GFI receptacles by garage door.

[don_resqcapt19]

GFCI receptacles are not required anywhere in the NEC.

Actually there is at least one rule in Chapter 6 that requires a GFCI receptacle, but in general that is a correct statement.

 

[iwire]

I have run into problem with GFCI's in the past with hand tools kicking a GFCI, and it was an inductive kick back problem,

Boy I sure would like to know what that means and how a cheaper GFCI has more problems with it.

I see large pipe threaders and large chop saws on GFCIs all the time without any issues.

 

[jxofaltrds]

Actually there is at least one rule in Chapter 6 that requires a GFCI receptacle, but in general that is a correct statement.

Don

Can you give me the Article to narrow down my search. I do not have the NEC on my computer so I can't cheat and hit Ctrl+F.

 

[jumper]

Don

Can you give me the Article to narrow down my search. I do not have the NEC on my computer so I can't cheat and hit Ctrl+F.

620.85 Ground-Fault Circuit-Interrupter Protection for
Personnel. Each 125-volt, single-phase, 15- and 20-ampere
receptacle installed in pits, in hoistways, on elevator car
tops, and in escalator and moving walk wellways shall be
of the ground-fault circuit-interrupter type.

 

[hurk27]

don

can you give me the article to narrow down my search. I do not have the nec on my computer so i can't cheat and hit ctrl+f.

620.85....

 

[jxofaltrds]

Thanks. Saves me a little time. I skipped over it earlier.

 

[hurk27]

Boy I sure would like to know what that means and how a cheaper GFCI has more problems with it.

I see large pipe threaders and large chop saws on GFCIs all the time without any issues.

Bob this problem is not new, it's been around for as long as GFCI's have been required, it has only be recently (in the last 10 years) that UL has been trying to address the issue in requiring GFCI's to have a design to lower the neusent tripping that this problem was causing, mostly because of the manufactures and code making panel wanting to expand the use of GFCI's into areas that was exempt for the requirement.
I identified this problem over 15 years ago for my own knowledge, using a capture scope with recording and a simple plug in motor to show how the high frequency pulse on a circuit can look like an imbalance to the GFCI circuit by not being in phase with the 60 Hz, the problem with this was easy to solve since this event was a one time shot if no residual resonance were to occur, but that is not always the case, other factors such as reactant and other things can cause harmonics to keep the spike going for a few cycles after the initial event, some will call this a ringing transient, this provided another problem to solve, well it was discovered by doing a gated time delay to delay the trip after a set time that would allow all oscillation to cease before allowing the electronics to view (monitor) the circuit after an event was the best answer to resolving the problem, so they redesign the National GFCI IC chip (LM1851) with this gated feature (some just call it a time delay) and most GFCI manufactures jumped on board to use this as well as other filters and TVSS devices to limit the nuisance tripping to come in to compliance with UL specifications for the newer GFCI's some of the other change in requirement was the neutral grounding issue and the detection of miss wiring which was also just lately incorporated in to them.

Much of this was posted in here a while back not sure if it was before the forum change of 2003 or not, but over the years I have posted about this problem allot.

Now that brings us to these what I call cheap Menard's GFCIs, well in the past couple years I have been finding them, and it not just from Menard's, CES has a version of them too, that have been quite a problem with inductive loads, not too long ago I posted a problem I encountered in an apartment building where the maintenance workers were using these to bring the apartments up to code to meet their HUD requirements, this was my first most profound experience with them, and I sent a few samples of them as well as some I went and purchased from Menard's, to UL for further analyzing, well the only thing I ever heard from UL on this was they were still trying to track down where in China they were being made, because these units were not using the UL listed National chip LM 1851 and didn't seem to have the delay that helped in stopping this problem, I have not made any more inquiry's to UL about the problem, but maybe I will after I get through this VFD install I'm working on.

Here is little of what Samuel M. Goldwasser had put in his book "Notes on the Troubleshooting and Repair of Small Household Appliances and Power Tools" back in 1996:

Why a GFCI should not be used with major appliances
A Ground Fault Circuit Interrupter is supposed to be a valuable safety device. Why not use them everywhere, even on large appliances with 3 wire plugs?
A properly grounded 3 prong outlet provides protection for both people and the appliance should a short circuit develop between a live wire and the cabinet.

Highly inductive loads like large motors or even fluorescent lamps or fixtures on the same circuit can cause nuisance tripping of GFCIs which needless to say is not desirable for something like a refrigerator.
Nuisance tripping of GFCIs
When used with highly inductive loads like motors or even fluorescent lamps, GFCIs may occasionally (or more frequently) trip due to the voltage/current spikes at power on/off. While the NEC/UL specifications apparently allow for some time delay in their response to combat this problem, it is not known if all manufacturers of GFCIs incorporate this into their product. However, the very common Leviton GFCI outlet probably does use the National chip (LM1851 Ground Fault Interrupter) referred to below. Also see the section: How does a GFCI work?.
(From: James Phillips (jamarno@juno.com).)

I quit having GFCI trouble after I fixed all the bad wiring connections, and I haven't had trouble at all with GFCIs and my workshop, which I wired myself. GFCI controller chips include a time delay to reduce false tripping. I used to think GFCIs always tripped at 5 to 6 mA, but the UL allows up to a whopping 200 mA if the GFCI stops the current within 30 ms, and 6 mA leakage is allowed to last 6 seconds.

According to National Semiconductor, their GFCI chips will stop a 200 mA fault in 20 ms, a 6mA fault in .5 sec.

And can be read in full HERE

Remember much of this book was written back before many of the newer requirements so his suggestions are not up to date to the newer requirements.

I do want to state that the other problems that you and other have mention are as real as this also, and have to be dealt with as needed just as finding any solution to a problem.

One of the most common symptoms of an inductive kick back is the very noticeable (if you look for it) tripping when power is removed from an inductive load, generally it does not occur when power is applied, this is a good telltale sign to the problem.

I know this was alittle long winded, but I hope I could share this info with the forum to help others in the understanding of this problem, and short cut some of their trouble shooting time.

 

[iwire]

Wayne I do not understand that.

The current in any part of a circuit at the same instant is always the same is it not?

In my limited knowledge there would have to be an alternate path for the current to flow on.

Kirchhoff's Current Law


At every node, the sum of all currents entering a node must equal zero. What
this law means physically is that charge cannot accumulate in a node; what
goes in must come out.

 

[iwire]

Bump ....

I was really hoping that someone could comment on the discussion Wayne and I had been having about GFCI trips.

 

[jumper]

Bump ....

I was really hoping that someone could comment on the discussion Wayne and I had been having about GFCI trips.

I will take a crack at it.

I believe that Wayne is saying that certain loads were creating distorted voltage/current waveforms that were causing the GFCI to trip.
The devices were sensing this distortion as current imbalance.

The solution was to put in time delay gate/chip/circuitry that allows the wave form to smooth out before it goes through the neutral current sensor in the GFCI.

We do this in digital electronics a lot. It is called "debouncing a switch".

The voltage and current leaving the time delay will be a cleaner waveform that GFCI can read better.

 

[iwire]

I will take a crack at it.

I believe that Wayne is saying that certain loads were creating distorted voltage/current waveforms that were causing the GFCI to trip.
The devices were sensing this distortion as current imbalance.

It is not making sense to me.

Lets say there are two different currents on a single two wire circuit one at the expected 60 Hz and one at say 120 hz.

Now run that two wire circuit through a single CT, it should still read zero.

 

[jumper]

Okay, let me try again.

The frequency does not change, the waveform does.

The current goes to the load as a nice smooth sinewave.

It returns as a jagged sawtooth wave of the same freq and amplitude.

The GFCI gets confused.

I put the jagged sawtooth waveform through a time delay/gate/filter and it comes out a smooth sinewave. The GFCI is happy.

I am not good at explaining stuff like this, maybe do a search on "debouncing a switch" can explain what I am trying to say.

 

[iwire]

But no matter how distorted the wave form gets it gets the same amount of distortion on both conductors at the same instant resulting on a net difference of 0 at the CT.

At least that is what it seems to me should be happening.

 

[jumper]

I think I understand now why my responses make no sense to you.

I am a moron. Sorry.

I closely reread what Wayne wrote and I misinterpeted it.

He was talking about inductive kick,

Inductive kickback: (a) Switch open. (b) Switch closed, electron current flows from battery through coil which has polarity matching battery. Magnetic field stores energy. (c) Switch open, Current still flows in coil due to collapsing magnetic field. Note polarity change on coil. (d) Coil voltage vs time.

When the pushbutton switch is actuated, current goes through the inductor, producing a magnetic field around it. When the switch is de-actuated, its contacts open, interrupting current through the inductor, and causing the magnetic field to rapidly collapse. Because the voltage induced in a coil of wire is directly proportional to the rate of change over time of magnetic flux (Faraday's Law: e = NdΦ/dt), this rapid collapse of magnetism around the coil produces a high voltage ?spike?.

http://www.allaboutcircuits.com/vol_3/chpt_3/9.html


and I was thinking of something like this, which is a switch closing with the contacts "chattering"

 

[hurk27]

Bump ....

I was really hoping that someone could comment on the discussion Wayne and I had been having about GFCI trips.

Sorry Bob this kind of got buried back a couple pages and I didn't see the response to it.

Back when I looked at the wave forms I didn't think to also look at the ground between the hot and EGC to see if any current is being induced into the grounded armature of a motor or solenoid, if it is then this can the "other" return path for the HV spike that is caused by inductive kick back.

At the time I was thinking that out of phase current was the cause of the GFCI trip, but your response got me thinking that it is passable that this kick back current might be induced or capacitive coupled to the grounded armature and flows back on the EGC thus causing the imbalance in the GFCI, a gated time delay as Samuel M. Goldwasser mentioned will still eliminate almost all nuisance tripping of GFCIs, as it still is only a one time event with a duration of a certain time, but maybe I should dig out my quad trace scope and do some experimenting with looking at the ground, I wish I had a power analyzer available like the one your using and posted screen shots.

In case there are a few who don't know who Samuel M. Goldwasser is, he was a engineer in the electronics field who also wrote the schematic books found in every TV shop around the world called "Sam Facts" and many other educational books.

Stay tuned Bob as I will keep digging into this.

 

[sinnadurai sripadmanaban]

Can different bodies(NEC/IEEE/UL) and different manufacturers specify different tripping times?. What is the National standard?.

 

[hurk27]

Can different bodies(NEC/IEEE/UL) and different manufacturers specify different tripping times?. What is the National standard?.

I would say they could, but I would also say they would also follow to what has been found scientifically and set as a safe level and tripping time that would protect a person from a shock hazard, so to me it would be in their best interest to follow a standard set for safety.

 

[don_resqcapt19]

The UL standard for GFCI trip time is:
[FONT=&quot]The maximum permitted time to trip in [/FONT][FONT=&quot]seconds[/FONT][FONT=&quot] is equal to the quantity (20/fault current in milliamps) raised to the 1.43 power. [/FONT]

 

[mikeames]

This thread is good brain food.

Don

I recgonize the 1.43 but where does the 20 come from in your formula?

 

[don_resqcapt19]

This thread is good brain food.

Don

I recgonize the 1.43 but where does the 20 come from in your formula?

Mike,
I have no idea...that is just the formula in the UL standard that the GFCI is required to meet.