Go Back  DoItYourself.com Community Forums > Electrical, AC & DC. Electronic Equipment and Computers > Electrical - AC & DC
Reload this Page >

Garage GFCI tripping. Believe a exterior lamp is the culprit. Any thoughts?

Garage GFCI tripping. Believe a exterior lamp is the culprit. Any thoughts?


  #1  
Old 11-02-15, 06:10 AM
H
Member
Thread Starter
Join Date: Nov 2015
Location: USA
Posts: 4
Upvotes: 0
Received 0 Upvotes on 0 Posts
Garage GFCI tripping. Believe a exterior lamp is the culprit. Any thoughts?

Hi All, I’ve been troubleshooting a garage GFCI that has been tripping for the past month.

Like a million google searches before me… Garage GFI tripped after heavy rain. It reset. Tripped again a week or so later. Reset, Tripped next day, reset, tripped within an hour and from that point it won’t reset.

Here are the steps I’ve taken so far. The GFI immediately trips as soon as I turn the fuse back on so I could only examine dead outlets.

1. Determined what is on the circuit. (Basement outlets before the GFI and 3 exterior outlets and a lamp post with daylight senor post GFI)
2. Check outside outlets. Found one with extremely rusted screws and was rusted so bad that I couldn’t even remove the wires. Cut wires, replaced outlet, sealed plastic casing. Found another with a little bit of rust, replaced it just to be safe and resealed casing. 3rd outlet was flawless.
3. Replaced GFI and examined load side. When removing the load wires the GFI would reset, so im pretty sure the GFI was fine but replaced anyway to just to eliminate any doubt moving forward.
4. In reviewing the load line, I see that the single load (hot) goes into a cap that has 3 wires coming out.
5. Isolated and energized each line. Line 1 (2 outlets, front and side), Line 2, 1 outlet (back porch), line 3, trips immediately and wont reset. Assuming this is the lamp (worst case..) All outlets work (when isolated) and test showing proper wiring using handy little “electrical plug testing thingy….”
6. Unscrew lamp top, make wiring diagram as there is a daylight senor. Remove lamp, cap wires. Turn fuse on. GFI trips and wont reset…
7. Senor? Remove sensor connections. Now the ground wire (coming up through the lamp post) is not connected to anything and the 3 wires are capped (white/black/copper). GFI trips and wont reset.
This is where I am… Has to be ground wire correct? Looks like im digging up the lamp post? Could it be wet? Could a mole or something have chewed through it?

A very important note here is I’m somewhat mechanically inclined but am 100% self and internet taught re: electricity. So if there are glaring errors in my above approach please let me know. I’m not against calling an electrician but I’m not sure he could tell me more than I know at this point… and this has taken me around 6 hrs over a few days going through all the testing and isolating to get to this point.

Right now, wire 3 is removed from load bundle behind the GFI. GFI outlet and all exterior outlets are working normally.
Any suggestions on where I can to go from here?

Thanks for your time.
 
  #2  
Old 11-02-15, 06:50 AM
I
Forum Topic Moderator
Join Date: Feb 2005
Location: Near Lansing, Michigan
Posts: 9,785
Upvotes: 0
Received 45 Upvotes on 43 Posts
Good approach on the troubleshooting. It does sound like the underground wiring is compromised to me. Rodents can do it, as can rocks shifting around in the soil or if the original installer didn't use underground rated cable the jacket will eventually rot through. A good place to start when working with a metal lamp post is to dig right down beside the post. There is often a hole cut or drilled through the post, and if that had a sharp edge may have cut through the cable jacket.

Next steps would be to isolate the cable at both ends so none of the wires are connected to anything. Then using the resistance setting on your meter, measure between Black-White, Black-Bare, White-Bare. These should have infinite resistance (no connection). If your meter detects a connection, you know for sure the cable is broken and shorted.
 
  #3  
Old 11-02-15, 07:05 AM
H
Member
Thread Starter
Join Date: Nov 2015
Location: USA
Posts: 4
Upvotes: 0
Received 0 Upvotes on 0 Posts
Next steps would be to isolate the cable at both ends so none of the wires are connected to anything. Then using the resistance setting on your meter, measure between Black-White, Black-Bare, White-Bare. These should have infinite resistance (no connection). If your meter detects a connection, you know for sure the cable is broken and shorted.
Thanks, my father-in-law and i dug up this post a while back (had sever sway) and also replaced the lamp top. So i am familar with what is down there. Just as you say, wire is in earth and enters pole about 6" from bottom through hole. I had that exact thought that the 90 degree turn of the wire in the pipe could have caught or snagged. Just odd that it worked for a while and then slowly started tripping GFI to the point of immediate trips. Its not like there is any force on that wire after install.

I do have a question about the resistance testing. I could isolate the hot wires by energizing, but there is no way for me to tell which white wires go with each back as they are in their own cap.

Do I need the corresponding neutral wire to my "wire 3" that i've identified? If so, how can i identify? I guess i could go through some time consuming trial and error again?

Right now I can easily locate lamp side wires and GFI side hot wire.

obviouslly i was trying to avoid digging up and resetting the post again but sounds like I may be at that point.
 
  #4  
Old 11-02-15, 11:36 AM
M
Member
Join Date: Oct 2015
Location: USA
Posts: 95
Upvotes: 0
Received 0 Upvotes on 0 Posts
For isolating which wires go to the same load, turn off power, separate all of the wires (cap the source wires), put a turned-on load on each circuit (e.g. a light bulb or a table lamp plugged into an outlet). Then use the ohmmeter to find the pairs. The ohmmeter should read only when wires from the same pair are measured.

Is it possible that the pipe of the lamp post cut the insulation? There should be a grommet, but it may have deteriorated.

Possibly the cable itself has filled with more and more water over time.
 
  #5  
Old 11-02-15, 11:50 AM
H
Member
Thread Starter
Join Date: Nov 2015
Location: USA
Posts: 4
Upvotes: 0
Received 0 Upvotes on 0 Posts
Thanks for your reply. anything is possible but i would consider it pretty unlikely that the post cut the wire after our install, and we were pretty careful when we reinstalled the post, and tested lamp before setting.

Im wondering if we created a void for water to fill since this all started after a heavy rain. Thanks for the info on the ohmmeter, forgive my electrical ignorance, but how do i put a "turned on load" to the wires when i have disconnected their line?

the load coming out of the GFI is 3 black and 3 white (i know which black is which). Are you saying i should hook one of the load wires up to the gfi (the back porch outlet for example) and then then plug a lamp/radio in that back porch outlet. Then try to read the ohms of all 3 white wires to see which one gives me a reading? All while everything is hot and exposed?

Or you you saying I will still get a reading from the disconnected circuit?

Sorry if I am misunderstanding. I have the basics down for wiring but thats about it.
Thanks again.
 
  #6  
Old 11-02-15, 12:12 PM
ray2047's Avatar
Member
Join Date: Mar 2006
Location: USA
Posts: 29,711
Upvotes: 0
Received 15 Upvotes on 13 Posts
Maybe I missed it but the cable to the light is UF not NM-b (Romex) isn't it?
 
  #7  
Old 11-02-15, 12:44 PM
I
Forum Topic Moderator
Join Date: Feb 2005
Location: Near Lansing, Michigan
Posts: 9,785
Upvotes: 0
Received 45 Upvotes on 43 Posts
There could have been a nick in the cable insulation which has allowed it to fill up and corrode. Check the bottom of the post -- doesn't need to be a massive excavation project, just enough to get to the cable and inspect it.

If the problem isn't right there at the bottom of the post and you have no other good candidates for a cable break (e.g. recently planted tree), I'd just go ahead and start digging a new trench to replace the cable with the assumption it's bad somewhere and difficult to find where without expensive testers or tedious spot digging and testing.

Another good proof would be to bypass the underground wiring using an extension cord and see if the light and daylight sensor work on the GFCI with the the underground cable bypassed.
 
  #8  
Old 11-05-15, 04:38 AM
H
Member
Thread Starter
Join Date: Nov 2015
Location: USA
Posts: 4
Upvotes: 0
Received 0 Upvotes on 0 Posts
Maybe I missed it but the cable to the light is UF not NM-b (Romex) isn't it?
Yes, I believe it is UF wire in the post/ground.
 
  #9  
Old 11-08-15, 10:21 PM
M
Member
Join Date: Oct 2015
Location: USA
Posts: 95
Upvotes: 0
Received 0 Upvotes on 0 Posts
Are you saying i should hook one of the load wires up to the gfi (the back porch outlet for example) and then then plug a lamp/radio in that back porch outlet. Then try to read the ohms of all 3 white wires to see which one gives me a reading? All while everything is hot and exposed?
No No No! NEVER use an ohmmeter with power applied to a circuit.

Disconnect all of the black wires and all of the white wires from the GFCI and from each other.

For one of these three circuits, either plug in a table lamp and turn the lamp's own switch ON, or if there is a ceiling or wall light on that circuit, turn its switch ON. Thus lamp will NOT light. but it provides a current path from the hot to the neutral of that circuit for the ohmmeter to find.

Now try the ohmmeter on each of the hot wires and on each of the neutral wires for each hot wire. When you have the hot wire and the neutral of the circuit you connected the lamp to, both wires go to the same circuit, and the ohmmeter will give a reading (usually under 1000 ohms).

Once you have found hot and neutral for the same circuit, mark both so you know which circuit they go to.

Now move the table lamp to one of the other circuits (or turn on a ceiling or wall light on that circuit). Again look for the hot and neutral that provides a reading (but do not test the already marked leads). Mark the leads giving the ohmmeter reading for the second circuit.

Repeat this process for the third circuit. In this case, there should be only one pair left to test.

Now you know which neutral goes with each hot.

Connect each identically marked pair alone to the GFCI (leaving the other two disconnected). This will find out which circuit is tripping the GFCI.
 
  #10  
Old 11-08-15, 11:52 PM
PJmax's Avatar
Group Moderator
Join Date: Oct 2012
Location: Jersey
Posts: 62,078
Received 3,422 Upvotes on 3,068 Posts
5. Isolated and energized each line. Line 1 (2 outlets, front and side), Line 2, 1 outlet (back porch), line 3, trips immediately and wont reset.
You have 3) two wire load cables there. If they are indeed cables....... and you know the hot/black wire.... just look into the box and find its white/neutral companion.
 
 

Thread Tools
Search this Thread
 
Ask a Question
Question Title:
Description:
Your question will be posted in: