GFCI INTERMITTENT TRIP

[Jpflex]

The break room where this 15a 125 volt GFCI is installed is first in circuit and supplies 3 other 15a receptacles from its load side.

However, nothing is plugged into the other receptacles or has been used yet. The GFCI apparently intermittently trips as it’s plugged into both a refrigerator and microwave

The refrigerator pulls only 0.10 ampere running and the microwave 13 amperes however the GFCI is rated at 15 amperes on a 20 ampere circuit (yellow romex nm cable as NEC allows for multiple receptacles per circuit.)

However, running over 80% of GFCI ampacity rating is my concern whether a GFCI will function also as an inverse time breaker and trip for this reason BESIDES normal trip for imbalance of current returning on neutral.

Additionally, although ground from NM cable is connected to GFCI it is possible ground does not continue to load center panel because this old property has a combination of old knob and tube with modern Romex

GFCI led stays green in my presence. Trying to find trip fault. Thanks

 

[Jpflex]

Amperes

 

[hillbilly1]

The receptacle contains no overcurrent protection, the only thing it monitors is current not returning on the same two wires feeding it. An equipment ground is not required for it to work. A fault can happen on either appliance, so divide and conquer. Use a different circuit for one of the appliances, if either trips the corresponding gfi receptacles, then you found your culprit. Also if the equipment ground is compromised, a simple voltage check can verify. If the equipment ground is open, check for voltage between it and the neutral. If voltage is present, unplug one or the other appliance to see if it goes away. If it does, then that appliance is causing the tripping.

 

[Jpflex]

Update line hot from GFCI to ground is 120 volts and line to neutral is 120 volts so it appears ground or equipment ground conductor at GFCI has continuity to panel

Also found GFCI to trip at only 0.7 amperes with ampere clamp on neutral wire. Will try to get two meteres one on line and one on neutral to compare amperes during shut down.

No amperes found in EGC yet

 

[letgomywago]

The amp clamp might be causing an imbalance. 15 amp gfi and 20 amp gfi are both rated for 20 amps feed through by UL and this is specifically allowed in the NEC you just can't put a 20 amp circuit to a dedicated simplex 15 amp. A duplex is OK though.

 

[letgomywago]

Microwaves trip gfcis often as do ice makers. Is the fridge outlet more than 6ft from the sink and not in a location requiring gfci?

There is the possibility that both appliances have 3 miliamps of leakage current so when both on would trip but putting line side gfci for all the counters would separate the potential leakage.

 

[Jpflex]

The microwave is not tripping the GFCI but it is drawling the higher current obviously.

Removing disconnecting the microwave from the GFCI and the GFCI still trips. Specifically it does this each time the refrigerator compressor kicks on. (I expected to see some current on EGC if there was a current leakage but have not so far)

The GFCI is not more than 6 feet from refrigerator nor factory cord longer than this.

I also followed the circuit and the branch circuit goes all the way to panel without any other box to connect any other receptacles or loads.

Looks like refrigerator is issue so I’m disconnecting it with another load attached followed by monitoring green LED in GFCI tomorrow morning.

I wanted to compare line and neutral current values during GFCI trip but it happens so fast and current difference may be in milliamperes difference to be able to tell

 

[hillbilly1]

The microwave is not tripping the GFCI but it is drawling the higher current obviously.

Removing disconnecting the microwave from the GFCI and the GFCI still trips. Specifically it does this each time the refrigerator compressor kicks on. (I expected to see some current on EGC if there was a current leakage but have not so far)

The GFCI is not more than 6 feet from refrigerator nor factory cord longer than this.

I also followed the circuit and the branch circuit goes all the way to panel without any other box to connect any other receptacles or loads.

Looks like refrigerator is issue so I’m disconnecting it with another load attached followed by monitoring green LED in GFCI tomorrow morning.

I wanted to compare line and neutral current values during GFCI trip but it happens so fast and current difference may be in milliamperes difference to be able to tell

Probably worn or cheap compressor leaking the current, especially if it is tripping on start up.

 

[Little Bill]

Leave the microwave and put the refrigerator on a non-GFCI circuit.

 

[synchro]

Removing disconnecting the microwave from the GFCI and the GFCI still trips. Specifically it does this each time the refrigerator compressor kicks on. (I expected to see some current on EGC if there was a current leakage but have not so far)

From these symptoms, it's very likely the refrigerator compressor is driven by a VFD. These types of refrigerators are getting more common. The VFD can have higher frequency common-mode currents that can trip GFCI's.

I wanted to compare line and neutral current values during GFCI trip but it happens so fast and current difference may be in milliamperes difference to be able to tell

The short duration can be difficult to measure as you say. If you really wanted to compare the line and neutral currents you should put the clamp around both the line and neutral conductors so that you're measuring the difference in currents directly. But most clamp meters don't have enough sensitiivity and resolution to accurately measure in the 0-5mA range, even with a steady current level.

 

[kwired]

Is the refrigerator self defrosting? Faulted defrost heater can trip a GFCI, but may only do so when there is a call for defrost, and would very possibly make it seem pretty random when it does trip.

 

[letgomywago]

The microwave is not tripping the GFCI but it is drawling the higher current obviously.

Removing disconnecting the microwave from the GFCI and the GFCI still trips. Specifically it does this each time the refrigerator compressor kicks on. (I expected to see some current on EGC if there was a current leakage but have not so far)

The GFCI is not more than 6 feet from refrigerator nor factory cord longer than this.

I also followed the circuit and the branch circuit goes all the way to panel without any other box to connect any other receptacles or loads.

Looks like refrigerator is issue so I’m disconnecting it with another load attached followed by monitoring green LED in GFCI tomorrow morning.

I wanted to compare line and neutral current values during GFCI trip but it happens so fast and current difference may be in milliamperes difference to be able to tell

Location of the gfci in relation to the fridge has nothing to do with is the fridge 6 ft or more from sink and is this a location needing gfci

 

[ptonsparky]

You need some pretty good meters to catch the few milliamp that trips a GFCI.
A Fluke 87 or equivalent.

 

[Jraef]

Aside from the tripping only when the defrost cycle comes on (very common problem with older fridges and GFCIs), the other concern is this:

…because this old property has a combination of old knob and tube with modern Romex

Old K&T systems were often wired as what we would now call an MWBC, where the neutral was common to multiple circuits. That can cause GFCIs to trip if it is monitoring the neutral of one circuit, but current is returning via another path.

 

[ELA]

Below is what I always recommend for intermittent GFCI trips. Of course it is hard to justify the cost for occasional use.​

Assuming it is a true current leakage and not an EMC issue.​

Fluke 369 True-rms Leakage Current Clamp Meter​

 

[Jpflex]

Aside from the tripping only when the defrost cycle comes on (very common problem with older fridges and GFCIs), the other concern is this:


Old K&T systems were often wired as what we would now call an MWBC, where the neutral was common to multiple circuits. That can cause GFCIs to trip if it is monitoring the neutral of one circuit, but current is returning via another path.

Sounds good but it appears the single circuit/ in single rigid metal conduit goes directly to load panel without any other branches. Current from other neutral loads is what I was also concerned about.

However if a GFCI TRIPS at a mere 6 to 7 ma difference between neutral and line I’m afraid I would need two oscilloscopes but I have only one plus the accuracy of each being within 4 ma agreement complicates this

 

[Jpflex]

a commercial electric brand amp clamp on white neutral shows 0.82 i with compressor going and fluke 98 with pico ta018 clamp on black lead shows 0.89 i with GFCI removed and replaced with standard 20i receptacle shown.

Under this condition GFCI TRIPS but can I really assume readings between tools are accurate? If it were this would be an excess of 4 to 7 mili ampere difference between line and neutral to load

 

[augie47]

I don;'t think it needs rocket science technology ..you have 3 basic culprits,,,the microwave, the refrigerator, a circuitry problem.
Most feel the refrigerator is the source, Run the refrig (only) for a time and see if it trips... if not, try the same with the micro.

 

[Jpflex]

Is the refrigerator self defrosting? Faulted defrost heater can trip a GFCI, but may only do so when there is a call for defrost, and would very possibly make it seem pretty random when it does trip.

I’m not sure as company bought this refrigerator brand new. All I can say is that GFCI trips when ac compressor kicks on

 

[Jpflex]

From these symptoms, it's very likely the refrigerator compressor is driven by a VFD. These types of refrigerators are getting more common. The VFD can have higher frequency common-mode currents that can trip GFCI's.



The short duration can be difficult to measure as you say. If you really wanted to compare the line and neutral currents you should put the clamp around both the line and neutral conductors so that you're measuring the difference in currents directly. But most clamp meters don't have enough sensitiivity and resolution to accurately measure in the 0-5mA range, even with a steady current level.

Frequency at receptacle is 59.8 hertz but is module internal to fridge is different or vfd then I don’t know. So far I have standard receptacle in place and there is no problem evident as before