Checking for hot/neutral reversal on GFCI onSho 2-wire?

Reply

  #1  
Old 09-23-07, 12:09 PM
Member
Thread Starter
Join Date: Sep 2007
Posts: 62
Checking for hot/neutral reversal on GFCI onSho 2-wire?

I've got a good-condition K&T wiring system. GFCI's have been installed (labeled as appropriate) on every outlet in the house, more or less. I haven't done all of these myself, or I wouldn't have an issue.

One problem I've had is distinguishing the hot and neutral. In some cases, I only have two black wires. Now, I do have a non-contact tester so I can pull the outlet, wirenut the wires, and then flip the breaker back on to test which one is hot - so that takes care of my newly installed ones.

What I'm wanting to make sure of is that the previously installed (not by me) GFCI's are wired correctly. This being a (mostly) 2-wire system, my receptacle tester just pops up the one light in the middle for open ground.

Is there any way to test for hot/neutral reversal other than pulling the outlets and testing which wire is hot? Or should the receptacle tester tell me about hot/neutral reversal if it existed, BEFORE open ground, since that's a more significant problem?
That's a lot of labor I'd LOVE to avoid, though I'll do it if I have to.

I do want to note that the earlier installed GFCI's are Leviton's of a recent vintage - probably within the last year or two. A good # of them have those always-on LED's. I don't believe Levitons check for hot/neutral reversal, though the new Coopers I'm installing do(IIRC), even if somehow I mess up with the non-contact tester.

Relatedly, I have at least 1 outlet that the receptacle tester indicates is wired correctly, even though I wired it myself and KNOW there is no ground. The only grounding in the house is to new circuits, plus one original ground wire up to the bathroom - all of those circuits check out fine. Any idea as to why this is the case? I am installing in the original metal boxes, which are thankfully large enough due to only 2 wires running into most boxes - but I'm screwing down the (unused) ground wire screw so it won't make contact with the box.
 
Sponsored Links
  #2  
Old 09-23-07, 12:43 PM
pcboss's Avatar
Forum Topic Moderator
Join Date: Jul 2002
Location: Maryland
Posts: 14,621
If you use your non-contact detector in the short slot of the receptacles it should indicate HOT.

Also the GFI device standard has changed. Now if they are mis-wired, or the internal circuitry has failed they should not reset.
 
  #3  
Old 09-23-07, 12:48 PM
Member
Join Date: Nov 2006
Location: Erie, PA
Posts: 305
Your plug in tester can't determine which wire is hot, because there is no 3rd ground wire to use as a reference.
 
  #4  
Old 09-23-07, 01:07 PM
Member
Join Date: Aug 2004
Location: NA
Posts: 1,065
Relatedly, I have at least 1 outlet that the receptacle tester indicates is wired correctly, even though I wired it myself and KNOW there is no ground. The only grounding in the house is to new circuits, plus one original ground wire up to the bathroom - all of those circuits check out fine. Any idea as to why this is the case? I am installing in the original metal boxes, which are thankfully large enough due to only 2 wires running into most boxes - but I'm screwing down the (unused) ground wire screw so it won't make contact with the box

Using the non contact tester in both slots would not imo be a 100% reliable method.

Using a neon tester or bulb tester and testing from both slots individually to a known good ground would be better. If your receptacle tester is showing good ground then it is there somewhere. Often a bootleg is installed by someone on the ground screw to the neutral screw and this will fool the tester into thinking there is a ground. but you installed this receptacle so must be a metal conduit on that circuit or someone ran a single ground wire to a grounding point from the metal box.

If you installed in a metal box you may have old bx or metal conduit of some type that is being used as the grounding means back to the main panel.

Screwing in the ground screw is fine but if it touched the metal box it still wouldnt cause a problem. Think about it,
the metal yoke of the receptacle is screwed to the metal box and the screw is threaded into the metal yoke.

Roger
 
  #5  
Old 09-23-07, 03:16 PM
Member
Join Date: Sep 2003
Location: Central New York State
Posts: 13,973
Your plug in tester functions just fine with no ground.
 
  #6  
Old 09-23-07, 10:52 PM
Member
Thread Starter
Join Date: Sep 2007
Posts: 62
I haven't used the non-contact tester in the slots. My reasoning is, once it gets that close, I can't 100% trust it not to pick up the other slot that's right there, even though it *shouldn't* on the lowest sensitivity setting.

Good thoughts on the BX cable, etc re: phantom grounded boxes - but I'm doubting that because we haven't seen *any* wiring upgrades in the house that weren't done within the last 5-7 years or thereabouts - ie, sunroom rewired with romex, basement dryer/washer circuit rewired with romex, kitchen & 1/2 bath remodel rewired with romex. Plus, we've been told by more than one electrician that they don't see a need to replace the K&T since it doesn't have any of "the usual problems" beyond lacking ground.

It doesn't hurt that the previous owner evidently spent hundreds and hundreds of dollars on GFCI's either

Anyway, I'm going to guess the so-called "ground" it's sensing might be the metal box nailed into studs, or in a couple of cases we've seen metal "fins" going to the stud. I SHOULD probably be replacing the old metal boxes with new plastic, but gosh, that's a lot of work - some of them are *really* in there, and I don't have a sawzall to get through the nails - I'd be scared of nicking a wire by accident anyhow.
Oh, and I'm screwing in the ground screw really just to make myself feel better, but I guess my original thinking went along the lines that if there WAS fault current dumped on the 3rd pin, it wouldn't have a direct path straight to the box, which is obviously attached to wood, etc.

I guess that's another question I have in this setup: if fault current gets dumped to ground pin on a GFCI with no ground wire, what happens to it? I understand the GFCI doesn't need/care about the ground, and I guess the only way for this to occur without tripping the GFCI imbalance detector is if overcurrent came from an OUTSIDE source somehow?

Also, Racraft, should I take that to mean that if hot and neutral were reversed, it should show me, even if there's ALSO an open ground?
 
  #7  
Old 09-24-07, 02:01 AM
Member
Join Date: Aug 2004
Location: NA
Posts: 1,065
I guess that's another question I have in this setup: if fault current gets dumped to ground pin on a GFCI with no ground wire, what happens to it?
If I'm getting the question correct your asking about what happens to the fault current. First the gfci is going to trip hopefully. So to give example take away the gfci and put in a grounding type receptacle but no grounding means present in the metal box. If we plug an appliance needing a 3 prong grounding power cord into this outlet and short the hot wire to ground inside the appliance ...since we don't have a completed path back to the source via a grounding means no appreciable amounts of current will flow..... Voltage will energize the case of the appliance and the ground wire in the cord and the the metal yoke of the receptacle and the metal box. There will be some leakage of current. In this case you have a booby trap and you could get yourself a nasty shock or worse if you touch that appliance case because it is at line voltage and waiting for you to give it a ground path so current can flow. Might not be a real good low impedance path but it's going to get your attention.

I understand the GFCI doesn't need/care about the ground, and I guess the only way for this to occur without tripping the GFCI imbalance detector is if overcurrent came from an OUTSIDE source somehow?
Gfci receptacles (not breakers) were never meant to trip on overcurent though they may or may not depending on the overcurrent fault. Usually in short circuits to ground it's a race with the circuit breaker on which will trip first given a circuit with a proper grounding path. They might both trip.

Gfci's shine at detecting small leakages of current in the ma range. Usually over 5 ma's difference between hot and neutral and they trip. They will not prevent you from getting shocked but should remove the fault before you are injured or worse.

What do you suppose would happen if , for example, you plugged a power cord into a gfci receptacle and then cut the cord and installed a single pole switch connected to the white neutral wire and the hot black wire then toggled the switch to "ON" ?

Please don't try this!!

Btw the metal fins are old work boxes and they hold the box in place because you already have the sheetrock installed and can't nail the box.

Roger
 
  #8  
Old 09-24-07, 02:33 AM
Member
Join Date: Aug 2004
Location: NA
Posts: 1,065
Oh...as bob said the tester doesn't need the ground. It makes a connection to the hot wire on the load side of the sensing coil to the neutral on the line side of the sensing coil when you push its test button.

Roger
 
  #9  
Old 09-24-07, 05:19 AM
Member
Join Date: Sep 2003
Location: Central New York State
Posts: 13,973
If you have an appliance plugged into a GFCI that has no ground and an short develops between the hot wire and the metal shell of the apliance, the following will happen:

Nothing, at least not initially. The shell of the appliance will be energized, but with no path to the source, no current will flow. However, as soon as someone touches the metal shall and provides a path to the course (back through say a wet floor or a metal water pipe, etc.), the GFCI will trip as the current will not be going back through the neutral of the GFCI.
 
  #10  
Old 09-24-07, 08:05 AM
Member
Join Date: Sep 2000
Location: United States
Posts: 18,497
Okay, I confess that I didn't read this whole thread. But I'll offer a few thoughts anyway.

Without a grounding connection to the receptacle, the plug-in tester can test for ground fault operation, but it cannot test for hot/neutral reverse.

Not 100% sure, but I'll bet that the new GFCIs that won't allow it to reset unless wired properly do not provide this extra feature without a grounding connection. I don't think they can tell whether it's wired properly or not without a grounding connection.

Stick the non-contact tester into the slots. If one side beeps and the other side does not, then you have found your hot. The shorter slit should beep. But if both sides beep, then you'll need to remove the receptacle and separate the wires as far apart as you can, and then use the non-contact tester. Generally, however, this isn't necessary.

You can also usually test this with a neon tester, even without a normal ground. Stick one probe into the narrow slot and hold the other probe with your fingers. You provide the ground. Granted, you don't make a great ground, but it'll be good enough for this test. Make sure the tester is in good shape (i.e., doesn't have an internal short) before you do this.
 
  #11  
Old 09-24-07, 11:22 AM
Member
Join Date: Aug 2004
Location: NA
Posts: 1,065
Bob is correct my statement about the gfci tripping if a ground fault occurs and no grounding means is not what would happen in the majority of cases. You have to touch the appliance to provide the grounding path.

Roger
 
  #12  
Old 09-24-07, 07:59 PM
Member
Thread Starter
Join Date: Sep 2007
Posts: 62
Okay, this has been quite instructional! I'm definitely glad we're replacing with GFCI's, and am in the process of testing outlets via the slots with the non-contact tester. I have already had to pull one outlet out to get greater wire separation using this method, but it's definitely saved me time since it's more reliable than I had initially surmised.

I am curious about whether or not (definitively) the newer GFCI's that won't reset unless wired correctly can't function without a proper ground. Is there no way for it to tell "hey, the short slot is getting power and the long slot is returning it? What about a manufacturer providing some internal sort of mini-ground just to test this, kind of like the hold-the-other-end-of-the-neon-tester concept? I'm probably misunderstanding something pretty big, but it's an interesting idea to me.

Oh, and I'll try not to "accidentally" wire any switches to extension cords for fun I'm thoroughly against opening up extension cords period, having seen some nasty pictures! Especially at today's prices...
 
  #13  
Old 09-24-07, 08:42 PM
Member
Join Date: Aug 2004
Location: NA
Posts: 1,065
All I can say is the best electrical engineers on the planet work for these companies that make gfci's and afci's and if they can't figure it out I sure as heck can't, which is rather an overstatement. My understanding is that the gfci will trip provided it can sense an inbalance in current in the neutral vs the hot wire. The only case I believe is when you similtaneously touch the hot and neutral on the load side of the gfci...there may be other situations where it won't trip. Some of the engineers here may be able to explain what your asking. A google search may get the answer for you.

I once watched a demonstration years back by some Sqaure d representatives at a school I was attending as part of my training. That is why I mentioned the switch deal. When that switched was closed there was a large arc inside the gfci and the branch circuit wires that they set up actually moved from the energy released in that short circuit to the neutral wire. This was an over current event. The gfci did not trip but the breaker did. They repeated this once more and the gfci fried but it did not trip the breaker tripped again. I was quite surprised but now that the newer technology is in the gfci's I'm not sure if this would happen. One other thing is people and electricians (notice I separated the two... ) have argued about rf tripping a gfci, let me tell you it sure can. We played heck in our plant with two way radios tripping out our conveyor control circuits that were wired to the load side of gfci service receptacles. You had to be about 2 or 3 feet away to get it to trip.

Roger
 
  #14  
Old 09-24-07, 08:55 PM
Member
Thread Starter
Join Date: Sep 2007
Posts: 62
Haha, Roger, sounds like you should be kept far away from portable radios

Yeah, I know a GFCI won't trip so long as the imbalance isn't too much. Is that situation that you mentioned (must have been a nice demonstration to see!) one thing AFCI's are intended, in theory, to help with?

I guess I should figure that engineers would figure out a way to do the hot/neutral/etc testing in a GFCI without ground if it were possible. I do wonder if it's POSSIBLE but not happening because of cost? I also wonder if the new "testing" GFCI's DEFINITELY do need ground to do their tests - John earlier said he wasn't 100% sure. I know that the load/line testing does work without ground, but that's a whole other issue, I'm assuming.

I might be tempted to call up Cooper and ask, but getting to someone who understands what I'm asking might be difficult, whereas here it's so easy...
 
  #15  
Old 09-24-07, 11:58 PM
Member
Join Date: Mar 2006
Location: NE Wis / Paris France{ In France for now }
Posts: 4,808
Originally Posted by Roger View Post

One other thing is people and electricians (notice I separated the two... ) have argued about rf tripping a gfci, let me tell you it sure can. We played heck in our plant with two way radios tripping out our conveyor control circuits that were wired to the load side of gfci service receptacles. You had to be about 2 or 3 feet away to get it to trip.

Roger

Roger ,, I done that also not only in commercal area but it did happend in one resdentail area it did the same thing too and one spot i did recall that i have a record service call on that day dang ,,, the Fire Dept was surronded one building and when their two way radios got pretty powerfull and it did trip all the GFCI repecales in that building including about 5 extra resdentails home right next to it that building have over 40 or so GFCI repectales so you get the idea of the " world fastest testing "

The building owner have to call me to come out and reset all the GFCI's took me a better part of the day to it.

Merci, Marc
 
  #16  
Old 09-25-07, 12:09 AM
Member
Join Date: Aug 2004
Location: NA
Posts: 1,065
Wow! that's one I haven't heard. But i'm a believer so it dang sure makes sense to me.

Roger
 
  #17  
Old 09-25-07, 12:19 AM
Member
Join Date: Mar 2006
Location: NE Wis / Paris France{ In France for now }
Posts: 4,808
Roger : yeah it is i was rather suprised how much it did wreck hovac on that one during that day it happend and that situation we have 6 fire trucks about half dozen police cars IIRC,, it was surround half of city block so it was pretty contrated two way radio traffic.

but i heard some Ham Radio operators do have some quirks with the GFCI in their place as well that have situation as well the only way i can slove it is ran the whole system in the EMT that useally cure it.

sorry for hijacking the thread and i think we better get back on track here

Merci , Marc
 
  #18  
Old 09-25-07, 03:56 AM
Member
Join Date: Aug 2004
Location: NA
Posts: 1,065
It's ok marc . Nice break in the action. I'm starting to get a little gun shy in my posts as have been making some careless errors of late.

It's really starting to annoy me. I fly around on 4 or 5 forums and it's apparent that I need to slow down and focus better.

take care
 
  #19  
Old 09-25-07, 05:21 AM
Member
Join Date: Sep 2003
Location: Central New York State
Posts: 13,973
A GFCI will trip properly with no ground wire attached to it when there is a current imbalance between the hot and neutral.

However, you cannot use an external GFCI tester of the plug in variety to make a test, because this tester attempts to send the excess current through the ground wire. Since there is no ground wire, no excess current flows.

When you have no ground to the GFCI you must test it with the internal test button or using an external tester that allows for the excess current (the test current) to be dumped somewhere else.
 
  #20  
Old 09-25-07, 08:20 AM
Member
Thread Starter
Join Date: Sep 2007
Posts: 62
I perfectly understand everything you said, Racraft - also wanted to note this isn't a GFCI tester, only a normal 3-prong outlet tester that's NOT supposed to make it trip.

My curiosity is around hot-neutral reversal detection in newer GFCI's - if it depends on the ground wire being connected or not. John earlier said he thought it did, and we discussed some methods of finding hot- neutral reversal externally assuming this to be true.

Peripherally, if it definitely does need the ground wire to detect hot-neutral reversal, I wonder if there's a way to make a GFCI that wouldn't need the ground - but that would amount to having a tiny tic tester in there, I guess, so size and cost likely become issues?

I don't mean to belabor the point - at this point, I understand 100% what's going on in MY setup and how to test and work with it, but I'm interested in a more academic sort of way.
 
Reply

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Thread Tools
Search this Thread
Display Modes