arc fault devices

GerryB

Senior Member
I plugged my gfi tester into some arc fault devices and they all popped. They were just arc fault not dual arc/gfi. The plug in tester never trips an arc fault breaker but will trip a dual purpose breaker.
The device was Legrand pass and seymour. I called tech support and they didn't really have an answer other than the tester induces voltage into the device and the accurate test is to push the test button.
I am curious as to any thoughts on this or if anyone tested any other brand arc device and had the same thing happen.
(also the tech support person was ignorant of any code wording about metal raceways to the first device, listed breakers, etc)
 

James L

Senior Member
Location
Kansas Cty, Mo, USA
Occupation
Electrician
The closest I've come was plugging in an incandescent nightlight under a bed. I couldn't reach it well. It tripped at that receptacle, but did not trip at any other receptacle on that circuit (or any other)

I deducted that I was slow plugging it in because of reaching. It made a slight arc because being on.

At every other receptacle I tried, it was swift and I had no trips. It's not real scientific, but it's all I've got.

Made me wonder about slow connections, so I tried a toggle switch - very very slow, where you can normally hear a pop inside the switch - and it tripped the same breaker.

Maybe you're not quick enough plugging in.

I know it sounds goofy, but the whole arc fault experiment debacle is goofy
 

kwired

Electron manager
Location
NE Nebraska
The closest I've come was plugging in an incandescent nightlight under a bed. I couldn't reach it well. It tripped at that receptacle, but did not trip at any other receptacle on that circuit (or any other)

I deducted that I was slow plugging it in because of reaching. It made a slight arc because being on.

At every other receptacle I tried, it was swift and I had no trips. It's not real scientific, but it's all I've got.

Made me wonder about slow connections, so I tried a toggle switch - very very slow, where you can normally hear a pop inside the switch - and it tripped the same breaker.

Maybe you're not quick enough plugging in.

I know it sounds goofy, but the whole arc fault experiment debacle is goofy
There is also videos on line of people trying to make an AFCI trip, 1000+ watt heater for a load, touching, scraping, etc. an open segment in the conductors with lots of arcing going on. They can do that possibly for hours and the AFCI won't trip.
 

rnatalie

Senior Member
Location
Catawba, NC
Occupation
Retired Electrical Engineer
Depends on how much current your litle pluggie tester leaks. Arc Faults will trip on a ground fault but at 30ma rather than the 6ma that the GFCI is supposed to guarantee.
 

kwired

Electron manager
Location
NE Nebraska
Depends on how much current your litle pluggie tester leaks. Arc Faults will trip on a ground fault but at 30ma rather than the 6ma that the GFCI is supposed to guarantee.
or is also possible there is already leakage in the circuit and the tester added enough to put it over the 30 mA trip point.
 

kwired

Electron manager
Location
NE Nebraska
I did not know that about the 30 ma trip point. The tester I have is a Sperry.
Does it say anything about what test current is applied for GFCI test? I'm guessing most intended to test a class A GFCI do draw more than 6 mA when "testing" but likely still draw less than 30mA and even more likely no more than maybe 10 mA.
 

hbiss

EC, Westchester, New York NEC: 2014
Location
Hawthorne, New York NEC: 2014
Occupation
EC
I plugged my gfi tester into some arc fault devices and they all popped. They were just arc fault not dual arc/gfi.
A GFCI tester would only put 6ma between hot and ground.

Arc Faults will trip on a ground fault but at 30ma rather than the 6ma that the GFCI is supposed to guarantee.
Are you sure of that? As far as I know a straight AFCI has no GFI function.

-Hal
 

Fred B

Senior Member
Location
Upstate, NY
Occupation
Electrician
There was another thread here that mentioned that some straight up AFCI's had built in GFI even though not marketed with that as a function. That might explain the gfi tester tripping the AFCI breaker.
 

Little Bill

Moderator
Staff member
Location
Tennessee NEC:2017
Occupation
Electrician
A GFCI tester would only put 6ma between hot and ground.



Are you sure of that? As far as I know a straight AFCI has no GFI function.

-Hal
All of the AFCI breakers had a form of GF function in them when they first came out. That's why you couldn't share neutrals on an AFCI circuit and also couldn't have a ground neutral fault. Then GE came out with some that didn't have any GF function. Now there are a few more that have taken it out. I can't explain it but something about having to pass a UL test was the reason they had the GF function.
I'm thinking that a very high percentage of the trips on them were due to the GF function then the arc signatures they didn't recognize.
Now we don't even know if they work on an arc fault or what's inside them.
 

GerryB

Senior Member
There was another thread here that mentioned that some straight up AFCI's had built in GFI even though not marketed with that as a function. That might explain the gfi tester tripping the AFCI breaker.
Just to restate my gfi plug in tester never tripped an afci breaker, only the pass and seymour legrand afci device. It also did not trip whatever brand afci device depot sells.
 

synchro

Senior Member
Location
Chicago, IL
Occupation
EE
I did not know that about the 30 ma trip point. The tester I have is a Sperry.
Does it say anything about what test current is applied for GFCI test? I'm guessing most intended to test a class A GFCI do draw more than 6 mA when "testing" but likely still draw less than 30mA and even more likely no more than maybe 10 mA.
The Sperry GFI tester I have (model HGT6520) measures 16.56 kΩ from the hot to ground pins when I push the button (vs. 212 kΩ when it's not pushed). That would be 7.55 mA at 125V.

The clamp-on Ames meter I have can only resolve 10 mA increments, and so I made a 10 turn coil using a twisted pair of THWN for the hot and neutral wires to increase the resolution. With the GFI tester I could then see around a 7 or 8 mA increase in the measured current when I pushed the button, but there was about a 22 mA reading even without pushing the button. I suspect that the GFI tester has some higher frequency switching circuitry that may be causing this leakage current and the Ames true RMS meter is responding to this current. However, I haven't seen the tester trip a GFCI receptacle or breaker when the button is not pushed. I suspect this is because there's some internal low-pass filtering to attenuate high frequency noise and thereby minimize false tripping. As Tesla showed you won't get a shock if the AC frequency is relatively high (at least in the 10's of kHz), but you can still get a burn if the current is large enough.

So in the OP's case where the GFI tester is tripping an AFCI receptacle it may be such high frequency noise from the tester that's causing this. As far as I know most (if not all) AFCIs rely on measuring the energy in high frequency bands in order to sense the presence of an arc.
 
Last edited:
Top