Meggering with a GFCI in the circuit

Status
Not open for further replies.

wptski

Senior Member
Location
Warren, MI
I know this has been discussed before here and I was involved in one of the threads at least.

I first tried a device called a Shock Buster a portable GFCI that you just plug in first. Using a AEMC 1060 with the device plugged(tripped) into an extension cord, stating at 50V and trying all possible connections between L-N-G. I got max resistance in 50V, 100V and 250V but at 500V the voltage dropped to zero as the resistance started to drop. It was actually resetting the device.

I didn't know at first that I had a few PASS&Seymour 20A duplex GFCI's. I first used a Ideal Suretest which has a GFCI test function(I know the test button is the only true test) before starting and noted the readings. Using all possible L-N-G connections on the input/load side all the way to 1000V and got the max. resistance in every range. The test button still works and the Suretest shows the exact same tets values. Now, I don't know about PI ten minute test as that might be pushing it????
 
Location
NE (9.06 miles @5.9 Degrees from Winged Horses)
Occupation
EC - retired
I know this has been discussed before here and I was involved in one of the threads at least.

I first tried a device called a Shock Buster a portable GFCI that you just plug in first. Using a AEMC 1060 with the device plugged(tripped) into an extension cord, stating at 50V and trying all possible connections between L-N-G. I got max resistance in 50V, 100V and 250V but at 500V the voltage dropped to zero as the resistance started to drop. It was actually resetting the device.

I didn't know at first that I had a few PASS&Seymour 20A duplex GFCI's. I first used a Ideal Suretest which has a GFCI test function(I know the test button is the only true test) before starting and noted the readings. Using all possible L-N-G connections on the input/load side all the way to 1000V and got the max. resistance in every range. The test button still works and the Suretest shows the exact same tets values. Now, I don't know about PI ten minute test as that might be pushing it????

Why are you leaving the GFCI in the circuit when you test? What are you trying to prove?
 

wptski

Senior Member
Location
Warren, MI
Why are you leaving the GFCI in the circuit when you test? What are you trying to prove?
I'm not leaving it in the circuit, I'm creating a circuit with one. I "trying" to prove that a circuit with a GFCI can be megged without removing it with no damage to the GFCI.
 
Location
NE (9.06 miles @5.9 Degrees from Winged Horses)
Occupation
EC - retired
You are trying to prove that it is safe to apply a voltage up to in excess of 8 times the nominal operating voltage of a GFCI? (1000 volt test range) How are you going to ensure you did not compromise the ability of the GFCI to detect future faults? Lack of visible smoke does not necessarily mean the device will still be functional.
 

meternerd

Senior Member
Location
Athol, ID
Occupation
retired water & electric utility electrician, meter/relay tech
I know this has been discussed before here and I was involved in one of the threads at least.

I first tried a device called a Shock Buster a portable GFCI that you just plug in first. Using a AEMC 1060 with the device plugged(tripped) into an extension cord, stating at 50V and trying all possible connections between L-N-G. I got max resistance in 50V, 100V and 250V but at 500V the voltage dropped to zero as the resistance started to drop. It was actually resetting the device.

I didn't know at first that I had a few PASS&Seymour 20A duplex GFCI's. I first used a Ideal Suretest which has a GFCI test function(I know the test button is the only true test) before starting and noted the readings. Using all possible L-N-G connections on the input/load side all the way to 1000V and got the max. resistance in every range. The test button still works and the Suretest shows the exact same tets values. Now, I don't know about PI ten minute test as that might be pushing it????

If the resistance drops, it's because you exceeded the breakdown voltage of the insulation to ground. Doesn't sound like a good thing to me.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
141201-1336 EST

wptski:

On the input side of a GFCI there is an MOV. Probably rated 150 V, meaning it starts to clip at about 212 volts DC across the MOV.

Specifically Littlefuse specifies a minimum of 216 V and a max of 264 V for a 1 mA test current on a V07E150P.

If both hot and neutral are disconnected from anything, then testing from hot or neutral to EGC should be OK. If either hot or neutral had an unknown path to EGC or earth, then if the Megger available power does not exceed the the power dissipation capability of the MOV, then it is likely that no harm to the GFCI would occur. This assumes the breakdown voltage of the GFCI SCR is not exceeded.

With a Megger voltage above 270 V DC and connected between input hot and neutral, and again that the power capability of the MOV is not exceeded, then Megger will read low, and unlikely to damage GFCI.

.
 

wptski

Senior Member
Location
Warren, MI
You are trying to prove that it is safe to apply a voltage up to in excess of 8 times the nominal operating voltage of a GFCI? (1000 volt test range) How are you going to ensure you did not compromise the ability of the GFCI to detect future faults? Lack of visible smoke does not necessarily mean the device will still be functional.
Yes, I far exceeded what would be the normal voltage for testing. I don't remember who the technical guru here that stated that there is "only" one true test of a GFCI's function ability and that is its built in test button which still works. The conversation included the use of GFCI testers or special functions included on a tester/meter.
 

wptski

Senior Member
Location
Warren, MI
If the resistance drops, it's because you exceeded the breakdown voltage of the insulation to ground. Doesn't sound like a good thing to me.
The resistance started to drop because on the that particular GFCI device, it was being RESET just as if the RESET button was pressed.
 

Smart $

Esteemed Member
Location
Ohio
The resistance started to drop because on the that particular GFCI device, it was being RESET just as if the RESET button was pressed.
That don't make any sense. All GFCI's, that I'm aware of, require manual reset.

Are you certain you are not seeing the effect of the MOV gar referred to?
 

Smart $

Esteemed Member
Location
Ohio
No, because when I applied power to it the LED was already ON. This is what it was: http://www.amazon.com/Tower-Shock-Buster-Portable-GFCI/dp/B001EJDN5U/ref=pd_sxp_f_pt and didn't happen with the PASS&SEYMOUR GFCI.
Trying to understand... Your test setup and results are not exactly published in an explicit manner. After reparsing, it simply sounds like you tested a defective GFCI. ???
...​
I first tried a device called a Shock Buster a portable GFCI that you just plug in first. Using a AEMC 1060 with the device plugged(tripped) into an extension cord, stating at 50V and trying all possible connections between L-N-G. I got max resistance in 50V, 100V and 250V but at 500V the voltage dropped to zero as the resistance started to drop. It was actually resetting the device.​

...​
 

wptski

Senior Member
Location
Warren, MI
Trying to understand... Your test setup and results are not exactly published in an explicit manner. After reparsing, it simply sounds like you tested a defective GFCI. ???

No, the first device, a Shock Buster simply plugs into an extension cord. I powered it and pressed the test button. That one gave good results till I tried 500V where voltage and resistance both dropped but I discovered that it was being RESET. I don't know why it was being RESET.

The PASS&SEYMOUR was wired to a extension cord with the female end removed and terminal lugs installed. I first tested on the input then the load side.

Both still trip with their test button and RESET.

This thread came about because of this thread: http://forums.mikeholt.com/showthread.php?t=166214 where I "might" be doing this with GFCI's installed. Previous threads asking the same question but nobody gave it a real life test just came to the conclusion that GFCI had to be removed before a megger test was to be performed.​
 
Location
NE (9.06 miles @5.9 Degrees from Winged Horses)
Occupation
EC - retired
the first device, a Shock Buster simply plugs into an extension cord. I powered it and pressed the test button. That one gave good results till I tried 500V where voltage and resistance both dropped but I discovered that it was being RESET. I don't know why it was being RESET.

The PASS&SEYMOUR was wired to a extension cord with the female end removed and terminal lugs installed. I first tested on the input then the load side.

Both still trip with their test button and RESET.

This thread came about because of this thread: http://forums.mikeholt.com/showthread.php?t=166214 where I "might" be doing this with GFCI's installed. Previous threads asking the same question but nobody gave it a real life test just came to the conclusion that GFCI had to be removed before a megger test was to be performed.

Test all you want but I would suggest not selling any of those tested GFCIs.
 

wptski

Senior Member
Location
Warren, MI
Test all you want but I would suggest not selling any of those tested GFCIs.
Of course not, I'll use them myself. Keep an open mind here. If you tested a cable and it passed, do you replace or proceed? The test button passes the device, the Suretest passes the device and a Fluke T+Pro passes it as well. This isn't a HiPot test.
 

Smart $

Esteemed Member
Location
Ohio
Of course not, I'll use them myself. Keep an open mind here. If you tested a cable and it passed, do you replace or proceed? The test button passes the device, the Suretest passes the device and a Fluke T+Pro passes it as well. This isn't a HiPot test.
Not a hi-pot test, but you are still substantially exceeding the device's rated voltage when you megger at 500V.

Yes we commonly exceed rated voltage of conductors in megger testing, but not substantially. We test 600V rated conductors at 1000VDC... but the 600V rating is applicable to AC, which has 849V peaks.

1000V/849V = 118%

You are testing a 120VAC rated device at 500VDC. 120VAC has 170V peaks.

500V/170V = 295%
 

wptski

Senior Member
Location
Warren, MI
Not a hi-pot test, but you are still substantially exceeding the device's rated voltage when you megger at 500V.

Yes we commonly exceed rated voltage of conductors in megger testing, but not substantially. We test 600V rated conductors at 1000VDC... but the 600V rating is applicable to AC, which has 849V peaks.

1000V/849V = 118%

You are testing a 120VAC rated device at 500VDC. 120VAC has 170V peaks.

500V/170V = 295%
You must have missed this post back then: http://forums.mikeholt.com/showthread.php?t=136699. Check Post #13 by Brian John.

If it showed anything but max in any range, I wouldn't have proceeded to the next higher range.
 

Smart $

Esteemed Member
Location
Ohio
You must have missed this post back then: http://forums.mikeholt.com/showthread.php?t=136699. Check Post #13 by Brian John.

If it showed anything but max in any range, I wouldn't have proceeded to the next higher range.
I don't read every thread, every post... :happyno:

Showing max in a megger range does not mean we can go to the next higher range. That premise is essentially hi-pot testing: increase voltage until the item(s) under test fail. Yes, some will withstand a higher voltage, some will not. But practically everything on the market that has a rated voltage will fail at some point... which is exactly why they have a rated voltage to begin with.
 

wptski

Senior Member
Location
Warren, MI
I don't read every thread, every post... :happyno:

Showing max in a megger range does not mean we can go to the next higher range. That premise is essentially hi-pot testing: increase voltage until the item(s) under test fail. Yes, some will withstand a higher voltage, some will not. But practically everything on the market that has a rated voltage will fail at some point... which is exactly why they have a rated voltage to begin with.
That's Hi-Pot testing in a lab. You have never used a Hi-Pot tester for anything other than to destroy what's under test? We used a Hi-Pot tester on special motors that turned 38,000 RPM.
 

Smart $

Esteemed Member
Location
Ohio
That's Hi-Pot testing in a lab. You have never used a Hi-Pot tester for anything other than to destroy what's under test? We used a Hi-Pot tester on special motors that turned 38,000 RPM.
It doesn't matter what you call the test if you exceed ratings to a point of failure.
 
Status
Not open for further replies.
Top