Home Articles Forums Gallery Marketplace Videos

WoodCentral Forums

Knowledge You Can Build On — Est. 1998

If you sell to woodworkers, your customers are already here

GFI conundrum

Posts

GFI conundrum

#1

GFI conundrum

Howard

I have a dedicated circuit for my small compressor (17 amps). The circuit has a GFI outlet that the compressor plugs into and a light bulb socket and is controlled by a switch in the shop. The compressor is in a separate room.

Yesterday I plugged in the compressor and there was a spark it the GFI and the GFI stopped working. I tried replacing the GFI but the new one and a second replacement did not work. The green indicator light on the GFI did not come on.

I verified that there was power going into the GFI and that power was properly connected to GFI.

Called the GFI manufacturer and was told that the problem must be in the wiring between he circuit breaker and the GFI???

Anybody have an idea what I can do next?

Thanks

Howard

Re: GFI conundrum

#2

Re: GFI conundrum

TomD

You say the power is properly connected, but just to ask, your ground is good?

No idea if all the wires are OK. What is the point to a GFI in that case. Is there any way your compressor could be grounding the circuit? When you put the GFIs in, did you try something like a light first? From what you said though, you couldn't activate the GFI, which is why I wonder if you have a bad ground, or conversely, something is grounding the gfi box, etc...

Is it possible you have a problem with the breaker?

I am not an electrician. But I do occasionally have to sort myself through similar problems. there is a good electrical forum on Canadian woodworking I could post this to if you run out of options.

https://forum.canadianwoodworking.com/forumdisplay.php?23-Shop-Electrical

Re: GFI conundrum

#3

Jesper, Denmark

Re: GFI conundrum

Jesper, Denmark

GFI relays work by comparing the current in the hot and the neutral return. This should be equal or there is a leak somewhere. Is there a chance that the neutral is shared with another circuit somewhere between the GFI and the panel, thus fooling the GFI?

If the neutral is not properly grounded at the incoming panel, there is likely always some small current/voltage in the neutral.

I am not really familiar with GFI's for individual circuits, as the regulations here require GFI at the feed to the whole system before distributing to the individual circuits. Great for overall safety, but a pain if tripped, as the whole house goes dark.

Re: GFI conundrum

#4

Re: GFI conundrum

Barry Irby

How did you test the GFI after you installed it (them)? I use one of the little plug in circuit testers daily on my home inspections. There are many things they will detect and some things they don't.

Locally, GFI's are not required on refrigerators. I would have to check, but it may be that they are permitted on refrigerators, not just frowned on, because Fridges and Freezers cause false trips and you lose the contents. So, even in garages where GFI protection is required, the code allows for one unprotected outlet for the spare fridge. The point to all this is that your compressor may not be a good thing to plug into a GFI.

Not sure, but it seems to me 17 amps is a large draw for a GFI. Does your compressor pull 17 amps continuously? Or is that the start-up surge? You are not supposed to have a continuous draw of more than 80% of the circuit rating. So if it is a 20 amp circuit, its overloaded.

None of this explains why the compressor won't even try to start.

Have you plugged the compressor into a different circuit to see if the problem is with the compressor?

Re: GFI conundrum

#5

Re: GFI conundrum

Howard

I used a voltmeter and a plug in circuit tester, both

showed no output voltage. I did not plug the compressor into the two new GFI's I tried.

The compressor works fine on another outlet. It is a Makitka Mac 2400. Check the specs 12.3 amps.

Howard

Re: GFI conundrum

#6

In and Out

David Yoho

You said there was no output voltage. I assume you checked there is voltage at the input side (meaning the hot wires that connect to the outlet)???

If the hot wires are hot and there's no voltage on the output side of the GFCI outlet then I'd have to figure the outlets were bad unless something isn't wired correctly.

Re: GFI conundrum

#7

Hmmm.....

Barry Irby

You mentioned the green light......

On some brands, the light is on when the outlet is on/hot. On others the light is on when the outlet is off/tripped. (this is a bit of a mind bender for me. How could the light be ON if the power is OFF?) That brand uses the light as an Alert, signalling "the problem is here". They will not reset if the power is not On to the outlet. That is, if you trip the GFI and Turn Off the breaker, you can not reset the GFI until you turn the breaker back on. The point is, with the power on tot he GFI, have you reset it? AS I recall, the buttons on this type do not pop out far and it is not obvious that the button should be pushed. Push the reset and give it a try.

Maybe?

Re: GFI conundrum

#8

Re: Hmmm.....

Bill Howatt

I'd first do what Barry suggests, make sure the device is reset properly.

If that doesn't work, then remove all the downstream wires (if any) from the GFI and see what happens.

The term ground-fault interrupter is indeed that and it will sense if the neutral is connected to ground anywhere even if there is no current flowing in the circuit. This happened to me when I was restoring some wiring after a reno. The neutral had been connected to ground and I could not reset the GFI (in my case the light out indicated a tripped GFI). Puzzled me at first because I was thinking the circuit had to have some misdirected load current flowing but there was nothing turned on. The GFI is designed to detect this "ground fault" condition without any load current.

What is strange is that it worked and then it didn't work but you did get a spark although I would have associated it with something in the compressor.

You could temporarily hook up the GFI not fastened to its outlet box (floating in the air) without the ground wire connected to it and no downstream connections and see if that works. GFIs do not need a ground wire to work. If that corrects the problem then you have something wrong likely with the neutral-ground wiring.

Bill

Re: GFI conundrum

#9

I detest GFI's

John in NM

Yes, yes, I know it's all for safety, probably for the children too, but I hate the darn things - with a passion I hate them! :D

Why? Because under normal use they stop working after a couple years. They suck.

Ok, so what to do. What I do is replace them with a normal outlet and follow the common sense advice so concisely put by Red Green - "If you're standing in a puddle, don't touch anything that hums"

So many people are less than comfortable with this solution. Nothing wrong with that, for those people we have solution #2 (suggested by my NM state electrical inspector). Go to Harbor Freight and buy a six pack of cheepie GFI's for $20 (or whatever they cost now). Replace as needed. They will live no shorter a life than the premium varieties in my experience.

For what it's worth, the one's I've had the worst luck with all were outlets that had fairly large and consistent current draws - rather like what you describe. Outlets that see little use and light use seem to work fine for decades.

Re: GFI conundrum

#10

Wait a sec... a spark?

John in NM

That's kinda weird when plugging in a compressor that is not turned on. Makes me think there may be a problem with the compressor.

OTOH, replacing the GFI with another didn't work either - presumably without the faulty compressor - have you checked both the ground and the common in the circuit. Are you certain that they are only bonded at the panel?

That is a kinda weird one.....

I would still recommend plugging your compressor into a normal outlet. Twelve amps through a 15 amp GFI on a regular basis is exactly what I've seen kill them, but that may be neither here nor there in this case.

Re: GFI conundrum

#11

Re: Wait a sec... a spark?

Bill Howatt

He says it works fine on another outlet. If the automatic low-pressure switch was in the run state it would have tried to start when he put the plug in and the motor current could have caused a spark.

Bill

Re: GFI conundrum

#12

Re: Wait a sec... a spark?

John in NM

That is what I call plugging it in while turned on :D

Understandably, that happens now and then, but it is also a good thing to specifically avoid.

So it sounds like the three possibilities are a collection of bad GFI's (even the replacements tried), a ground fault elsewhere in the same circuit, or a ground fault in the compressor itself.

I would probably simply abandon the GFI's myself, but that is not an option for everyone. Otherwise, I would try plugging the compressor (turned OFF) into a different GFI circuit. If that works fine, I would try a new from the package GFI outlet - preferably not purchased from Home Depot or another vendor with a generous, even lax, return policy.

Should be easy enough to locate the actual problem.

Re: GFI conundrum

#13

OK, let's try a couple things

Tim Greif - West of Chicago

First off, let's understand what the GFCI is supposed to do. As others stated, it simply compares the current flowing out on the hot wire to the current returning on the common wire. If there's a difference of over 5 milliamps, it trips under the assumption that the missing current may be flowing through YOU. You can demonstrate this by plugging a test light into the hot side of the outlet and touching the common leg of the light to the box screw terminal. This should trip the GFCI if using an incandescent light. It will NOT trip if using a neon light since the current flow through the neon light is way below 5 milliamps. You might be able to detect 5 milliamps as a tingle, but it generally takes in the neighborhood of 250 milliamps to electrocute a normal human. Keep in mind, this is only a quarter of an amp.

Out of the box, a GFCI is normally in a "tripped" condition. Meaning it will not conduct if wired properly until you reset it. The little indicator light (on most) illuminates when there is power on the load terminals and the outlet socket terminals. Wired correctly, the light will glow when the GFCI reset button is released. When you push the reset button in, the light goes out until you release it, actually performing another test.

If you wire it up backwards (incoming line wires connected to the load terminals on the GFCI), the indicator light will always be on. The outlet will power anything plugged into it. Try that first on one of the replacements. If wiring it backwards doesn't let the compressor or anything else work plugged into it, you have a problem with the wiring coming into the box. Incidentally, wiring the GFCI backwards is not inherently dangerous, it just defeats the purpose of the GFCI, turning it into a normal outlet receptacle.

If the compressor works with the GFCI wired backwards, and will not power anything when wired correctly and reset, then I'm pretty sure it's a faulty GFCI. If you wire it backwards and indicator light does not illuminate, you either have a problem with the wiring coming into the box or you have a completely toasted GFCI.

If you're not comfortable doing any of this, then don't. But find someone who is. You may well have an electrical issue that needs attention.

Let us know how this turns out.

Re: GFI conundrum

#14

Agreed!

mdclor

We have one that that is dead, that is to say that nothing I plug into it works even though they work in other outlets. The curious thing is that the rest of the outlets that are downstream from the dead GFI all work fine. We don't ever need to use the dead outlet, so I haven't bothered replacing it. My father would have called it a "modern inconvenience". :D

Re: GFI conundrum

#15

Re: GFI conundrum

don stephan

As mentioned a new GFCI in the box won't work until reset, as if it was tripped at the factory. And there are two sets of terminal screws, Line and Load.

If you don't currently have a circuit tester with probes, I'd suggest one. I love the ETCON VT154. If you're not comfortable using a two probe tester, get someone that is. Electrocution is always a risk around wire connections.

With the breaker off, and the wires confirmed dead with your two probe circuit tester, remove the GFCI and leave the wire ends separated and clear of the box and each other. Ask someone to turn on the breaker for the circuit, and carefully test the not and neutral for 120V, then test each individually with the ground. The hot and ground will show 120V, the neutral and ground will show continuity. Have the other person turn off the breaker and join you while you connect the supply wiring to the GFCI. Some people would leave the GFCI proud of the box at this time, making sure the GFCI and outlet terminals are not touching the box. If the other person will turn on the breaker again you can test for voltage and continuity across the three connected terminal screws. Turn the breaker off long enough to reset the GFCI safely, noting that the reset button is now not as proud of the face of the GFCI. With the breaker on again you can test for voltage and continuity on the two LOAD terminals and the ground. Touching the probes to the LOAD hot and ground will trip the GFCI. With the breaker off again, reset the GFCI (it's not fastened to the box yet) and with the breaker on push to TEST button and the GFCI will click and trip. (It can be difficult with probe tips to connect with the female openings in any outlet plug.)

At this point you've confirmed 120V to the GFCI, that the LINE and LOAD are not reversed, and that the GFCI trips. With the breaker off one more time, secure the GFCI in the outlet box and add the cover plate.

Re: GFI conundrum

#16

Way TOO embarrassed

Howard

Got it working. I had not pushed the reset button in far enough. I the back of my mind I felt that the solution would either be very complex or stupidly simple.

Thanks for all the relies.

Bye the way the compressor is a Makita mac2400 and has served me very well for more than five years.

Howard

Re: GFI conundrum

#18

Re: Way TOO embarrassed

Bill Howatt

If your wife, should you have one, doesn't know about the cause, you are still ahead.

Nothing worse than when you are working on the next problem to get, "Well, did you press the reset thing properly?".

My problem is that my wife, who may know little about how gizmos work, seems to have a logical troubleshooting ability. Causing me to have to admit many times the problem was what she suggested and nothing to do with things like the "increased coefficient of friction due to relative humidity changes in the room" I was postulating.

Bill