality check, it may be well to scope the circuit to verify that the leakage is, in fact, capacitive.
My first rule in trouble shooting is:
Make sure that you know what the source of the problem really is.
My third rule:
Re-check the first rule.
How many of us have seen this:
First find the problem and fix it.
Second, remove, repair or otherwise deal with the last three attempts by previous repairmen to "fix" the wrong problem.
Back to rule one.
Verify that we are seeing capacitive leakage.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[John2025]

FWIW, I ran a 400 ft, each way, 110 volt line with UF to run my driveway gate opener and 8 light poles along the way. I kept getting "nuisance" trips in thunderstorms. I made the "it must be lightning" excuse for a while, but finally realized the lights leaked enough water to trip the GFCI. Added gaskets where the manufacturer didn`t and never had another trip. I don't think I've ever seen what I would really call a nuisance trip.

 

[LionelHutz]

Geeze, people keep posting that the ground fault path would still kill the swimmer yet there is no path. Protecting against a failure of the isolation transformer does have to be considered.

The most likely cause is the pump has a small amount of moisture in the windings lowering it's resistance to ground. But, if you can't get the damn pump fixed then you have to start looking for other solutions.

I have no idea how the voltage and current of an isolation transformer could double...

 

[itsmoked]

Hutz; I'm not seeing how you can say "there is no path"? This could certainly be a cable capacitive issue, it likely is (no external path). So put a in a transformer and we'll stop having trips - any trips, even those caused by anything after the transformer, including bonified real leaks due to future conductor failures.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

With all respect, Lionel, there is no ground path because the transformer secondary is ungrounded. If there is a ground fault, there will be no ground fault current. Agreed. Almost.
However, there will be a capacitive current to ground from each of the ungrounded conductors.
Grounding either line will shunt the capacitance to ground and current will flow.
Further in the event of a second ground fault a ground fault current will flow. I will accept that this current may be low enough to be non-threatening on a small circuit but we can't assume that this circuit is small enough.

The wire run is several hundred feet.

Double?
Well with a grounded 120:240 Volt circuit the voltage driving a ground fault current will be 120 Volts.
However, if one line of a 240 Volt circuit goes to ground, then a second ground fault current will be driven by 240 Volts.
Isolation and safety is well addressed in the code for life safety isolated circuits in operating theatres. Not only must the circuit be isolated from ground but monitoring equipment must be installed to prove the absence of any leakage paths to ground.
The isolated circuit is great, but it must be isolated and there must be a system to prove that there is no inadvertent or unintended path to ground.
This is addressed here:
Excerpt from the Canadian Electrical Code; I am sure the NEC has a similar rule.
(2) Wiring systems supplied by an ungrounded supply shall be equipped with a suitable ground fault detection
device to indicate the presence of a ground fault.
(3) Ground fault indication activated by the ground fault detection device required by Subrule (2) shall be
clearly
(a) labelled as to its purpose; and
(b) visible to persons monitoring the status of the system.
And
10-112 Isolated circuits
Special circuits shall be permitted to be supplied from the ungrounded secondaries of transformers having the
primary and secondary windings separated by a grounded metal shield if
(a) installed under the provisions of other Sections of this Code; or
(b) this is required to recognize a particular accident or fire hazard.
A long cable will have capacitance to ground. If the cable is under water we must also consider the fairly high capacitive constant of water.
When I started out there were a lot of legacy ungrounded three phase systems in industrial plants.
The ground detection lights were primarily to avoid equipment damage in the event of a second ground fault on another phase.
The capacitive leakage current in many plants was well above lethal levels.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Compositepro]

I just saw this on the news:

http://www.msnewsnow.com/story/3198...utostart=true&clienttype=generic&sf26391519=1

 

[waross]

Thanks Compositepro. That kind of takes it out of the realm of theoretical discussions. We are talking about life safety.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[LionelHutz]

You're assuming the pump is 240V Bill.