Mustwin351
Senior Member
- Location
- Texas
Step down transformer and gfci
- Thread starter Mustwin351
- Start date
- Status
LarryFine
Master Electrician Electric Contractor Richmond VA
- Location
- Henrico County, VA
- Occupation
- Electrical Contractor
As long as that grounded wire is connected to the earth through an electrode or two, a GFCI should function correctly.
Mustwin351
Senior Member
- Location
- Texas
How does the gfci read an imbalance between the two lines on the secondary side in this setup? Since there are only two lines and one of those is grounded any fault to ground is the same thing as a phase to phase fault. I don’t see how the gfci read any imbalance in currents. (FYI this setup would be 480 to 120). If there was a center tapped neutral I would understand but that doesn’t exist in this setup.
ptonsparky
Tom
- Occupation
- EC - retired
One wire is bonded to the enclosure. Any current that does not return via the GFCI will cause an imbalance.
LarryFine
Master Electrician Electric Contractor Richmond VA
- Location
- Henrico County, VA
- Occupation
- Electrical Contractor
Inside, a GFCI passes both (or all) circuit conductors through a ring with a small coil on it. As long as all current on the hot returns on the neutral, no magnetism is generated in the ring, and no voltage is generated in the little coil.
If someone gets a shock, then some current returns to the source through the person, whatever grounded surface they happen to be in contact with, through the electrodes, and back to the source grounded conductor.
When that happens, the difference in current causes a voltage in the little coil, which triggers the GFCI to trip. It doesn't matter how many hots there are, as long as all of them, along with the neutral if there is one, pass through the ring.
That's why a GFCI requires that the system on which it's used has a grounded conductor. A shock (or equivalent leakage to earth or other grounded surface) has to occur for a GFCI to react.
If someone gets a shock, then some current returns to the source through the person, whatever grounded surface they happen to be in contact with, through the electrodes, and back to the source grounded conductor.
When that happens, the difference in current causes a voltage in the little coil, which triggers the GFCI to trip. It doesn't matter how many hots there are, as long as all of them, along with the neutral if there is one, pass through the ring.
That's why a GFCI requires that the system on which it's used has a grounded conductor. A shock (or equivalent leakage to earth or other grounded surface) has to occur for a GFCI to react.
ptonsparky
Tom
- Occupation
- EC - retired
The enclosure could be isolated from earth. It needs no GEC. The transformer has one conductor bonded to the box. Consider that box earth if it helps. By definition the secondary of a two wire transformer does not have a neutral, but it can have a grounded conductor. As long as there is enough current flow via that metal enclosure, the GFCI will trip.
synchro
Senior Member
- Location
- Chicago, IL
- Occupation
- EE
The key difference is that a fault current to ground occurring after the GFCI returns on a conductive path back to the point before the GFCI where you said there's "one line grounded to the frame of the box it’s mounted in." And so this conductive path will form a "ground loop" around the GFCI in which the fault current bypasses around the GFCI and does not return back thru the other conductor on the load side of the GFCI. This causes an imbalance in the currents on the wires going through the CT of the GFCI which can cause it to trip.
As Larry said there needs to be a grounding electrode connected to the grounded conductor on the line side of the GFCI. Effectively, this is an intentional ground fault to earth on the line side of the GFCI
. Then if there's a ground fault on the load side of the GFCI, a ground loop circuit that bypasses around the GFCI will be completed and the GFCI can detect the fault.
- Location
- Chapel Hill, NC
- Occupation
- Retired Electrical Contractor
A gfci on the line side of a transformer will not protect the secondaries as far as I know at least where there is only 2 wires on the secondary as in 120v to 12v.
- Location
- Wisconsin
- Occupation
- PE (Retired) - Power Systems
Except for an auto transformer configuration, a single phase transformer cannot pass a line to ground fault between the secondary and primary.
ptonsparky
Tom
- Occupation
- EC - retired
Where is the GFCI located? I don’t see where he states primary or secondary.
My SEWAG was secondary, because, WITH would you put it on the primary and expect it to work.
My SEWAG was secondary, because, WITH would you put it on the primary and expect it to work.
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
OP states that a 480 to 120 transformer is being used, implying a GFCI on the secondary side.
To the OP, perhaps instead of thinking in terms of 'imbalance' current, consider the term 'alternate path' current.
The GFCI will have something to detect if there is a path which current can follow 'around' the GFCI back to the transformer secondary. If current can go through one GFCI leg, through a person, and back to the secondary _not_ through the GFCI, then the GFCI will trip.
If there are no alternative paths, then the the shock risk is reduced but the GFCI has nothing to detect.
-Jon
To the OP, perhaps instead of thinking in terms of 'imbalance' current, consider the term 'alternate path' current.
The GFCI will have something to detect if there is a path which current can follow 'around' the GFCI back to the transformer secondary. If current can go through one GFCI leg, through a person, and back to the secondary _not_ through the GFCI, then the GFCI will trip.
If there are no alternative paths, then the the shock risk is reduced but the GFCI has nothing to detect.
-Jon
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
Basic operation of a 2 wire GFCI is you bring L1 and L2 (for the basic functionality doesn't matter which or if either of them is grounded) through a current transformer within the GFCI device. As long as current on L1 is equal to current on L2, the magnetic fields cancel one another and the CT produces no output voltage.
If there is a fault in either conductor beyond the CT (the "load side" of the GFCI) that allows more current than the trip setting to flow on some path that doesn't return through the CT, then that unbalances the current in the conductors passing through the CT and does result in enough output from the CT to initiate the trip process of the device.
The GFCI itself doesn't care what is or is not grounded, all it monitors is whether what current comes in is equal to what current that returns or in a Class A GFCI application is allowed to be up to 4-6 mA difference before initiating trip sequence. It will work on totally ungrounded system, but there needs to be more than one ground fault in that situation to yield the necessary current to unbalance the CT, the first fault just creates a ground reference for the system but no stray current flow.
If there is a fault in either conductor beyond the CT (the "load side" of the GFCI) that allows more current than the trip setting to flow on some path that doesn't return through the CT, then that unbalances the current in the conductors passing through the CT and does result in enough output from the CT to initiate the trip process of the device.
The GFCI itself doesn't care what is or is not grounded, all it monitors is whether what current comes in is equal to what current that returns or in a Class A GFCI application is allowed to be up to 4-6 mA difference before initiating trip sequence. It will work on totally ungrounded system, but there needs to be more than one ground fault in that situation to yield the necessary current to unbalance the CT, the first fault just creates a ground reference for the system but no stray current flow.
LarryFine
Master Electrician Electric Contractor Richmond VA
- Location
- Henrico County, VA
- Occupation
- Electrical Contractor
Any current flowing through a person is current flowing on a hot conductor that is not flowing back on the grounded conductor.
That's the difference current that the GFCI detects. Normally, all current flowing on a circuit mathematically totals to zero.
That's the difference current that the GFCI detects. Normally, all current flowing on a circuit mathematically totals to zero.
Not necessarily. People do get shocked by touching both conductors and that is not detected by GFCI.
LarryFine
Master Electrician Electric Contractor Richmond VA
- Location
- Henrico County, VA
- Occupation
- Electrical Contractor
I didn't mean to imply that every shock will trip a GFCI; only those that involve earth or other grounded surfaces.
A line-to-neutral (or -line) shock is seen as a load, does not pass through earth, and is not detectable by a GFCI.
Mustwin351
Senior Member
- Location
- Texas
I got it now. Thanks guys for the explanation.
LarryFine
Master Electrician Electric Contractor Richmond VA
- Location
- Henrico County, VA
- Occupation
- Electrical Contractor
To add, the reason the test button on a plug-in tester doesn't work with no EGC is:
A GFCI device has access to both line conductors within the device, and mimics a shock by connecting a resistor to the hot wire after the sensor to the neutral ahead of the sensor.
A plug-in tester only has access to both line conductors after the sensor, so can only mimic a shock by connecting its internal test resistor between the hot wire and the circuit EGC.
For those dealing with (home) inspectors who use a plug-in tester: Use a grounding "cheater" and extend the green wire with an alligator clip on the other end, clipped to a ground.
A GFCI device has access to both line conductors within the device, and mimics a shock by connecting a resistor to the hot wire after the sensor to the neutral ahead of the sensor.
A plug-in tester only has access to both line conductors after the sensor, so can only mimic a shock by connecting its internal test resistor between the hot wire and the circuit EGC.
For those dealing with (home) inspectors who use a plug-in tester: Use a grounding "cheater" and extend the green wire with an alligator clip on the other end, clipped to a ground.
- Status