Equipment Ground on a GFCI

[don_resqcapt19]

The issue with using something like a solenoid tester or the like is that it may verify the GFCI is basically functioning, but it does not confirm it has adequate sensitivity to ground fault leakage.

For a lark I measured a resistance of 14.6kΩ from the line out hot to line in neutral on a Pass & Seymour GFCI when the test button was pushed. So that would apply 8.2 mA of unbalanced current for a 120V input.
A Sperry plug-in tester had 16.6kΩ when the test button was pushed which would apply 7.2 mA with a 120V input.
I couldn't measure a Leviton GFCI because pushing the test button also mechanically trips open the contacts.

UL 943 requires that the test resistor be sized so that it will trip and 85% of rated voltage and so that the test trip current will not exceed 9mA at rated voltage.

Note that these tests are simply functional tests and are not intended to verify the sensitivity of the device.

 

[kwired]

The issue with using something like a solenoid tester or the like is that it may verify the GFCI is basically functioning, but it does not confirm it has adequate sensitivity to ground fault leakage.

For a lark I measured a resistance of 14.6kΩ from the line out hot to line in neutral on a Pass & Seymour GFCI when the test button was pushed. So that would apply 8.2 mA of unbalanced current for a 120V input.
A Sperry plug-in tester had 16.6kΩ when the test button was pushed which would apply 7.2 mA with a 120V input.
I couldn't measure a Leviton GFCI because pushing the test button also mechanically trips open the contacts.

Seems like a bad design to me. How does this confirm the sensing components are working if it mechanically trips it at same time?

 

[synchro]

Seems like a bad design to me. How does this confirm the sensing components are working if it mechanically trips it at same time?

From what I can gather, Leviton seems to have a different design philosophy. The test button does not actually perform the test, such as by connecting a resistor as in many GFCIs. Instead, the test button puts the GFCI into an initial state for testing by mechanically actuating the "trip portion" of the GFCI (which opens the contacts as described in the paragraphs below excerpted from the attached patent). After the GFCI has been tripped then the testing is actually performed when the reset button is pushed. If the "circuit interrupting portion" is not working properly then "reset lockout" will prevent the GFCI from being reset.
By the way, the test button is called "trip actuator" below.

"The trip portion preferably operates independently of the circuit interrupting portion so that in the event the circuit interrupting portion becomes non-operational the device can still be tripped. Preferably, the trip portion is manually activated and uses mechanical components to break the electrical connections."
"A trip actuator 26, preferably a button, which is part of the trip portion to be described in more detail below, extends through opening 28 in the face portion 16 of the housing 12. The trip actuator is used, in this exemplary embodiment, to mechanically trip the GFCI receptacle, i.e., break the electrical connection between input and output conductive paths, independent of the operation of the circuit interrupting portion."

"The circuit interrupting and reset portions preferably use electromechanical components to break and make the conductive path between input and output conductors. More particularly, the circuit interrupting portion is used to break electrical continuity between input and output conductive paths (or conductors) upon the detection of a fault. Operation of the reset and reset lockout portions is in conjunction with the operation of the circuit interrupting portion, so that the electrical connection between conductive paths cannot be reset if the circuit interrupting portion is non-operational and/or if an open neutral condition exists."
"A reset actuator 30, preferably a button, which is part of the reset portion, extends through opening 32 in the face portion 16 of the housing 12. The reset button is used to activate the reset operation, which re-establishes electrical continuity between the input and output conductive paths, i.e., resets the device, if the circuit interrupting portion is operational."

"Using the reset lockout feature … permits the resetting of the GFCI device or any of the other devices in the family of circuit interrupting devices only if the circuit interrupting portion is operational. Thus, testing of the circuit interrupting portion occurs during the reset operation. Further, if the circuit interrupting portion becomes non-operational after the device is set, the independent trip mechanism can still trip the device. In other words, the circuit interrupting device according to the present application can be tripped whether or not the circuit interrupting portion is operating properly."

https://patentimages.storage.googleapis.com/0d/e1/32/151a8037ca6365/US7545244.pdf

US7545244B2 - Circuit breaker with independent trip and reset lockout - Google Patents

Resettable circuit breakers having an independent trip mechanism and a reset lockout are provided. The trip mechanism operates independently of the fault protection operations, and the reset lockout prevents the resetting of the circuit breaker if the fault protection is non-operational or if an...

patents.google.com

 

[don_resqcapt19]

From what I can gather, Leviton seems to have a different design philosophy. The test button does not actually perform the test, such as by connecting a resistor as in many GFCIs. Instead, the test button puts the GFCI into an initial state for testing by mechanically actuating the "trip portion" of the GFCI (which opens the contacts as described in the paragraphs below excerpted from the attached patent). After the GFCI has been tripped then the testing is actually performed when the reset button is pushed. If the "circuit interrupting portion" is not working properly then "reset lockout" will prevent the GFCI from being reset.
By the way, the test button is called "trip actuator" below.

"The trip portion preferably operates independently of the circuit interrupting portion so that in the event the circuit interrupting portion becomes non-operational the device can still be tripped. Preferably, the trip portion is manually activated and uses mechanical components to break the electrical connections."
"A trip actuator 26, preferably a button, which is part of the trip portion to be described in more detail below, extends through opening 28 in the face portion 16 of the housing 12. The trip actuator is used, in this exemplary embodiment, to mechanically trip the GFCI receptacle, i.e., break the electrical connection between input and output conductive paths, independent of the operation of the circuit interrupting portion."

"The circuit interrupting and reset portions preferably use electromechanical components to break and make the conductive path between input and output conductors. More particularly, the circuit interrupting portion is used to break electrical continuity between input and output conductive paths (or conductors) upon the detection of a fault. Operation of the reset and reset lockout portions is in conjunction with the operation of the circuit interrupting portion, so that the electrical connection between conductive paths cannot be reset if the circuit interrupting portion is non-operational and/or if an open neutral condition exists."
"A reset actuator 30, preferably a button, which is part of the reset portion, extends through opening 32 in the face portion 16 of the housing 12. The reset button is used to activate the reset operation, which re-establishes electrical continuity between the input and output conductive paths, i.e., resets the device, if the circuit interrupting portion is operational."

"Using the reset lockout feature … permits the resetting of the GFCI device or any of the other devices in the family of circuit interrupting devices only if the circuit interrupting portion is operational. Thus, testing of the circuit interrupting portion occurs during the reset operation. Further, if the circuit interrupting portion becomes non-operational after the device is set, the independent trip mechanism can still trip the device. In other words, the circuit interrupting device according to the present application can be tripped whether or not the circuit interrupting portion is operating properly."

https://patentimages.storage.googleapis.com/0d/e1/32/151a8037ca6365/US7545244.pdf

US7545244B2 - Circuit breaker with independent trip and reset lockout - Google Patents

Resettable circuit breakers having an independent trip mechanism and a reset lockout are provided. The trip mechanism operates independently of the fault protection operations, and the reset lockout prevents the resetting of the circuit breaker if the fault protection is non-operational or if an...

patents.google.com

Not sure how that would comply with the UL 943 requirement.

5.15.1 A ground-fault circuit-interrupter shall be provided with a supervisory circuit that will conveniently
allow for periodic, manual testing of the ability of the device to trip by way of a simulated ground fault.{/quote]

 

[SSDriver]

Another way to test is with a Fluke T+ Pro. It has a GFCI test function. You put one lead in the positive terminal and the other either in the ground terminal (if it has one) or the neutral terminal and press the GFCI test button. It leaks a certain amount current to trip the GFCI. You can also use it for testing any style of GFCI breaker or device you access to put your leads on. Here is their rated levels/specs for the GFCI test current

GFCI test current

100 V - 150 V @ 6 mA - 9 mA AC, 150 V - 240 V < 12 mA

 

[LarryFine]

Another way to test is with a Fluke T+ Pro. It has a GFCI test function. You put one lead in the positive terminal and the other either in the ground terminal (if it has one) or the neutral terminal and press the GFCI test button. It leaks a certain amount current to trip the GFCI. You can also use it for testing any style of GFCI breaker or device you access to put your leads on. Here is their rated levels/specs for the GFCI test current

If you only have access to hot and neutral, you cannot simulate a shock.

 

[kwired]

Another way to test is with a Fluke T+ Pro. It has a GFCI test function. You put one lead in the positive terminal and the other either in the ground terminal (if it has one) or the neutral terminal and press the GFCI test button. It leaks a certain amount current to trip the GFCI. You can also use it for testing any style of GFCI breaker or device you access to put your leads on. Here is their rated levels/specs for the GFCI test current

No ground - put lead in the neutral instead does not place current outside the monitoring CT - device sees it as an acceptable load, if there is no ground and you want to trip it with any external load (which is all the meter you mentioned is doing) then you need to somehow bring a ground to your testing device from elsewhere.

 

[SSDriver]

If you only have access to hot and neutral, you cannot simulate a shock.

You are correct. Now that I think of it, I always go to ground or find a ground to go to. Well at least with the T+ Pro you can test other types of GFCI as long as you have a ground/bond

 

[kwired]

You are correct. Now that I think of it, I always go to ground or find a ground to go to. Well at least with the T+ Pro you can test other types of GFCI as long as you have a ground/bond

What other types?

GFCI is 4-6 mA trip.

GFPE is 30mA or even up to 100 mA trip. You could test those also if you have enough test load to ground or any other conductor at different voltage not protected by the GFCI for that matter.

 

[SSDriver]

What other types?

GFCI is 4-6 mA trip.

GFPE is 30mA or even up to 100 mA trip. You could test those also if you have enough test load to ground or any other conductor at different voltage not protected by the GFCI for that matter.

You could use the tester for GFCI breakers or receptacles as long as you have a ground to go to. Not all two pole breakers are for equipment. There are a lot of different sized pumps for swimming pools that need to be protected for personal and not for the equipment. A standard plug in GFCI tester would not work. However you could test the pump GFCI at the pump disconnect for a 240v motor, if you didn't want to just test at the breaker.

 

[kwired]

You could use the tester for GFCI breakers or receptacles as long as you have a ground to go to. Not all two pole breakers are for equipment. There are a lot of different sized pumps for swimming pools that need to be protected for personal and not for the equipment. A standard plug in GFCI tester would not work. However you could test the pump GFCI at the pump disconnect for a 240v motor, if you didn't want to just test at the breaker.

I agree, but I thought you were talking about a meter with two separate leads as the testing device. Can test between any two points with that.

 

[SSDriver]

I agree, but I thought you were talking about a meter with two separate leads as the testing device. Can test between any two points with that.

I was talking about a T+ Pro Meter. I can put one lead on either conductor of a Class A GFCI circuit and the other lead on any ground that is bonded back to the same source and it will trip the breaker. I should add. I bought the T+Pro because I like having this style of tester for general purposes. It works great for checking voltage with or without batteries and I don't see ghost voltages. The GFCI test function is just a plus. The size also fits well in small bags without bulk.

 

[SSDriver]

I was talking about a T+ Pro Meter. I can put one lead on either conductor of a Class A GFCI circuit and the other lead on any ground that is bonded back to the same source and it will trip the breaker. I should add. I bought the T+Pro because I like having this style of tester for general purposes. It works great for checking voltage with or without batteries and I don't see ghost voltages. The GFCI test function is just a plus. The size also fits well in small bags without bulk.

I should say any UN-GROUNDED conductor of the Class a GFCI circuit and it will trip the breaker or receptacle.

 

[kwired]

I was talking about a T+ Pro Meter. I can put one lead on either conductor of a Class A GFCI circuit and the other lead on any ground that is bonded back to the same source and it will trip the breaker. I should add. I bought the T+Pro because I like having this style of tester for general purposes. It works great for checking voltage with or without batteries and I don't see ghost voltages. The GFCI test function is just a plus. The size also fits well in small bags without bulk.

I should say any UN-GROUNDED conductor of the Class a GFCI circuit and it will trip the breaker or receptacle.

All low impedance meters will draw enough current at 120V+ to trip a GFCI if you test one protected line to any non protected point of differing potential.

Test non GFCI "hot" to the GFCI "neutral" that also will trip the GFCI. Test GFCI "hot" to a different "phase" that is not also protected by the GFCI and it will trip the GFCI. Test "hot" of one GFCI to "hot" of another GFCI (that is on a different phase) and at least one trips, though both likely will. All this has to have a low impedance meter or other current path that allows more than the 4-6 mA to flow.

 

[SSDriver]

Just made this of the T+ Pro before reading your post. My whole point is how useful the meter is for testing without the button. The meter will read from my kitchen faucet to the hot of the GFCI receptacle without tripping. When I press the test button it then trips. Not sure what the impedance is but its not enough to trip the GFCI without using the button. Tried it on multiple receptacles and have used it in the past at facilities with the same result. Also have used it on 2 pole breakers to any ground. Customers like to see simple tests, this is simple enough to show them it works.

 

[LarryFine]

A voltmeter (needed only if you need to know the exact voltage) and a solenoid voltage tester will do the same thing.

 

[SSDriver]

A voltmeter (needed only if you need to know the exact voltage) and a solenoid voltage tester will do the same thing.

T+ pro also does continuity and resistance. Its Cat III for 1000v, Checks AC/DC. If they added an amp clamp it would be the perfect meter for 99% of daily tasks.

 

[ptonsparky]

We have used the T+Pro for years but would not trust it when looking at ghost voltage. Applying a second T+ will drop the Voltage to half what the 1st indicated. A solenoid tester would drop it to nothing. My 87s may not change at all, or very little.

 

[Fred B]

And an Inspector would shoot it down ..

If Receptacle is marked as required in 406.3(E)(2)(b)(c) "GFCI protected No Equipment Ground" the inspector should recognize plug in tester limitations.

 

[synchro]

Not sure how that would comply with the UL 943 requirement.

5.15.1 A ground-fault circuit-interrupter shall be provided with a supervisory circuit that will conveniently allow for periodic, manual testing of the ability of the device to trip by way of a simulated ground fault.

Good point. Regarding that, I found the following in the Hubbell white paper at the link below:

"However, some competitive units do not conduct any test when the “test button” is activated. These devices rely on an exemption in the standard that allows for a review when the unit is reset. An artificial mechanical trip only takes place when the test button is operated. Portions of the electronics are tested during the reset mode by actuating the solenoid to unlock the device. During this process, the mechanical trip system is never tested. A hand held GFCI receptacle tester, such as the Hubbell GFT2G, must be used to fully test these competitive devices to ensure that the device is operational."

https://www.joexpert.net/wp-content/uploads/2016/11/Com_ST_White_Paper_Launch.pdf