2 pole GFCI breakers for 120/208 single phase

C

ChesterDawg

Member
Location
CA
Occupation
Electrician
Doing some apartments with a 120/208 3 phase service and 125 amp 120/208 single phase to each apartment.

Having trouble finding any of the usual panel manufacturers with 2 pole GFCI breakers rated for 208v.

What are you guys using for something like this?
 
ChesterDawg said:
Interesting. We won't be needing any 2 pole AFCI. Just the GFCI for the dryer and range in some of the kitchens.

I got a support request in to Eaton and Square D, so if they ever reply I'll post what they say here.
Click to expand...
GFCI works on the concept of passing all protected conductors through a current transformer. If there is no leakage outside of the protected conductors said CT should produce no output voltage. No output voltage from CT means no initiation of trip logic circuitry. Even three pole GFCI on a 208/120 supply will work fine, High leg supply system would theoretically work fine as well, but they were not necessarily designed for the high leg voltage and/or should the GFCI electronics happen to end up on the high leg, it likely is only designed for 120 volts.
 
kwired said:
GFCI works on the concept of passing all protected conductors through a current transformer. If there is no leakage outside of the protected conductors said CT should produce no output voltage. No output voltage from CT means no initiation of trip logic circuitry. Even three pole GFCI on a 208/120 supply will work fine, High leg supply system would theoretically work fine as well, but they were not necessarily designed for the high leg voltage and/or should the GFCI electronics happen to end up on the high leg, it likely is only designed for 120 volts.
Click to expand...
Would the two pole GFCI be more sensitive on a wye system? With a single phase system a 7 mA ground fault on one leg and a 3 on the other, would result in the CT seeing 4 mA. With two phases of a wye system, the CT would see 10 mA under the same conditions.

CORRECTION. I did not do the wye calculation correctly. The two ground faults of 7 and 3 mA would sum to 6.1 mA using two phase of a 208Y/120 volt system
 
Last edited:
kwired said:
GFCI works on the concept of passing all protected conductors through a current transformer. If there is no leakage outside of the protected conductors said CT should produce no output voltage. No output voltage from CT means no initiation of trip logic circuitry. Even three pole GFCI on a 208/120 supply will work fine, High leg supply system would theoretically work fine as well, but they were not necessarily designed for the high leg voltage and/or should the GFCI electronics happen to end up on the high leg, it likely is only designed for 120 volts.
Click to expand...

Yup. Couldn't think of any reasons it wouldn't work, just trying to avoid any headaches with inspectors. It's my first time dealing with 120/208 in a residential setting like this.
 
Eaton confirmed their BR GFCI breakers are compatible with 120/208.
 

Attachments

  • Screenshot_20251006_064940_Gmail.jpg
    Screenshot_20251006_064940_Gmail.jpg
    362.1 KB · Views: 5
don_resqcapt19 said:
Would the two pole GFCI be more sensitive on a wye system? With a single phase system a 7 mA ground fault on one leg and a 3 on the other, would result in the CT seeing 4 mA. With two phases of a wye system, the CT would see 10 mA under the same conditions.
Click to expand...
No, the gfci may equal sensitivity in both, 10mA
 
Deltaforce said:
Neutral they move opposite direction
Ground they move same direction to source
Click to expand...
The currents behave exactly like neutral currents when the ground faults are from opposite legs. The actual current path is from the ground fault to some path back to the main bonding jumper to the neutral.
 
don_resqcapt19 said:
The currents behave exactly like neutral currents when the ground faults are from opposite legs. The actual current path is from the ground fault to some path back to the main bonding jumper to the neutral.
Click to expand...
But neutral conductor through CT gfci breaker carry total load return current (-)not leak to ground, other two power conductors through CT carry currents each load current (+') and earth leakage current e1 one power conductors and load current (+") and earth leakage current e2.other power conductor
So summation all currents through CT
(-)+(+')+(+") cancel
only e1+e2 unbalance remain ie 10mA
 
Deltaforce said:
But neutral conductor through CT gfci breaker carry total load return current (-)not leak to ground, other two power conductors through CT carry currents each load current (+') and earth leakage current e1 one power conductors and load current (+") and earth leakage current e2.other power conductor
So summation all currents through CT
(-)+(+')+(+") cancel
only e1+e2 unbalance remain ie 10mA
Click to expand...
Ground current does not return through the breaker GF CT, therefore it always creates an imbalanced condition.
 
Deltaforce said:
But neutral conductor through CT gfci breaker carry total load return current (-)not leak to ground, other two power conductors through CT carry currents each load current (+') and earth leakage current e1 one power conductors and load current (+") and earth leakage current e2.other power conductor
So summation all currents through CT
(-)+(+')+(+") cancel
only e1+e2 unbalance remain ie 10mA
Click to expand...
So you are saying that if I have a line to line load of 5 amps, that CT will not sum to zero?
It is the same with any ground fault currents on opposite legs of the 120/240 volt system. They sum in the CT just like any other loads and ground fault current on one leg acts to cancel ground fault current on the other when the CT sums those currents.
 
don_resqcapt19 said:
So you are saying that if I have a line to line load of 5 amps, that CT will not sum to zero?
It is the same with any ground fault currents on opposite legs of the 120/240 volt system. They sum in the CT just like any other loads and ground fault current on one leg acts to cancel ground fault current on the other when the CT sums those currents.
Click to expand...
I apply Kirchoff current law at node, two power lines and neutral line meet:
I1+I2+In=0
So In=-(I1+I2)
I1,I2 and In two line currents and one neutral current and all phasor quantities
This I use for neutral current through CT.
But currents in lines through CT include ground leakage currents e1 and e2 also
So total
I1+e1+I2+e2-(I1+I2)=e1+e2, the unbalance leakage current ie 10mA
 
Deltaforce said:
I apply Kirchoff current law at node, two power lines and neutral line meet:
I1+I2+In=0
So In=-(I1+I2)
I1,I2 and In two line currents and one neutral current and all phasor quantities
This I use for neutral current through CT.
But currents in lines through CT include ground leakage currents e1 and e2 also
So total
I1+e1+I2+e2-(I1+I2)=e1+e2, the unbalance leakage current ie 10mA
Click to expand...
Oops e1 and e2 as phasors also 180degree phase difference. Sorry Don you correct
 
don_resqcapt19 said:
Would the two pole GFCI be more sensitive on a wye system? With a single phase system a 7 mA ground fault on one leg and a 3 on the other, would result in the CT seeing 4 mA. With two phases of a wye system, the CT would see 10 mA under the same conditions.

CORRECTION. I did not do the wye calculation correctly. The two ground faults of 7 and 3 mA would sum to 6.1 mA using two phase of a 208Y/120 volt system
Click to expand...

I agree. I had mentioned that having 2 phases from a 208V wye would be a safer alternative for GFIs, especially for a marina application, in a previous post:
https://forums.mikeholt.com/threads/transformer-question.2566109/post-2733254

Let's assume we have a 2-pole GFCI which trips at a 5mA common-mode current. With the two phases 120 degrees apart, the plot below shows the resulting trip threshold as a function of the leakage current for one phase shown on the x-axis, and the other phase on the y-axis. Any currents above or to the right of the curve will trip. Worst case, there's only about a 15% increase of the trip threshold above 5mA on either phase. Ignore any negative currents because that would require a negative resistance.


2-pole_GFCI_trip_threshold_208V_system.jpg

It's still possible that you could have leakage currents on two 208V wye phases that will cancel. For example, if you put enough capacitive leakage on the leading phase to get an additional 60 degrees of phase lead, and with the other phase having only resistive leakage. Also, the magnitudes of these dissimilar types of leakage currents would have to be equal. I think that could be considered an outlier condition.
 
You must log in or register to reply here.
Top