DIY Home Improvement Forum banner

GFCI Trips

7K views 28 replies 3 participants last post by  campbellpt 
#1 ·
Hope I can explain this clearly, as I am very new to working with electrical stuff. I have a GFCI outlet in my garage that is middle of the run and has two other outlets run from it (another outlet in the garage and one on my front porch).

I have 2 problems with this outlet:

1. While attempting to replace GFCI outlet yesterday (would not reset properly), I broke the ground wire right where the 3 ground wires entering the box are pigtailed. I really don't have enough slack to run a solid ground wire to the outlet. I did manage to slightly undo the pigtail and wirenut a bit of the same gauge wire into one of the grounds. Is this OK? Is there a better way? Am I violating codes?

2. The GFCI is middle of the run. I have the line wires going to the correct line holes. I have 2 seperate sets of load lines--one controls another outlet in the garage, the other goes to one (or more) outside outlets. If I connect just the line wires and test it, it is OK. If I connect the line wires and the load wires for the garage outlet, they both work OK. As soon as I try to connect the load wires for the outdoor outlet(s), it will not reset at all. Do you think it might be that the outdoor outlets might have some moisture in them? Could I have some bad wiring somewhere? Could my crazy rigged ground described above be throwing it off?

I'm pretty sure it is not the GFCI outlet, because I bought 2 new ones and they both do the same thing. Would it be better to replace the breaker witha GFCI breaker?

Thanks in advance for any suggestions. I'm at a loss and trying to avoid a pricy bill right before the holidays!
 
See less See more
#2 · (Edited)
Do you think it might be that the outdoor outlets might have some moisture in them?
Good possibility (check that first) from your description the problem certainly lies with that branch of the circuit. Anything plugged into those outside outlets? If so unplug them and see what happens. It is possible there is some other outlet on the load side your not thinking of ?.., lights or the like...so be sure that you aren't missing something plugged in elsewhere. In general check all outlet boxes on the problem branch for moisture or iffy contact of wires to the bare of each other or metal. If these things don't correct the problem then come back and we will tell you how to test the wiring. If nothing is plugged into these outside outlets then no current should be flowing in the wires ....unless there is contact with the hot wire letting a small amount of current flow to the neutral or ground wires. As a home owner that is inexperienced I don't reccomend you doing any hot (power on troubleshooting) other than voltage checks. As you know be sure to focus when you have the power on and when you have it off when your working on this.

So let us know what you find.



Could my crazy rigged ground described above be throwing it off?
Nope
 
#3 · (Edited)
Will Test

Thanks for your reply, I will check outside outlets for moisture (I'm at work right now).

I'm 98% sure there are only 3 outlets controlled by the GFCI in question--the GFCI itself, a second outlet in the garage, and one outdoor outlet (that trips the whole mess). I will check that when I get home as well.

I used to have a small fridge plugged into the GFCI, a stereo plugged into the other garage outlet, and low-voltage landscape lights on a timer plugged into the outside outlet (under a covered porch).

I unplugged everything before replacing the outlet in question, so I can't really point at any appliance.

Driving me crazy!
 
#5 · (Edited)
Thats fine....if nothing is plugged in then you have a problem with the wiring or the outside outlet on the branch that causes the gfci to not reset. Keep checking back others may reply. If you are comfortable with a few power on tests we can go into that also.

If you broke a ground wire off it is very possible that you can get a push connector on it then a pigtail to the other grounds or connect the others into the push connector. You can get them at lowes or HD. They look like this. GFCI's take up a lot of space in a single gang box so it might be crowded but should work.

 
#7 ·
I would love to try some tests (this will bug me until I can get it working), you'll just have to keep the description simple. The only thing I've done so far is to use a multitester to check between hot black and white neutral (which seemed to skip around quite a bit in the 120 range). I'm reading up on some of its other functions right now.

Fortunately, the main panel is only about 3 feet away from this outlet, so it is easy to switch it off, mess withe the wires, flip it back on, and test. An unfortunate side effect of this is that I switched the circuit breaker on/off so many times in rapid succession that I did forget once....ouch. Glad this isn't 240.

Also, I got flustered at one point and I unhooked everything from the initial replacement GFCI without labeling. Thought I had them straight in my head, reconnected them wrong (or maybe just the neutral and ground got too close?), hit reset button and a big flash/spark. No shock to me, didn't seem to be any damage to the outlet (it functioned--or should I say failed to function--the same as it had been all along), the neutral line wire had some burnt-looking discoloration going from about 1/2 inch to 1 inch from the end--doesn't look like it went through. Still, I can hook up just the GFCI outlet and it works, I can add the other garage outlet and it works too, but the outside outlet, once added, prevents the outlet from resetting.
 
#8 ·
Going back to the conditions that existed in your first post, with the GF tripped, confirm that there is no voltage downstream of the GF (because it's tripped) and then measure the resistance between any known good ground and the hot side (short slot or black wire) of the line that keeps tripping the breaker.

If it's less than 120v/4mA = 30 kilo-ohms you've found the problem (assuming there is nothing plugged in this line).

If not, there may still be line-to-ground resistance that only shows up with 120v applied. That's another post.

BTW, the GF "does not know" about ground.
It only knows that, of the current that was sent out, 4 mA or more of it are not returning through the same line, and so this missing current must be giving someone a shock.
 
#9 ·
OK, just want to make sure I have this straight, because I'm (clearly) not too sharp with this stuff.

I am going to leave the circuit braker ON and leave the GFI tripped, then go to the outside outlet (the one that keeps tripping it) and use my my multimeter or noncontact tester to see if there is voltage (there should not be any).

Then I am going to use my multimeter set to Ohms (which Ohm setting I will need to check...still learning about my multimeter) and leave the circuit breaker ON and the GFI tripped. I'll put the red probe to the black wire/short slot and the black probe...maybe in the ground hole to an extension cord plugged into an outlet with a ground that I know is OK (there are no other outlets nearby close enough to use).

Check to see if it's less than 120v/4mA. If so, I found the problem?
 
#11 ·
I discovered that the load lines in question cover a total of 3 outdoor outlets (not just the one that I mentioned initially). None have anything plugged into them,

I have a Greenlee DM-20 multimeter (I tried to post a link but it won't let me). Just not really good at using it, and the instructions that came with it are terrible.

I'll try the test you described with the MM set at 20K.
 
#12 ·
I can't seem to get any type of steady measurement to report back. I mean the numbers jump all over the place. Not sure if I'm just doing it wrong or what. Plan right now is to wait until daylight and rip open all those outdoor outlets and see what is going on inside. I will check back on here to see if there is anything else I need to be doing. Thanks for your help!
 
#13 · (Edited)
I can't seem to get any type of steady measurement to report back. I mean the numbers jump all over the place.
If the ohm reading is steady with the meter leads touched together, then the meter is good.

You might have found an intermittent short. This could be hopeful.

Very unlikely is that there is still somehow some voltage in the wires; I don't see how this is possible.

Your meter can't read AC current; slight change of plans if a leakage test is necessary.
 
#15 · (Edited)
Good. Everything is unplugged or switched off on this line, right?
You should read infinity (OL ?) from the hot cable lead to ground, same as if the leads were not touching.
I'd put it on the 200k ohms range, to start with.

If it reads infinity, or substantially more than 30k, the next step is the homebuilt "megger."
If it passes the megger test also then the problem is very likely not in the cable.
 
#16 · (Edited)
I think you should do a few things a little less complicated before doing ohms tests. The idea to find a fault is to divide and locate. You have 3 outside receptacles connected to the load side of the gfci as I understand things on the branch that is causing the problem. This means you have another junction in another box somewhere in one of those other boxes and you may have one or two cables leaving that box besides the supply cable from the gfci box. You need to disconnect those from the receptacle and keep the wires isolated in wire nuts if possible just don't let bare of the wire touch themselves or metal. This gets the other two receptacles and wiring disconnected from the gfci load protection. So now you just have the gfci and one outside receptacle connected. See if the gfci holds with just that one receptacle connected. If it doesn't then the problem is between the gfci and that one receptacle.
Now if you don't have a trip out turn off the power reconnect a cable or one of the cables to the receptacle. Now you have the next branch connected to the receptacle and to the gfci. Turn power back on and see if it trips out. If no then connect the last outlet in the same manner careful to remember to turn off power and test for power to be sure at the outlet.

See where I'm going?
Isolate the branches of wiring on the load side of the gfci where the fault lies then you only have to look at that one branch. One of the branches when connected back to the protection of the gfci is going to show the fault and you have narrowed down your search.
 
#18 ·
Of course! Somehow that never registered in my electronically challenged brain.

Correct, there are a total 3 outdoor outlets (1 front porch, 1 back porch, 1 in the middle of the back of the house) "downstream" from this GFCI.

Your post makes perfect sense, and I will try as soon as I get some daylight. Anything in particular I should be on the lookout for in these outside outlets that causes common faults but mught not be obvious to the untrained eye?


I think you should do a few things a little less complicated before doing ohms tests. The idea to find a fault is to divide and locate. You have 3 outside receptacles connected to the load side of the gfci as I understand things on the branch that is causing the problem. This means you have another junction in another box somewhere in one of those other boxes and you may have one or two cables leaving that box besides the supply cable from the gfci box. You need to disconnect those from the receptacle and keep the wires isolated in wire nuts if possible just don't let bare of the wire touch themselves or metal. This gets the other two receptacles and wiring disconnected from the gfci load protection. So now you just have the gfci and one outside receptacle connected. See if the gfci holds with just that one receptacle connected. If it doesn't then the problem is between the gfci and that one receptacle.
Now if you don't have a trip out turn off the power reconnect a cable or one of the cables to the receptacle. Now you have the next branch connected to the receptacle and to the gfci. Turn power back on and see if it trips out. If yes then connect the last outlet in the same manner careful to remember to turn off power and test for power to be sure at the outlet.

See where I'm going?
Isolate the branches of wiring on the load side of the gfci where the fault lies then you only have to look at that one branch. One of the branches when connected back to the protection of the gfci is going to show the fault and you have narrowed down your search.
 
#19 ·
Just wanted to thank both of your for your help. It's been a long and frustrating couple of days, and I'm going to try to get some sleep (just don't want you to post more tonight and wonder why I don't respond). Aside from a brief appointment, I'm off all day tomorrow. Plan on spending all of the daylight on this problem. Will check back throug hthe day and keep you posted on my progress (or lack thereof).
 
#20 · (Edited)
When you isolate the branch that has the problem remove the receptacle and keep all wires separated go back where you disconnected it and reconnect it to the gfci circuit this will tell you if it is the receptacle. If the gfci does not hold then it is a wiring problem. Something like a driven nail piercing the cable somewhere.

Heres the concept..... gfci's measure the current in the black wire (hot) and the white wire (neutral) (tell us if any red wires are in any of the boxes). If there is 5 amps flowing thru the black then there must be 5 amps in the neutral with a 4 milliamp or so tolerance. If the gfci sees 5 amps in the black but only 4 amps in the white it trips. This means that 1 amp is leaking to ground. Think of it like this... the amps flow out of the load side of the gfci on the black wire and return to the gfci on the white. The gfci must see this as being equal. If not then it trips. If nothing is plugged into any of the outlets then current cannot flow under normal conditions and the gfci just sees zero amps in both the black and white. Without something plugged in and turned on the circuit cannot be completed from the black wire to the white wire unless a hot wire has made a bridge by milliamps fault (small leakage of current) to the ground either at the receptacle or somewhere in the wiring. Lots of times water can accumulate on a receptacle and will allow current to flow from the black to the ground so you get the idea I think. Only the ground is not monitored by the gfci. How the gfci sees the problem is like this... if current is leaving the black wire to the bare ground wire somewhere on the load side of the gfci this current must flow thru the gfci on the black hot wire It is returning on the ground wire (not the white which is monitored by the gfci). So the gfci sees current on the black but nothing on the white and trips. Hope that makes sense.

So your gfci is tripping without anything plugged into the protected side of the gfci. This almost has to be a leakage of current from the black hot wire to bare ground. Also if your boxes are metal they are connected to the bare ground so leakage to metal can also cause the gfci to trip. Now this current loss to ground must be very very small. If not and the current loss exceeds the circuit breaker rating by large amounts the circuit breaker for the branch circuit the gfci is on will trip.

If you want a quick way to see if this is the problem disconnect all the grounds at the gfci with all the outlets connected to it correctly. Energize the circuit and the gfci will hold. This is because you have broken the return path and current cannot flow.....there isn't a complete circuit anymore. Without current flow the gfci senses no amps on the black wire even though it is leaking current to ground. Once you reconnect the bare grounds the fault returns and the gfci trips.
 
#21 · (Edited)
Here are a couple diagrams to assist you in understanding how your gfci works. I put three receptacles protected by the gfci to relate to your situation. The diagrams depict no loads (things plugged in) being operated from the receptacles. Bear in mind that it is possible for current to be leaking to both neutral and ground from water contamination. However the ground is not monitored and essentially is the root cause of the trip of a gfci in this situation. If the current was only leaking to neutral then both hot and neutral would be equal as far as the gfci is concerned and the gfci would hold all things being perfect.
 

Attachments

#22 ·
Fixed!!!!

Thanks for all the help. Here's what I did. I took apart all downstream outlets and figured our which was thge end of the run. That outlet (right above my AC units, which is never used) also looked suspect to me somehow. Can't say what, just looked "wrong." Went to lowers and picked up 4 new vanilla outlets ( 2 bucks a pop). Started with the suspect outlet. Whoever installed it used those stab holes in the back. When I tried to pull them out to check the condition oif the wires I couldn't get them to pull out. Seemed kind of strange. I snipped the wires, stripped off just enough to make a good curl and wrap it around the screws of a new outlet. Went to the garage and checked it out and voila! Power!!

Put all the others back together (with a healthy dose of silicone to keep future moisture out) and I was good as gold.

Just wanted to thank you again for the help. Sometimes the answer is right in front of you, but you just don't see it. Anyway, I don't use that outlet, and would like to put up some motion-detecting floodlights out back using those wires, although I won't be tinkering with wires till after thanksgiving at least. When I'm ready, I know who to check with. You guys are great!

Slainte!
PC



Here are a couple diagrams to assist you in understanding how your gfci works. I put three receptacles protected by the gfci to relate to your situation. The diagrams depict no loads (things plugged in) being operated from the receptacles. Bear in mind that it is possible for current to be leaking to both neutral and ground from water contamination. However the ground is not monitored and essentially is the root cause of the trip of a gfci in this situation. If the current was only leaking to neutral then both hot and neutral would be equal as far as the gfci is concerned and the gfci would hold all things being perfect.
 
#24 ·
Had to report back, you were a great help!

Still, I am a little irritated that I don't understand my multimeter yet. Weird, I've googled the heck out of it and still don't understand all its features (or even most of its features). Can you recommend a site or a book or something? Everything I've seen seem to skirt around the issue like it is some secret electrician's code or something. Even the instructions that came with the thing are kind of cryptic. Just hate owning any piece of equipment and not really understanding it.

Thanks!
 
#28 ·
Digital meters are not the best for house wiring but do have there place. They are more commonly used in the electronic field. However most electricians carry one for one reason or another. A anaolg multimeter is better. Digital meters are high impedance (unless labeled otherwise) and cause a lot of wandering readings when testing house wiring. Ghost or phantom voltages are a distraction when using these type meters and can be easily misinterpreted by the unknowing.

I'm not sure of a good site to give a tutorial on using a digital.. might try googling 'instructions for digital multimeter' or other search keywords.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top