GFCI trip point

Status
Not open for further replies.
ptonsparky

ptonsparky

Tom
Location
NE (9.06 miles @5.9 Degrees from Winged Horses)
Occupation
EC - retired
Related to 'VFD & GFCI'

I decided to test the trip levels of my small GFCI protected splitter. I've had it for years,+ a couple more, and typically use it with a small genset or locations that have no GFCI readily available.
My poor mans method is to use my Fluke 87 on the ma range and catch the MAX values of a Neutral to EG fault via the meter. I then compared it to a newer GFCI receptacle on my work bench.
100 msec. Yes, I know pressing the Test Button occasionally is all that I am required to do.

Splitter: 9.6, 7.2, 4.8, 3.2, 7.2
GFCI: 4.8, 4.8, 5.6, 4.8, 4.8

The old splitter is headed to the scrap barrel but I am wondering if the new self testing GFCIs actually check the levels at which it trips? ​
 
kwired

kwired

Electron manager
Location
NE Nebraska
ptonsparky said:
Related to 'VFD & GFCI'

I decided to test the trip levels of my small GFCI protected splitter. I've had it for years,+ a couple more, and typically use it with a small genset or locations that have no GFCI readily available.
My poor mans method is to use my Fluke 87 on the ma range and catch the MAX values of a Neutral to EG fault via the meter. I then compared it to a newer GFCI receptacle on my work bench.
100 msec. Yes, I know pressing the Test Button occasionally is all that I am required to do.

Splitter: 9.6, 7.2, 4.8, 3.2, 7.2
GFCI: 4.8, 4.8, 5.6, 4.8, 4.8

The old splitter is headed to the scrap barrel but I am wondering if the new self testing GFCIs actually check the levels at which it trips? ​
Click to expand...
AFAIK the new "self testing" units don't actually perform the test that the test button performs. The test button actually puts a load (resistor) between a line terminal and a load terminal - this unbalances the sensing coil and should trip it.

The "self test" feature is only testing the electronics to verify they are still in working condition, and from my understanding is constantly monitoring this. Whether there is any failure modes it can miss - IDK, but there is more certainty then before they added this feature.
 
G

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
180323-1547 EDT

From memory the test resistor is 15,000 ohms or 8 mA at 120 V. Possibly it was 18,000 ohms or 6.7 mA, but this seems too close to the trip spec.

The trip characteristic of a GFCI is time dependent. Shorter times require a higher current.

From http://www.bender.org/documents/LifeGuard_datasheet_NAE1082321.pdf
6 mA Trip Level with Inverse Time CurveThe standard LifeGuard model features a 6 mA trip level. The units trip in accordance withUL943, the standard for personnel protection. The minimum response time is 25 ms at leakagecurrents of 250 mA and above. A built-in inverse time curve helps to prevent nuisance trippingissues, particularly in systems with variable frequency drives (VFDs).Models up to 100 A using system voltages 240 V line-to-line and below are listed as Class Aground fault circuit interrupters.
Click to expand...

I used to be able to find a reference that showed the trip time curve. My present Google search brings up all sorts of useless junk. Very superficial discussions.

The ON datasheet does not show the triptime curve. See http://www.onsemi.com/pub/Collateral/NCS37010-D.PDF

.

.
 
I

Ingenieur

Senior Member
Location
Earth
not sure where the 4-6 mA comes from
it's a curve
depending on ground path, etc it could be 100 mA
but as long as fast enough no injury

you shorted N-gnd thru a meter set to mA?
what was the voltage?
 
G

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
180323-2043 EDT

ptonsparky:

You performed an invalid experiment to determine your GFCI trip point.

You are lucky you did not burn out the Fluke internal fuse, about $15.

To determine the trip current of your GFCI connect a 50 k Ohmite or AB 2 W pot in series with a 5.6 k 2 W resistor. Maximum RMS current at 120 V is 120/5.6 = 21 mA.

Set the adjustable resistor to maximum resistance, current at 120 V is about 2 mA. At 120 V a current of 2 mA is 60,000 ohms. This should not trip the GFCI.

This series resistance should be only connected momentarily to create a current because when no trip occurs the current could be too high for continuous application to the resistors. A 2 W 50,000 ohm pot can tolerate about 6.3 mA on a continuous basis.

Set to a resistance value, then momentarily connect the variable resistance between a GFCI hot output terminal, and a neutral input terminal, about 2 mA.This should not trip the GFCI. Gradually adjust the resistance, and retry until trip occurs. Measure the resistance and calculate current using the input hot to neutral voltage.

.
 
Last edited:
I

Ingenieur

Senior Member
Location
Earth
try this

get a lamp with 20 W bulb
a 500 Ohm R, 5 W or so
on the neut side: wire R and meter in series
connect to gnd screw (or use a 3 p plug reconfigured)
plug lamp into gfci
turn it on

max i ~ 100 mA
 
ptonsparky

ptonsparky

Tom
Location
NE (9.06 miles @5.9 Degrees from Winged Horses)
Occupation
EC - retired
gar said:
180323-2043 EDT

ptonsparky:

You performed an invalid experiment to determine your GFCI trip point.

You are lucky you did not burn out the Fluke internal fuse, about $15.

To determine the trip current of your GFCI connect a 50 k Ohmite or AB 2 W pot in series with a 5.6 k 2 W resistor. Maximum RMS current at 120 V is 120/5.6 = 21 mA.

Set the adjustable resistor to maximum resistance, current at 120 V is about 2 mA. At 120 V a current of 2 mA is 60,000 ohms. This should not trip the GFCI.

This series resistance should be only connected momentarily to create a current because when no trip occurs the current could be too high for continuous application to the resistors.

Set to a resistance value, then momentarily connect the variable resistance between a GFCI hot output terminal, and a neutral input terminal, about 2 mA.This should not trip the GFCI. Gradually adjust the resistance, and retry until trip occurs. Measure the resistance and calculate current using the input hot to neutral voltage.

.
Click to expand...
Why would I blow a fuse when I am causing an imbalance for the GFCI to detect?
From the grounded conductor to the eguipment ground.
A Basic check when you have GFCI trouble on new installation is to look for that inadvertent load side bond. A 5 ma fault on the grounded conductor should trip the same as the hot side. All the GFCI looks for is an imbalance.

Why do the math when I have meters that tell me the current? I could connect a pot in series with the meter and slowly turn it until the GFCI trips but I don’t think that’s how it happens in real life.

The part that makes me question my test now is the trip level of my basement GFCI.
 
G

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
180323-2159 EDT

ptonsparky:

I am the one that mentioned the $15 fuse.

An ammeter using a "shunt" (low resistance resistor, an electrical instrumentation term) uses a very low resistance in series with the load current to be measured. For many shunts this will be a resistance such that at full rated current of the shunt the voltage drop will be 50 mV.

The shunt resistance is so low that peak current thru the shunt will be extremely high when not limited very much by the total loop impedance. A 1 A shunt at 50 mV is a resistance of 50 milliohms. Apply 120 V to this shunt and I = 2400 A. External loop resistance of the 120 V source will help reduce this maximum current some.

Your GFCI tripping responded fast enough that the Fluke 87 fuse I^2*T was not exceeded and hopefully the 87's shunt was not damaged.

The single number trip current rating of a GFCI is based on the lowest current applied over a long time, several seconds, that will not trip the GFCI. There is a lower and upper limit for this spec. The value 6 mA is a sort of nominal value.

Your test with the 87 produced a very large trip current, but this was variable in its integrated value as seen by the GFCI electronics, because your time of application was random relative to the the AC line voltage at the instant you made contact. The magnitude of the 87 reading varied. Time to trip of the GFCI is variable based on current and other factors.

The 87 current reading was not a measure of the trip current of the GFCI. This reading was a function of how the 87 makes current measurements, the duration of the current, the waveform, and how the GFCI trips.

A scope and the correct kind of shunt would allow you to see the the kind of current you produced. Any current above of what the GFCI considers its trip point, including timing, will trip the GFCI. But that larger current is not the trip current of the GFCI.

.
 
Last edited:
ptonsparky

ptonsparky

Tom
Location
NE (9.06 miles @5.9 Degrees from Winged Horses)
Occupation
EC - retired
Gar,

My fault is from the Grounded Conductor, Neutral, or White wire, to the Equipment Ground.

My fault is NOT from the Ungrounded, Hot, or Black wire to the Equipment Ground. That would possibly blow the $15 fuse or worse cause damage to the meter. Yes, I know you mentioned the fuse.
 
ptonsparky

ptonsparky

Tom
Location
NE (9.06 miles @5.9 Degrees from Winged Horses)
Occupation
EC - retired
Ingenieur said:
not sure where the 4-6 mA comes from
it's a curve
depending on ground path, etc it could be 100 mA
but as long as fast enough no injury

you shorted N-gnd thru a meter set to mA? Yes
what was the voltage?
Click to expand...

Voltage N-gnd is 8.5 mV at my basement GFCI. Load side. My shop work bench was probably similar. I did not check. Both are in conduit less than 8’ from the SE.

Note to self...remove router from work bench GFCI. Posts are lost during testing.
 
ptonsparky

ptonsparky

Tom
Location
NE (9.06 miles @5.9 Degrees from Winged Horses)
Occupation
EC - retired
My test results from the first post were not the only ones I did. Originally the splitter was plugged into a work bench GFCI protected outlet. IDNR the 100ms results for those. The bench GFCI always won the horse race to tripping.

I moved the splitter to the office and tested it there to get the results I posted. I also have 1ms Max values but they, like the 100ms, are all over the place. Roughly 180 times the magnitude of the 100ms values. Distance to the office recept is an additional 80'+-.

The bench GFCI is very consistent in its max values with the 1ms values only 22 times the magnitude.

Based on the previous readings I will buy a new splitter and compare.



The basement GFCI will have to wait until later. I have code class today.
 
G

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
180324-0943 EDT

ptonsparky:

My post last night has an error in the way I worded it that I will correct later.

That you had a much lower source voltage in your experiment was greatly to your benefit. But it still is an incorrect way to run the test. You need to do the test approaching the trip point from a lower current. More on this later.

.
 
Status
Not open for further replies.
Top