GFI Problems in Beauty Parlor

Status
Not open for further replies.

flashlight

Senior Member
Location
NY, NY
Occupation
Electrician, semi-retired
Did service call today, new hair salon.

Two adjacent chairs are served by 3 110V receptacles on different phases of a 4-wire mwbc protected w/ 3 20A cb's (05 code) When they use one of their electric clippers it trips the GFI which is part of the male end of the blow dryer. It does this with different blow dryers and clippers. They are not plugged in on the same phase of the circuit.

The clippers are all rather old, metal frame w/ 2-wire power supply, no ground. They don't draw more than 2 amps. There
are 3 of them plugged into a power strip, but it seems they all can cause the tripping.

Could this be some kind of induction in the neutral that is fooling the GFI into seeing an imbalance ? Typically, the blow dryers are not even on when this happens so there is no current passing through the cords.

I am thinking I am going to have to get them a separate 2-wire circuit for
the clippers.

Would greatly appreciate any insight on this.
 
Last edited:

infinity

Moderator
Staff member
Location
New Jersey
Occupation
Journeyman Electrician
I don't see how separate neutrals will help when your 120 volt receptacles are being fed from a system that has a shared neutral. If you shut off two of the three circuits then you'll have only one circuit using that particular neutral. You'll then be able to see if it makes a difference.
 

grasfulls

Senior Member
Leakage / Differential current flows

Leakage / Differential current flows

I do think running a two wire circuit may help.

There is a good, though older,article in EC&M written by the then VP at Pass & Seymour/Legrand, here is an excerpt (I hope this is OK to post):
"Also, a differential current can result from other causes such as sharing the neutral with another ungrounded conductor or from capacitive leakage between the ungrounded circuit conductor and ground."
Jack Wells Vice President of Corporate Development for Pass & Seymour/Legrand. Copied from:
http://ecmweb.com/mag/electric_think_gfci/
 

flashlight

Senior Member
Location
NY, NY
Occupation
Electrician, semi-retired
I don't see how separate neutrals will help when your 120 volt receptacles are being fed from a system that has a shared neutral. If you shut off two of the three circuits then you'll have only one circuit using that particular neutral. You'll then be able to see if it makes a difference.

Rob, we can't shut off 2 of 3 circuits because they need at least 2 circuits
to run blow dryer and clipper at the same time.
 

Article 90.1

Senior Member
My canned answer to GFCI troubles: is it a ground fault? Sometimes the devices are actually doing their jobs. But back to the program... Have you tried removing the surge protector strip?
 

ELA

Senior Member
Occupation
Electrical Test Engineer
I do think a separate circuit could help. It sounds like the hair dryer GFCIs are sensitive to the noise generated by the clippers.

Increasing the impedance between the clipper and the GFCI will reduce the noise level (if a separate circuit means longer wire lengths between the two).

You might also try inserting an EMI filter on the clippers circuit to isolate the noise from other circuits.
 

DavidA

Member
Location
Fresno, CA
Yeah maybe check each receptacle individually with the clippers while the other two are de-energized. May at least help you gain a little insight into the cause of the problem. May sound silly, but are all 3 circuits really on different phases?
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
100224-0811 EST

I concur with ELA on the probable cause. Also may imply the GFCIs are poorly designed.

A simple quick test or two.

Get a long extension cord and plug it in to some other source a long distance from where the blow dryers are located. This might be an outlet near the main panel. Or maybe an outlet in an adjacent business.

Get a low pass noise filter, Corcom is one company. Put this filter between the outlet to which the clipper is connected, and the clipper to the output of the filter.

.
 

ramsy

Roger Ruhle dba NoFixNoPay
Location
LA basin, CA
Occupation
Service Electrician 2020 NEC
IMHO Load current passes thru GFCI (N-G) if shared neutral is only wire with load.

Simple Proof:
1) 2.0-vac is a typical potential (N-G) for 120v receptacles with grounding.
2) Using Ohms law W=(V*A), 6.0w = 2.0v*(3.0A clipper load) max neutral load on MWBC.
3) 6.0w also =(120v*0.05A). GFCI's may trip near 5.0mA measured thru (N-G) terminal.
4) (N-G) voltage > 2.0 may push load thru GFCI prongs, even with a measured resistance.

Simple Test using LOTO/PPE. Prohibit GFCI's from measuring shared neutral loads (N-G):

A) Jumper duplex prongs shorting (N-G). Test 0-vac (N-G) & see if nuisance trip is fixed.

B) Or, plug hair dryers at End Of Line (20A extension cord), or swap location with clippers. Neutral-load path must flow from panel without passing thru hair dryer outlet w/ GFCI's.

C) Or, for 120v loads only, at panel temporarily pig tail a few hots to the same breaker.
 
Last edited:

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
100224-0859 EST

ramsy:

I do not understand your post.

On the supply side what does the neutral to ground potential have to do with tripping the GFCI? The EGC is simply a pass thru conductor in a GFCI. Therefore, there is only a very small amount of capacitance between the EGC and any of the electronics, hot, or neutral in the GFCI.

Same comment applies to neutral to EGC on the output side of the GFCI.

On a Leviton 7899 the capacitance from EGC to the remainder of what is in the receptacle is about 22 pfd.

Very fast large transients might be coupled thru this small 22 pfd, but then there should be noise filtering in the basic design of the GFCI, and this coupled noise, if it existed, should not be a problem.

.
 

flashlight

Senior Member
Location
NY, NY
Occupation
Electrician, semi-retired
My canned answer to GFCI troubles: is it a ground fault? Sometimes the devices are actually doing their jobs. But back to the program... Have you tried removing the surge protector strip?

If it is a ground fault, it would not be on the circuit monitored by the GFI, unless somehow current is induced in the cord. As I mentioned, the hair dryers are not even energized when it trips.

We will be returning there on Monday (they're closed on Mondays) and plan
to remove surge protector, as well as other suggestions, including filters and testing with long extension cord in case we have to run new ckt. Thanks everyone for suggestions, we have done other hair salons and this is the first time I've seen this one.

BTW, how would the surge protector strip be contributing to this ?
 

ramsy

Roger Ruhle dba NoFixNoPay
Location
LA basin, CA
Occupation
Service Electrician 2020 NEC
The EGC is simply a pass thru conductor in a GFCI.
If GFCI's don't measure (N-G) then your right, but listed diagrams may deliberately omit proprietary designs, a common practice to protect intellectual property. A patent application would provide more reliable design diagrams.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
100224-1031

ramsy:

Of the several GFCIs I have taken apart none have any connection between neutral and EGC, or monitor this voltage,


flashlight:

A surge suppressor at a minimum is probably one to three MOVs connected to limit transient peak voltages, and possibly filtering in addition to the MOVs. An MOV does not limit the rate of rise of a voltage, it simply softly limits the peak voltage. Softly means the V-I curve does not have sharp threshold like a Zener diode.

I do not see an obvious reason for a surge suppressor on the input side of a GFCI to cause the GFCI to trip. Output side is a different story.

A well designed GFCI should not be tripped by any perturbations of the source voltage. A poor design may be susceptible to rapidly changing voltages on the source voltage.

Quite obviously these comments do not mean there are not voltages on the input side that will falsely trigger or fail a well designed GFCI. If you apply a high enough and long enough input voltage relative a device's rating you will fail any device. I have tried to falsely trip the Leviton GFCI with transient voltages that might be generated on a normal AC 120 V circuit and I have not been able to trip this device. Very high RF field intensities might trip it, but these are unlikely.

.
 

ramsy

Roger Ruhle dba NoFixNoPay
Location
LA basin, CA
Occupation
Service Electrician 2020 NEC
Of the several GFCIs I have taken apart none have any connection between neutral and EGC, or monitor this voltage
No connection necessary for current transformers, toroids in GFCI's, to measure current.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
100224-1931 EST

ramsy:

Since you seemed to suggest in a prior post that a neutral to ground voltage might be the cause of the tripping problem I simply stated, not by conjecture but by actual circuit analysis, that actual physical observation of several different GFCIs proved that for those units there was no functional use of the EGC relative to the GFCI operation.

In prior threads there are those that have believed that the ECG was necessary for the GFCI to perform its task. This is not so and the easy way to prove to someone that the EGC is not required for GFCI operation is to show that it is not connected to the sensing circuitry.

.
 

flashlight

Senior Member
Location
NY, NY
Occupation
Electrician, semi-retired
100224-1031

...

A well designed GFCI should not be tripped by any perturbations of the source voltage. A poor design may be susceptible to rapidly changing voltages on the source voltage.
.

Gar,

These GFIs are incorporated in the male end of the hair dryers and
are much smaller than a typical gfi receptacle and probably not top quality. Nonetheless, several different ones will trip under these conditions.

Thanks all for your thoughts on this and will keep you posted.
 

jahilliard

Senior Member
I am familiar with this problem because it happens at my house with the bathroom GFCI protected outlet when I use my personal clippers to cut my hair. Instead of walking into the other bath every time to reset the GFCI, I simply don't turn the clippers off with the switch on the clippers. It never trips as long as I simply pull out the cord while thy're still on. May not be THE solution you're looking for but may give some insight to the actual problem.:D
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
100224-2102 EST

jahilliard:

Your comment is interesting. This type of information is vital.

Do you have a way of knowing the manufacturer of your GFCI and its possible age? The only GFCIs I have opened are recent units. I expect that older GFCIs may be more susceptible to line transients than the newer ones, and I have really only done some testing for transient problems with the Leviton.

When one breaks an inductive circuit while current is flowing that inductor will try to do whatever is necessary to maintain that same current flow. What this means is that when you are supplying a current to an inductor and then open the source, the source becomes a very high impedance. What happens to the voltage across the inductor if you have 1 A before the circuit is opened and then after the circuit opens the the external impedance becomes 1 megohm? For 1 A and 1 megohm it will try to go to 1,000,000 V. But the gap in the opened circuit will arc over much below this high voltage, but still likely a large voltage.

This is how switching transients are generated.

What is interesting in your case is that breaking the circuit with the internal switch trips the GFCI, but unplugging the cord does not. It is possible the switch opens quicker and produces a higher voltage discharge than when you unplug. You might try a faster unplugging and see if this will trip the GFCI. If not than the difference may result from the different impedances that exist based on the switching location.

The reason for wanting to know about brands and date of manufacture of GFCI devices is that if this information is generally known, then persons troubleshooting false tripping problems can better understand what to do to try to solve the problem.

In your case it would be interesting to know if a Leviton 7899 in place of your current GFCI would eliminate the false tripping. To date any attempts I have made to try to trip the Leviton have not caused it to trip.

.
 

PetrosA

Senior Member
I'm going to guess it's transients from the clippers' magnetic motors backfeeding into the GFI'd neutrals as well. Maybe testing the clippers with the hair dryers on would verify this? The customer will have to decide whether it's cheaper to upgrade to better clippers or run a new line. Obviously, cordless clippers would solve the problem, but they might not be heavy-duty enough for cutting (I don't recall seeing full-sized clippers in a cordless version).

Are there maybe better grade GFI cord ends that could be put on the hair dryers? I'm guessing that installing GFI recepts. and swapping the cord ends to non GFI would violate their UL listing.
 

mxslick

Senior Member
Location
SE Idaho
I have to jump in here and say that hairdryers do not have GFCI's as we know them.

They have an "Immersion cutoff" device which is supposed to cut the power if the hair dryer falls into liquid. Most hairdryers of that kind have a two-prong plug (which is the first clue that it's not a GFCI) and the power cord's third wire (which would be an egc normally) or shield wire is the immersion devices' "sense" line. ANY sensed voltage (above whatever threshold the device is designed for) will trip the IC device. It DOES NOT look at current imbalance at all.

My brother-in-law owns a salon and I will see if he has any old hairdryers with those plugs on them that I can take apart and give some answers to us all.

The cause of the OP's problem is simple, the clippers are inducing voltage onto the AC line which is causing the hair dryer's IC to react.

Try placing either the clippers OR the hairdryers on the surge strip, and put the other straight into the receptacle. I'm willing to bet that will solve the problem.
 
Status
Not open for further replies.
Top