afci breakers

Status
Not open for further replies.

lexxx

New member
Location
arizona
why will one manufactuers afci breaker trip when tested at a recptacle with a hand held tester and another manufacturers breaker won't. I have tested one manufaturer with a hand held tester for 2 years.
 

charlie b

Moderator
Staff member
Location
Lockport, IL
Occupation
Retired Electrical Engineer
I didn't know there were hand-held AFCI testers that you plug into a receptacle. How do you know that it is a difference in manufacturer? By that I mean did you successfully perform the test, and then have the breaker replaced with one by a different manufacturer, and then run the test again at the same receptacle? I guess I am suggesting that it might be a wiring issue, rather than a breaker issue. But not knowing how the tester itself would operate, I can't be any more helpful.
 

roger

Moderator
Staff member
Location
Fl
Occupation
Retired Electrician
I don't have an answer as to why one manufactirers breaker will trip and anothers won't but, it really doesn't matter, hand held (remote) type testers are only indicators and shouldn't be relied on for approval. The only recognized test is the test button on the device or breaker.

Roger
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
lexxx said:
why will one manufactuers afci breaker trip when tested at a recptacle with a hand held tester and another manufacturers breaker won't. I have tested one manufaturer with a hand held tester for 2 years.
Because the actual arc signature information that the AFCI looks for is proprietary and each manufacturer does it differently. There is no way that the tester manufacturer can know what to simulate, unless the tester is made by the breaker manufacture, but even then it would only work with that brand of breaker.
 

ELA

Senior Member
Occupation
Electrical Test Engineer
don_resqcapt19 said:
Because the actual arc signature information that the AFCI looks for is proprietary and each manufacturer does it differently. There is no way that the tester manufacturer can know what to simulate, unless the tester is made by the breaker manufacture, but even then it would only work with that brand of breaker.

Don,
This seems very strange to me. I have heard this before and it makes me wonder about UL1699 testing on pg 3 of this doc:

http://www.ul.com/regulators/afci/AFCI_scenarios020502.pdf


If these devices have to meet certain UL test requirements doesn't that mean there are certain criteria that each manufacturer must meet?
If so why couldn't someone create a tester to produce those conditions?
It might not be easy but seems like it has to be doable?
 
L

Lxnxjxhx

Guest
there are certain criteria that each manufacturer must meet?

there are certain criteria that each manufacturer must meet?

I can't find it now, but Square D seemed to be saying that they made improvements in their AFCI design, which makes me wonder what spec it now falls under.

UL stuff reads like a patent claim (. . .a means to detect an arc. . .) in order to cover as many cases as possible.
I suppose manu's have a lot of wiggle room within a spec., but if UL names actual numbers, I'd say all manu's need to meet those numbers or reference a new spec. or lobby to change the spec.
I guess UL just wants to detect arcs reliably and it's up to each company to decide how best to do that.

A real tester would generate many good arcs and many bad ones and rate the performance of the AFCI. This would mean filling in each category below:

Trips/doesn't trip with no arc?
Trips/doesn't trip with a bad arc?
Trips/doesn't trip with a good arc?
Trips/doesn't trip with temp changes plus all of the above?
Trips/doesn't trip with power glitches plus all of the above?

This would be a whole week of testing with a lab full of equipment at NIST.

Any AFCI is so complex that almost any handheld tester needs to make assumptions and take shortcuts.
If you find an AFCI that gets a score of 100% on all the categories above hang onto it. Same if you find a tester that can reliably test this perfect AFCI.

And I used to be able to set the breaker point gap with a matchbook cover.
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
ELA,
The UL standard does not say how the device has to work, just the conditions under which it must open the circuit. I expect that there are many different things or combinations thereof that the AFCI can look at to "detect" the arc. The tester would have to simulate the exact combination of things that the AFCI is looking for. Without knowing what the AFCI is looking at, I don't see how you can make a functional tester.
 

ELA

Senior Member
Occupation
Electrical Test Engineer
don_resqcapt19 said:
The UL standard does not say how the device has to work, just the conditions under which it must open the circuit. .

Exactly!
The conditions under which it must function. This implies a test specification to me.

I simply do not buy the argument that a tester cannot be made. Each manufacturer may use a different software algorithm but they all must be able to detect an arc when exposed to one via a test instrument that produces the type of arcs that I believe would be defined in UL1699?

When a third party tests these devices do you believe that all they do is push the test button on the device? I do not.
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
ELA said:
When a third party tests these devices do you believe that all they do is push the test button on the device? I do not.
No, they create an actual arc...something a tester cannot do. I am sure that there are any number of factors that the AFCI firmwear looks at to decide if there is an arc. A plug in tester would have to know exactly being looked at to work.
 

roger

Moderator
Staff member
Location
Fl
Occupation
Retired Electrician
Texhunter7 said:
Every inspector in Colo. has a AFCI tester and they use them all the time on AFCI circuits to test them at the plug.

And if they use the results of these testers to pass or fail an inspection they are in the wrong.

Roger
 
They do that all the time and when they trip the AFCI thats when they check for the smoke detectors to drop. They sell the same tester at Home Depot here for around $25-$30. It's called a ARC FAULT tester almost the same as the GFI tester.
 

ELA

Senior Member
Occupation
Electrical Test Engineer
Im with Ideal on this one.
These AFCI are "shaky" devices that use independent software algorithms to recognize the arc signature. Some do it better than others.

They are getting better over time but until that time:

To back their design, their push buttons duplicate the signature "that they want to see".

I stand by my contention that a "tester" is totally possible based on a common signature to be recognized. It would probably cost more but totally possible.
 

roger

Moderator
Staff member
Location
Fl
Occupation
Retired Electrician
ELA said:
I stand by my contention that a "tester" is totally possible based on a common signature to be recognized. It would probably cost more but totally possible.

That's fine and more power to you and the people that will create this tester but, the fact of the matter is, there is not a true tester available at this time.

Roger
 

ELA

Senior Member
Occupation
Electrical Test Engineer
Never said it was available, just that it is possible. It should have came before AFCI devices were mandated.

It will come, it has to.
Since when, in the name of safety ,would you rely on a device to test itself?
 
Status
Not open for further replies.
Top