GFI tripping on low voltage??

[Cavie]

Yes of course... I am still confused on what Cavie is getting at.

I'll get more info on Monday. This was asked of me by one of my field supervisors late in the day and I didn't get any details. I could not answer his question.

 

[curt swartz]

GFCI receptacles will trip if voltage is removed. this has been an OSHA requirement in 'construction cords' for years, but it is only a recent requirement in 'normal' box mounted devices (a primary reason homemade extension cords were discouraged).

Jim, I have never seen a GFCI receptacle that will trip when the voltage is removed. They will not reset if tripped with out voltage but they don't automatically trip. We frequently install exterior GFCI receptacles that are controlled by switches.

 

[Dennis Alwon]

Jim, I have never seen a GFCI receptacle that will trip when the voltage is removed. They will not reset if tripped with out voltage but they don't automatically trip. We frequently install exterior GFCI receptacles that are controlled by switches.

The ones I have seen were part of a cord and plug extension- about 2' long. If you unplug it or if the circuit went off then you had to reset the gfci. They were a PITA.

 

[jumper]

The ones I have seen were part of a cord and plug extension- about 2' long. If you unplug it or if the circuit went off then you had to reset the gfci. They were a PITA.

Yep, that is a requirement for GFCI cords as Jim D stated.

http://www.osha.gov/SLTC/etools/construction/electrical_incidents/gfci.html

 

[jim dungar]

Jim, I have never seen a GFCI receptacle that will trip when the voltage is removed. They will not reset if tripped with out voltage but they don't automatically trip.

Thanks, for the clarification. The GFCI devices on 'OSHA approved' extension cords are not 'receptacles'.

 

[kwired]

I believe the reason for tripping when losing power for these "portable" devices is to prove there is both circuit conductors present. Conventional GFCI devices will not necessarily trip if not seeing nominal voltage across the trip unit. Break one wire in a cordset and you potentially have lost any protection but still may have hot conductors in the cordset.

Newer generation GFCI receptacles will not reset without full input voltage, but will not trip when input voltage is lost. If they do trip at loss of voltage, there are other complications causing the trip, like maybe inductive kickback.

 

[Cavie]

more info

more info

This is a light on a dock powered by GFI breaker. Works fine on 3 way switching. A 4 way was introduced into the system adding footage to about 450'. the issue seems to be the added footage in the system.

 

[Robatt]

GFI TRIPPING ON LOW VOLTAGE

GFI TRIPPING ON LOW VOLTAGE

Firstly, a long cable run with inadequate compensating surface area to lower resistance will definitely result into a voltage drop in excess of allowable norm. If the Power source is constant then an increase in resistance will drop the voltage and increase the current. Increased current beyond the rated current of the GFI will trip the GFI if it has an over current protection only. Without an OCPD, it will not trip.

 

[Dennis Alwon]

Firstly, a long cable run with inadequate compensating surface area to lower resistance will definitely result into a voltage drop in excess of allowable norm. If the Power source is constant then an increase in resistance will drop the voltage and increase the current. Increased current beyond the rated current of the GFI will trip the GFI if it has an over current protection only. Without an OCPD, it will not trip.

I don't believe I have ever seen a gfi that would trip on overcurrent. It may melt the unit or the electronic but it will not trip the unit, afaik

 

[kwired]

Firstly, a long cable run with inadequate compensating surface area to lower resistance will definitely result into a voltage drop in excess of allowable norm. If the Power source is constant then an increase in resistance will drop the voltage and increase the current. Increased current beyond the rated current of the GFI will trip the GFI if it has an over current protection only. Without an OCPD, it will not trip.

GFCI's do not respond to overcurrent, they only monitor for imbalanced current in the monitored circuit conductors. What goes out must come back on a monitored conductor, any imbalance of more than the 4-6 mA range will trip the unit.

If a 20 amp GFCI is carrying 50 amps the GFCI will not care as long as 50 goes out and 50 comes back.

 

[ELA]

Op said gfi breaker. Would that be a Gfci breaker?
If so do they not have over current protection?
Not that it really matters in this case.
Too much capacitance due to long cable length and moisture , leakage currents likely the cause.

 

[kwired]

Op said gfi breaker. Would that be a Gfci breaker?
If so do they not have over current protection?
Not that it really matters in this case.
Too much capacitance due to long cable length and moisture , leakage currents likely the cause.

OK, GFCI breaker will have overcurrent protection but the GFCI component of it will not respond to overcurrent conditions.

 

[hillbilly1]

This is a light on a dock powered by GFI breaker. Works fine on 3 way switching. A 4 way was introduced into the system adding footage to about 450'. the issue seems to be the added footage in the system.

Yeah, it's the distance, I don't know if its capacitance, or total leakage due to length of wire, but it's not voltage drop. I've seen this issue with runs that have been properly adjusted in size for voltage drop. A lot of small leakages in insulation add up on that long of distance.

 

[kwired]

Yeah, it's the distance, I don't know if its capacitance, or total leakage due to length of wire, but it's not voltage drop. I've seen this issue with runs that have been properly adjusted in size for voltage drop. A lot of small leakages in insulation add up on that long of distance.

Capacitance between the protected conductors should not make any difference, capacitance between a protected conductor and anything else will result in some current leakage to that "anything else". Go beyond 4-6 mA of leakage combined with any other leakage and you will trip the GFCI

 

[hillbilly1]

Capacitance between the protected conductors should not make any difference, capacitance between a protected conductor and anything else will result in some current leakage to that "anything else". Go beyond 4-6 mA of leakage combined with any other leakage and you will trip the GFCI

Capacitance to ground is what I'm talking about, believe it or not there is capacitance between the conductors and grounded (bonded) surfaces around it. The longer this exposure, the more capacitance.

 

[kwired]

Capacitance to ground is what I'm talking about, believe it or not there is capacitance between the conductors and grounded (bonded) surfaces around it. The longer this exposure, the more capacitance.

Location of the GFCI in relation to the long run makes a difference also. If you have a really long run you are better off having a GFCI receptacle at the load end instead of a GFCI breaker at the supply end.

 

[hillbilly1]

Location of the GFCI in relation to the long run makes a difference also. If you have a really long run you are better off having a GFCI receptacle at the load end instead of a GFCI breaker at the supply end.

But we are talking about a gfi breaker here, since it is a boat dock, and it is a switch leg circuit, the OP does not have the receptacle option, unless......he changes the circuit to non gfi, puts a relay in a box on shore, then feeds a switched gfci circuit from that point on.

 

[kwired]

But we are talking about a gfi breaker here, since it is a boat dock, and it is a switch leg circuit, the OP does not have the receptacle option, unless......he changes the circuit to non gfi, puts a relay in a box on shore, then feeds a switched gfci circuit from that point on.

Or some alternate low voltage, wireless, or power line carrier signaling type of switching instead of adding the long line voltage switch loop to the GFCI protected circuit.

 

[hurk27]

GFCI's will work just fine with voltage drops, whither it is a breaker or receptacle version, the design of the circuit in them consist of a bridge rectifier feeding into a resistor voltage divider network with a ziner diode to maintain regulation of 26 volts, but the electronics will still function between 22 to 30 volts, the voltage drop would have to fall below 22 volts before there was any kind of problem.

I have worked on Tech power isolating supply's for server rooms that have GFCI breakers and with only 60 volts to the neutral buss, 120 volts line to line the power to the electronics in a two pole GFCI breaker is always line to the neutral and they work just fine at 60 volts.

Long runs on the load side of a GFCI can be a problem as was said from the capacitance and or any leakage in the circuit, and or the type of lighting use such as HID type that might exhibit more leakage then incandescent.

Here is a link to the most common chip used in most GFCI products:

National Simiconductor LM1851 GFCI IC

 

[ELA]

Hurk27,

I suspect that your comments that " GFCI's will work just fine with voltage drops" and "the voltage drop would have to fall below 22 volts before there was any kind of problem" are total speculation.
Define "work just fine". While I would not expect a false trip on low line values I also would not expect the device to "work just fine" at any specific low line value without direct test results vs speculation.

I have studied the GFCI in some depth and I have reviewed the LM1851data sheets many times over many years.
If you look a little closer you will see that all specifications are at Iss=5ma & the shunt regulator can be as high as 30V..

Since they do not state a minimum Iss you cannot guarantee proper operation below that value.