Remember that a UL label does not ensure safety

[don_resqcapt19]

Bill,

You'd be wrong in replacing a GCFI that took a few seconds to trip with the test button as up to 5.6 seconds is allowed.

True, but I really doubt that I would hold the test button in that long. Most seem to be almost an instantanous trip when you push the button.
Don

 

[wptski]

Bill,

True, but I really doubt that I would hold the test button in that long. Most seem to be almost an instantanous trip when you push the button.
Don

Don:

It's a "press to test" not "press/hold to test". Or does it say "press/release? I think it's just "press". Nothing really clear including the specs!

 

[ELA]

I appears that the UL standard provides a guideline for a worst case trip time.
From the two links I supplied earlier you can see that the "Bender" unit takes advantage of the curve and may take as long as 5 seconds to trip at 6ma.

My guess is that there design they purposely slow down the trip time at low level currents to provide a greater immunity to noise. This units looks pretty industrial strength and may have to deal with a higher level of noise than a residential GFCI might have to.

In the other link of the LM1851 (most likely used in residential GFCIs) they indicate a typical trip time of 18 ms at the 0.25A fault level.
I did some calculations based on the information given on that data sheet and got a trip time (at 6 ma) of roughly 0.45 seconds under the same fault conditions. NOTE: they take into account both faults from H-G and N-G causing more variability in the trip time as well..

From this I would guess that most residential type GFCIs will trip in well under one second. As the UL graph indicates it could take much longer in other types of GFCI's if so designed.

 

[wptski]

The UL plot line means anything that falls on the line is allowed and 5600ms trip time with 6ma is on that line.

If so, what about the 6-8 residential GFCI's that tripped at 28ms with 7.5ma? They would be more prone to noise, going by what your saying.

Here's a typical Pass&Seymour spec sheet: http://www.passandseymour.com/pdf/E03.pdf. The maximum trip time and minimum current are the same even for a hospital grade GFCI.:-?

 

[don_resqcapt19]

Bill,

It's a "press to test" not "press/hold to test". Or does it say "press/release? I think it's just "press". Nothing really clear including the specs!

I don't think the instructions say anything about how long to hold the test button, but the simulated ground fault only exists as long as you hold the button.
Don

 

[don_resqcapt19]

ELA,

In the other link of the LM1851 (most likely used in residential GFCIs) they indicate a typical trip time of 18 ms at the 0.25A fault level.

That is 2/3s of the maximum permitted trip time for that fault level.
Don

 

[wptski]

Bill,

I don't think the instructions say anything about how long to hold the test button, but the simulated ground fault only exists as long as you hold the button.
Don

Don:

Hmmm, I guess that you hold it in till something happens. A Fluke T+ Pro tester's GFCI test times out after 6.5 seconds, I think and the SureTest max trip time is 6.6 seconds, I think again.

EDIT: Look at what this page shows for current and trip times: http://www.fda.gov/ora/Inspect_ref/itg/itg22.html. The GFCI receptacles that tested at 28ms trip time were found to be 20A Pass&Seymour brand.

 

[wptski]

I never heard back from Pass&Seymour or Leviton.

I received two P&S 2095-I GFCI receptacles today. I connected one as I did with the Leviton. The Ideal SureTest read 30ms trip time and the Fluke 1653 read 4.7ms trip time compared to 132ms and 66ms on the Leviton.

I also found this file which has a graph showing average GFCI trip times: http://www.fireshield.com/04_0301TechNotes.pdf