Gfci wont trip with tester

Status
Not open for further replies.
Okay, so a HO is selling his VERY OLD house. The HI uses his "tester" on the 2 outside GFI's and, sure enough no trip. There was also a few open splices in the attic and a broken receptacle.

I send a J-man out there. He fixed the open splices and broken recep. He can't find anything wrong with the 2 GFI's, but changes them anyway. HO pays the bill.

HI comes back and fails the GFI's again... today! HO calls, made as heck. I send the same J-man back and he tells me that he changed the working GFI's last time. (Had not told me this earlier) and these are working too. Okay, now we know why.

Now, what do suppose the chances are, that the HI will not pass the GFI's or will require grounding or something. :-?
 
tallgirl said:
If neutral and ground are swapped, the GFCI will open under any load that is present downstream of the swap. That's because the current return that would normally be on the neutral at the GFCI will now be on the EGC. And if there is no load, the GFCI cannot tell the difference.
Click to expand...
You're right. But consider what happens when you plug a tester into a receptacle that has neutral and ground swapped. When you push the button on the tester, it connects a resistor between the hot and ground pins. This is intended to draw a small amount of current on the hot wire, which returns via the ground rather than the neutral. If the receptacle is miswired with the neutral where the ground should be, the current flows from the hot to the neutral, there is no imbalance in current between the hot and neutral, and the GFCI does not trip.
 
360Youth said:
but I test only with a plug tester or a Wiggy hot to ground (I think the load of the Wiggy trips it).
Click to expand...

The wiggy works but so does the test button on the unit.

The test button on the unit is a true test of the GFCI....well at least as true of a test as you going to get without some real fancy equipment to determine how many milliamps it takes to trip.

Using a wiggy to test an GFCI is a lot like using a 100 amp load to test a 5 amp fuse.

Yes it will open the fuse but you will not know if the fuse would have opened with a 6 or 10 or 20 amp load.
 
I always carry one of the little plug-in testers on finals. Mostly just to check polarity, but also as a convenience when checking receptacles fed downstrem from GFI's. If it trips the GFI fine, if not, I leave the tester plugged in and hit the test button on the GFI or grab my Wiggy and test them with it. Did a final on a garage yesterday where the little plug in tester wouldn't trip the GFI. Between the test button on the GFI recept and my Wiggy, I'm satisfied things were right.

I always assumed that when the plug in testers didn't trip a properly installed GFCI circuit, it was because it used a fixed trip point. Probably 5ma, right in the middle of the 4-6ma of the GFI. If the actual trip point of the GFI is 6ma, that little tester ain't gonna trip it. However, I could be wrong.
 
mvannevel said:
I always assumed that when the plug in testers didn't trip a properly installed GFCI circuit, it was because it used a fixed trip point.
Click to expand...

A plug in tester not tripping a GFCI that does trip when the GFCIs test button is used indicates no EGC present or other wiring problems.


Probably 5ma, right in the middle of the 4-6ma of the GFI.
Click to expand...

I think you will find that the plug in testers put a ground fault in place with more than 6 ma.

With a class A GFCI having a published trip point between 4 and 6 ma a tester with a 5 ma test point would be worthless.
 
iwire said:
A plug in tester not tripping a GFCI that does trip when the GFCIs test button is used indicates no EGC present or other wiring problems.
Click to expand...
Propery polarity at all the devices and an EGC present. The wiggy trips the GFCI when testing from hot to ground and neut to ground.

I think you will find that the plug in testers put a ground fault in place with more than 6 ma.

With a class A GFCI having a published trip point between 4 and 6 ma a tester with a 5 ma test point would be worthless.
Click to expand...

Could very well be the case with those testers. But then it'd be a lot like your example of testing a 5 amp fuse with a 100 amp load. You know it trips but is it actually doing it within it's trip range.
 
Julie,
And after you do that, go from neutral to ground and make it trip again. Then you know that part works as well.
Click to expand...
That won't work unless there is current on the neutral.
Don
 
don_resqcapt19 said:
Julie,

That won't work unless there is current on the neutral.
Don
Click to expand...

GFCIs induce a signal on the neutral even when there is no load. If that signal makes it back to ground the GFCI should open.

And there is almost always current available to flow between the neutral and ground because neutral and ground are typically not at the same voltage potential due to voltage drop between wherever the neutral and ground are bonded and the point where the measurement is made.

This article explains the principles of operation for the neutral to ground fault detection circuitry.
 
Julie,
And there is almost always current available to flow between the neutral and ground because neutral and ground are typically not at the same voltage potential due to voltage drop between wherever the neutral and ground are bonded and the point where the measurement is made.
Click to expand...

It is very rare case where I can get a GFCI to trip when putting the leads between the grounded and grounding conductors. Also I am not sure that most GFCIs work like the one shown in your link. Most diagrams I have seen only show a single sensor around the hot and grounded conductor.
Don
 
don_resqcapt19 said:
It is very rare case where I can get a GFCI to trip when putting the leads between the grounded and grounding conductors.
Click to expand...
Same here. A direct connection, maybe, but even my solenoid tester (Knopp K-60) neutral-to-EGC doesn't cause a GFCI to trip.
 
This document by Fairchild Semiconducor (one manufacturer of GFCI "brains") gives quite a bit more detail on the operation of GFCIs, including how neutral ground fault detection is implemented.
Click to expand...
I don't think that the UL standard requires neutral ground fault protection. I have not seen the dual coils required for this in GFCIs that I have taken apart or manufacturers drawings. If neutral ground fault protection is not required by the standard, it is unlikely to be included in the commonly available GFCI products. A single sensor GFCI will provide protection in the event of current flow exceeding 5 mA between the grounded and grounding conductors.
Don
 
In reading that Fairchild info it struck me that perhaps that particular product is not for the US market.

Does anyone know if another country requires the neutral fault protection?

I noticed that they mentioned 110 and 220 volts.
 
iwire said:
In reading that Fairchild info it struck me that perhaps that particular product is not for the US market.

Does anyone know if another country requires the neutral fault protection?

I noticed that they mentioned 110 and 220 volts.
Click to expand...

All the GFCI's I've ever tried to trip with a neutral to ground fault have done so. I say, give it a try at home. I play with computers at home, you can play with some wires :)
 
don_resqcapt19 said:
... I don't think that the UL standard requires neutral ground fault protection ...

... If neutral ground fault protection is not required by the standard, it is unlikely to be included in the commonly available GFCI products ...
Click to expand...
That reference to the UL Std caught my eye too. At best it was poorly quoted in that Fairchild document, but I assume that they meant to say "... when the neutral-to-ground resitance falls below 1.6 ohms ..."

Being one who ALWAYS checks the source ... I researched the applicable UL standard, but couldn't gain access to the standard's "actual guts" w/o paying a huge sum of $$$.

Anecdotally ... I've recently installed a number of brand-new Cooper Wiring Devices 20A GFCIs (on branches serving only the GFCI -- all with EGCs) and they ALL trip when the neutral terminal is shorted to the grounding terminal.

[Edit] to add: There were no connected loads either "on" or "downstream" of the GFCIs at he time of test; therefore I was not creating a main sense transformer current inbalance by creating a "bypassing" parallel neutral back to the panel via the EGC.
 
Last edited:
Anecdotally ... I've recently installed a number of brand-new Cooper Wiring Devices 20A GFCIs (on branches serving only the GFCI -- all with EGCs) and they ALL trip when the neutral terminal is shorted to the grounding terminal.
Click to expand...
Maybe that is a new requirement. I know that the older ones won't trip under those conditions. If it is a new requirement, I wonder about the reasons for it. There is no real safety hazard with a grounded to grounding connection that goes away with ~5mA of load and that is what happens with the older GFCIs. Maybe this is part of the new safety standard for GFCIs that also really isn't fail safe like the manufacturers have strongly implied.
Don
 
Status
Not open for further replies.
Top