GFCI Protection for two-wire circuits....

[wrobotronic]

Hi everyone,
Can someone explain the electrical theory behind replacing an old two-wire receptacle (with no EGC) with a GFCI receptacle? Im reading 406.4D2b-c and just trying to figure out how this adds a level of safety. If there is no EGC how is the GFCI protecting the circuit from ground faults per se. I'm not arguing what the article is saying, but I want to be able to explain to customers why this upgrade is benefitting them, instead of just saying "that's what the code says".

Thank you all very much.

 

[petersonra]

Hi everyone,
Can someone explain the electrical theory behind replacing an old two-wire receptacle (with no EGC) with a GFCI receptacle? Im reading 406.4D2b-c and just trying to figure out how this adds a level of safety. If there is no EGC how is the GFCI protecting the circuit from ground faults per se. I'm not arguing what the article is saying, but I want to be able to explain to customers why this upgrade is benefitting them, instead of just saying "that's what the code says".

Thank you all very much.

the gfci trips when there is an imbalance in current between the two conductors. It does not depend on a ground at all.

 

[Chamuit]

Check here, post #12 has a diagram.

http://forums.mikeholt.com/showthread.php?t=171743

 

[al hildenbrand]

If there is no EGC how is the GFCI protecting the circuit from ground faults per se.

406.4(D)(2)(b,c) is not about the "circuit". It's about "personnel."

Also, if there is a current leak from the insulated path from the energized conductor through the load and back on the grounded conductor (neutral), a leak high enough to trip the GFCI, it trips and, in de-energizing the wiring downstream of the GFCI device, behaves as if it were an overcurrent protective device responding to a very tiny current. . .

 

[JFletcher]

Hi everyone,
Can someone explain the electrical theory behind replacing an old two-wire receptacle (with no EGC) with a GFCI receptacle? Im reading 406.4D2b-c and just trying to figure out how this adds a level of safety. If there is no EGC how is the GFCI protecting the circuit from ground faults per se. I'm not arguing what the article is saying, but I want to be able to explain to customers why this upgrade is benefitting them, instead of just saying "that's what the code says".

Thank you all very much.

There doesnt have to be a ground wire for the circuit to have a ground fault. The GFCI senses any imbalance of current between hot and neutral; if above 6mA, it trips. If a person is touching something conductive at a lower potential (like a stainless steel sink) and grabs a running appliance (like a margarita mixer) with a fault in it (say a mouse-chewed cord) and comes into contact with the bare wire, a portion of the mixer's current will flow thru the person and into the sink rather than on the mixer's conductors, and back thru the GFCI. That imbalance will trip the GFCI receptacle if it's 0.006A or more.

 

[ActionDave]

My boss set up a GFCI circuit to show someone how GFCIs keep you from getting shocked. So he sets everything up on test bench, grabs the hot wire, grabs the conduit that feeds the test bench and then yells and yanks his hand back as fast as the GFCI tripped. Five milli amps doesn't sound like much and they say it's not enough to kill you but based on the way he jumped and yelled it still hurts.

 

[al hildenbrand]

My boss set up a GFCI circuit to show someone how GFCIs keep you from getting shocked. So he sets everything up on test bench, grabs the hot wire, grabs the conduit that feeds the test bench and then yells and yanks his hand back as fast as the GFCI tripped. Five milli amps doesn't sound like much and they say it's not enough to kill you but based on the way he jumped and yelled it still hurts.

Great anecdote! ! !

It's amazing how many think that a GFCI "prevents" shocks. I've never experienced a GFCI tripping ground fault through my body, to this day, nor do I need to. I've had enough other experience to satisfy me many times over.

But I'll tell one of the editors that I do consulting for that they still get a shock, and it is a surprise to them. Even when I explain the millisecond to millisecond happenings, they're still not sure.

Just because the GFCI trips at 5 milliamps doesn't mean, in any way, that the GFCI limits the amount of current that flows until the mechanism clears continuity.

 

[K8MHZ]

My boss set up a GFCI circuit to show someone how GFCIs keep you from getting shocked. So he sets everything up on test bench, grabs the hot wire, grabs the conduit that feeds the test bench and then yells and yanks his hand back as fast as the GFCI tripped. Five milli amps doesn't sound like much and they say it's not enough to kill you but based on the way he jumped and yelled it still hurts.

It's still 120 volts, and may be more than 5 mA for a teeny fraction of a second.

 

[kwired]

Plug a GFCI tester into a receptacle with metal weatherproof cover and no EGC. Tester will indicate no EGC. If no EGC is indicated don't press the test button (on your tester not the GFCI device), if you don't believe it can shock you, then make sure you are touching the metal cover when you do press the test button to get maximum effect.

 

[kwired]

Hi everyone,
Can someone explain the electrical theory behind replacing an old two-wire receptacle (with no EGC) with a GFCI receptacle? Im reading 406.4D2b-c and just trying to figure out how this adds a level of safety. If there is no EGC how is the GFCI protecting the circuit from ground faults per se. I'm not arguing what the article is saying, but I want to be able to explain to customers why this upgrade is benefitting them, instead of just saying "that's what the code says".

Thank you all very much.

Here is basics of the internals of a GFCI receptacle



The red ring not labeled in the image but is a current transformer. If what passes through on L1 comes back on L2, the magnetic fields passing through the CT cancel one another and there is no current in the CT. If current passes through on L1 and comes back via some path other then L2, or at least if more then the 4-6 mA doesn't come back, the processor is programmed to initiate trip mechanism.

Does not matter if that current not returning is via some grounded object, a neutral from another circuit, an ungrounded conductor from another circuit - the CT is imbalanced and it causes the device to trip. It will trip if you connect neutral conductor from another circuit to the load side of the GFCI as well.

You can also see in that image how the devices test button is wired to a resistor and when you press the test button current passes one way through the CT but bypasses the CT on the return path, causing an imbalance and makes the device trip.

Because the device trips whenever current is not the same on both protected lines it is very effective at protecting you in situations with no EGC. Any stray current over 4-6 mA will make the device trip, where a 15 amp fuse or breaker needs more then 15 amp flowing in the wrong places before there is even a chance of a fuse or breaker opening the circuit, EGC or not.