GFI question -unique scenario

J

jcrafty

Member
Location
New York
Occupation
Electrical Engineer
Hello,
I have a question regarding GFI circuit breakers. We currently have a situation that is kind of unique and want to get some thoughts. We have a 3 phase system that is an old antiquated grounded B phase system. We use 240 volt single phase power to power 4 receptacles which come from 4 standard 20a breakers on a machine and some of these receptacles reference A-B , B-C and A-C for balancing as we have over a 125 machines with this setup. We are looking at installing a 2 pole GFI breaker for each of these receptacles which i see requires a neutral to the breaker for the electronics. However, the grounded B system does not have a neutral with it. We do have a control transformer on the machine that steps down the 240 to 120v with a neutral to run the 120 v power on the machine. So my question is two-part. First. Do you think the GFI would work correctly if the neutral on the breaker was tied to this secondary neutral. Second, if this did work, if the breaker was tied across A-B phase would it function properly as well if the one phase is grounded. In this system we measure 240v to ground on phase A and C but 0 volts to ground on B phase. My preliminary thought is this grounded phase may trip the breaker as it will think it is a ground path. Possibly a secondary transformer install would isolate the grounded B phase but unsure of that either. Appreciate any thoughts on this scenario. Thx
 
GFCIs can only be used where the voltage to ground does not exceed 150 volts, and that is not the case with your installation.
I think if the breaker is connected to the conductors from the corner grounded system and you used the neutral from the CPT, the breaker electronics will still have 240 volts, assuming the secondary of the CPT is grounded. If it is not grounded, there will be no real voltage between the CPT neutral and the conductors of the corner grounded system.

If you can use 208 volts at the receptacles the easiest would be 240 delta, to a 208/120 wye transformer.
If not you could use single phase 240 to 12/240 transformers for each receptacle.
 
A separate/independent 120V source will probably not work.

What are you trying to accomplish?
 
To provide ground fault protection for 240V single phase line to ground loads off a corner grounded delta I would recommend keeping the panelboard as is and using a IEC supplemental Residual Current Device (RCD).
Manufacturers such as a ABB and Allen Bradly offer them.
 
jim dungar said:
A separate/independent 120V source will probably not work.

What are you trying to accomplish?
Click to expand...
Thanks for the info, i will do some research on the RCB's. We are trying to add ground fault protection to the receptacles on machine. These receptacles are controlled via solid state relays and heat controllers. Mold heaters on the machine get plugged into the receptacles. The molds also have water lines for cooling purposes. We had an instance where one of the water lines was leaking on the mold and an operator went in to adjust something on the mold and he received a mild shock. Therefore we want to try and prevent this as best we can through a GFI of some sorts.
 
If I understand you correctly, I think I see a transformer and a pannleboard installation in your future. My customers hate me because I just flat out tell them the bad news straight up front. but they also return to me over and over again because I do it right, if you do some creative hodge podge solution, it’s going to be more expensive in the long run and unreliable. Possibly unsafe.
 
jcrafty said:
Hello,
I have a question regarding GFI circuit breakers. We currently have a situation that is kind of unique and want to get some thoughts. We have a 3 phase system that is an old antiquated grounded B phase system. We use 240 volt single phase power to power 4 receptacles which come from 4 standard 20a breakers on a machine and some of these receptacles reference A-B , B-C and A-C for balancing as we have over a 125 machines with this setup. We are looking at installing a 2 pole GFI breaker for each of these receptacles which i see requires a neutral to the breaker for the electronics. However, the grounded B system does not have a neutral with it. We do have a control transformer on the machine that steps down the 240 to 120v with a neutral to run the 120 v power on the machine. So my question is two-part. First. Do you think the GFI would work correctly if the neutral on the breaker was tied to this secondary neutral. Second, if this did work, if the breaker was tied across A-B phase would it function properly as well if the one phase is grounded. In this system we measure 240v to ground on phase A and C but 0 volts to ground on B phase. My preliminary thought is this grounded phase may trip the breaker as it will think it is a ground path. Possibly a secondary transformer install would isolate the grounded B phase but unsure of that either. Appreciate any thoughts on this scenario. Thx
Click to expand...
GFCI measure the difference in current between the ungrounded conductor and neutral but not the EGC ground circuit. Therfore i do not think your corner grounded delta would neccesarily result in gfci tripping.


You would have a delta high leg which would normally only allow you to use 2/3 phases to derived neutral. Better is a y secondary for balancing line to neutral loads

Delta is used on 3 phase motors more often
 
A standard North American 2 pole GFCI breaker is designed to operate on 120/240V single phase power. (Or 120/208V 'psingle' phase power.)

Internally the GFCI almost certainly uses one of the 120V H-N legs to power its electronics. (But there is no requirement of this; perhaps there are some GFCI breakers that run their internal electronics on 240V.) If you connect such a breaker to a supply that has 0V 'hot to white' on the wrong leg, then it won't work; if you have 240V 'hot to white' on the wrong leg then the magic smoke will come out.

Now if you set up a transformer so that you get the proper 120V 'hot to white', you might get the electronics to function correctly, but then you will still have the problem that the thermo-magnetic portion of the breaker is probably only rated 120V 'hot to ground', so you risk a breaker that will fail during fault conditions.

(This is also a reminder that you check your existing panel to make sure your existing breakers are 'straight rated' 240V devices, and not 'slash rated' 120/240V devices.)

@tortuga 's suggestion of using a 240V IEC device to _supplement_ your existing protection. But even here you will need to make sure that the devices are rated to have either or both poles 240V to ground. You should also know that IEC type RCB devices are not as sensitive as GFCI devices; they are more like GFPE devices.

IMHO if you want GFCI protection you will need to either install a transformer to provide 120/240V (or 120/208V if your systems can handle it) or you will need to contact a manufacturer to get GFCIs specifically designed for your existing system.

https://www.nssltd.com/ claims that they have GFCIs for straight 240V three phase; they might be able to come up with a box that monitors and shuts off your multiple single phase 240V circuits. They also sell plug in GFCIs made for 240V; perhaps you could just buy a few of these and temporarily install them when maintenance is working on a machine.

Yet another approach is to forget about GFCI type automatic shut-off devices, and install ground leakage monitoring on each machine. Essentially you have the 'sense' electronics of a GFCI without all of the safety related automatic shutoff stuff. Unfortunately I know how to do this in theory, but can't recommend any hardware to make this happen.


-Jonathan
 
winnie said:
Now if you set up a transformer so that you get the proper 120V 'hot to white', you might get the electronics to function correctly, but then you will still have the problem that the thermo-magnetic portion of the breaker is probably only rated 120V 'hot to ground', so you risk a breaker that will fail during fault conditions.
Click to expand...
That is an issue that is going to apply to all the breakers in the OP's corner grounded 240V delta system. Square D makes 3 pole GFCIs--I thought 3 pole breakers are always "full voltage" rated rather than slash rated?

If so, then in terms of "making it work" seems like you could use 3 pole GFCIs and figure out what combination of the 4 line side conductors the GFCI uses to power up its electronics (as long as it is just 2, you can do the test on a 120V/240V system, connecting just 2 line side conductors at a time and then using a GFCI tester on the load side to see if the GFCI is working or not). I'm going to assume that the GFCI is powered line to neutral and call the pole that is used "1". [If it's powered line to line, then things are much simpler, it will work on a corner grounded delta as is without the white pigtail connected to anything.]

Then make sure the the control transformer creates a line D that is 120V to the grounded B phase. And put the GFCI breaker in a 3 position enclosure--on the line side connect grounded B phase to the white pigtail, and A, C, and D to the 3 poles, making sure D goes to pole "1". On the load side, connect the 240V receptacle to whatever pair of A, B, and C you like.

As far as I can see, should work. And if the square D breakers are in fact rated for full 240V, the only potential code violation I see would be that they seem to be labeled 208Y/120V.

Cheers, Wayne
 
winnie said:
You should also know that IEC type RCB devices are not as sensitive as GFCI devices; they are more like GFPE devices.
Click to expand...
I am going to nitpick a little on that my interpreation of the definition of GFCI in the NEC is a performance requirement for Class A protection, not product manufacturing one there is no 5ma trip level or trip time in the NEC definition.
IEC and UL standards both address ground fault protection for personnel, and a RCD is tested with protection of personal as its key focus, UL GFPE is not.
However I am just some guy on the internet, but I hope to inspire some inspectors and state AJH's to look into it as it would save consumers $$ and solve lots of problems were having with EV's and HVAC equipment to just use the RCD.

For example if you ever swim in a pool in Europe (which I hope everyone does its nice) it's pool equipment will be on a 240V RCD, not a GFCI. I think there is good evidence suggesting that a RCD is all that is needed for 240V ground fault protection of personnel, keep GFCI for 120V.

1738350541288.png
 
Jpflex said:
GFCI measure the difference in current between the ungrounded conductor and neutral but not the EGC ground circuit. Therfore i do not think your corner grounded delta would neccesarily result in gfci tripping.


You would have a delta high leg which would normally only allow you to use 2/3 phases to derived neutral. Better is a y secondary for balancing line to neutral loads

Delta is used on 3 phase motors more often
Click to expand...
There is no high leg on a corner grounded system. Two phases are 240 volts to ground and the third is zero volts to ground.
GFCIs are only suitable for use on circuits that do not exceed 150 volts to ground, and you have more than 150 volts to ground on a 240 volt corner grounded system.
 
tortuga said:
I am going to nitpick a little on that my interpreation of the definition of GFCI in the NEC is a performance requirement for Class A protection, not product manufacturing one there is no 5ma trip level or trip time in the NEC definition.
IEC and UL standards both address ground fault protection for personnel, and a RCD is tested with protection of personal as its key focus, UL GFPE is not.
However I am just some guy on the internet, but I hope to inspire some inspectors and state AJH's to look into it as it would save consumers $$ and solve lots of problems were having with EV's and HVAC equipment to just use the RCD.

For example if you ever swim in a pool in Europe (which I hope everyone does its nice) it's pool equipment will be on a 240V RCD, not a GFCI. I think there is good evidence suggesting that a RCD is all that is needed for 240V ground fault protection of personnel, keep GFCI for 120V.

View attachment 2575457
Click to expand...
The NEC requires the use of listed GFCIs and the only product standard that GFCIs can be listed to is UL 943. There is no other listing standard named Ground Fault Circuit Interrupters.

Sure the other methods may provide adequate protection, but those other methods are not provided for in the NEC.
I have never seen a Public Input to permit the use of RCD devices in the NEC.
 
don_resqcapt19 said:
GFCIs are only suitable for use on circuits that do not exceed 150 volts to ground
Click to expand...
Is this a purely regulatory statement (listing standards for GFCI only cover this case), or is there a technical (physics) reason for this statement?

Edit: to make up one possibility, with 240V to ground, the capacitive leakage current will be higher, so 6ma trip threshold is too strict in practice.

Cheers, Wayne
 
Last edited:
wwhitney said:
Is this a purely regulatory statement (listing standards for GFCI only cover this case), or is there a technical (physics) reason for this statement?

Edit: to make up one possibility, with 240V to ground, the capacitive leakage current will be higher, so 6ma trip threshold is too strict in practice.

Cheers, Wayne
Click to expand...
Yes, it is part of the listing standard and there have been no proposals submitted to UL 943 to provide for the voltage to be higher.
Yes, there would be higher leakage at the higher voltage. Probably one of the reasons that RCDs have a 30 mA trip point. It might also require changes in the internal power supply to operate at higher voltages, but I don't think there is any technical reason.

The somewhat impractical and very expensive Special Purpose GFCIs operate at higher voltages. There is not yet a product standard for SPGFCI. UL 943C covers them, it is an outline of investigation and not yet a product standard.
 
don_resqcapt19 said:
The only product standard that GFCIs can be listed to is UL 943.
Click to expand...
I agree with you up till that statement, I am not seeing that in the definition of GFCI nor 110.2.
Perhaps I missed it, according to those I live with I am often wrong, but I don't see any reference in the NEC to UL 943, other a fine print note which is about as enforceable as the fine print note on voltage drop.
I think going by 110.2 a AHJ is within their rights to approve whatever products or listing standards they deem appropriate.
Otherwise UL would be a legislated monopoly.
 
don_resqcapt19 said:
There is no high leg on a corner grounded system. Two phases are 240 volts to ground and the third is zero volts to ground.
GFCIs are only suitable for use on circuits that do not exceed 150 volts to ground, and you have more than 150 volts to ground on a 240 volt corner grounded system.
Click to expand...
I know there is no high leg on a corner grounded delta because a neutral is not being "derived" at any mid point of 1/3 phase windings.

I am speaking of delta systems in general
 
Jpflex said:
I know there is no high leg on a corner grounded delta because a neutral is not being "derived" at any mid point of 1/3 phase windings.

I am speaking of delta systems in general
Click to expand...
The OP is speaking of 2-pole breakers feeding two wire 240V branch circuits off a corner ground delta 3-phase system so there could be 240V line to line or 240V line to ground.
If it were hi-leg sure there even could be 120 or 208 to ground depending on the delta system.
In theory a universal branch GFCI *could* be manufactured as 240V 'straight rated' to work on 2-wire 240 branch circuit on any US 240V delta system.
All the breaker would need to do is just measure the imbalance of the two line wires and power its electronics off the 240V line.
But no such GFCI breaker exists for US panelboards that I know of.
A IEC RCD breaker does exist for this and you simply mount it on DIN rail in a separate enclosure after the panelboard before the equipment.
Most if not all major manufacturers make RCD's that can mount on din rail.
 
Last edited:
tortuga said:
The OP is speaking of 2-pole breakers feeding two wire 240V branch circuits off a corner ground delta 3-phase system so there could be 240V line to line or 240V line to ground.
If it were hi-leg sure there even could be 120 or 208 to ground depending on the delta system.
In theory a universal branch GFCI *could* be manufactured as 240V 'straight rated' to work on 2-wire 240 branch circuit on any US 240V delta system.
All the breaker would need to do is just measure the imbalance of the two line wires and power its electronics off the 240V line.
But no such GFCI breaker exists for US panelboards that I know of.
A IEC RCD breaker does exist for this and you simply mount it on DIN rail in a separate enclosure after the panelboard before the equipment.
Most if not all major manufacturers make RCD's that can mount on din rail.
Click to expand...
Yes, if UL 943 was modified for other voltages you could do that. There is no technical reason. Not sure what type of power supply is used internally to power the electronics, but switching mote power supplies work over a very large voltage range and could probably be made small enough to fit in a GFCI breaker. I have seen such power supplies with an input voltage range from 65 volts to 300 volts. The CT part of the circuit would not care about the voltage of the circuit conductors. The current scope of UL 943 only provides for the GFCI to work on AC circuits supplied from 120 V, 208Y/120 V, 120/240 V, 127 V, or 220Y/127 V, 60 Hz systems. As far as I know the technical panel for UL 943 has not received a proposal to change the voltages. Even if there was such a proposal, the process to change a listing standard is pretty much the same as changing the NEC and takes at least as long.

The work to modify UL 943 to work with equipment that has high frequency leakage current has taken over 5 years so far, but they are close to issuing the standard for an optional GFCI-HF to address this issue.

Special Purpose GFCIs work with other voltages, but are expensive stand alone devices. There are no plans to make a breaker type SPGFCI that I am aware of, and the Outline of Investigation, UL 943C does not prohibit a breaker type SPGFCI. It may just be physical space issue in a breaker as the SPGFCI has an additional terminal for what I call a "ground check" conductor. This is a conductor run from the SPGFCI to the protected equipment. The SPGFCI flows a small amount of current via the ground check conductor and the EGC and if that path becomes open, the SPGFCI opens the circuit.
 
You must log in or register to reply here.
Top