ARC Fault puzzle

Status
Not open for further replies.

craig65

Member
I need some help with this one...
I have an Arc fault breaker problem. At an assited living facility several rooms on the west side of the building the arc fault breakers trip, with nothing being pluged in to the outlet. They have been reset and have stayed on anywhere from several hours to several days before triping again.(Still with nothing being pluged in). I have rung out the circuits to make sure nothing got crossed at the outlet and at the panels. I have also replaced the breaker in some cases..... but that wasn't the problem either :confused: :confused:

The manufacturer of the breakers keep telling me it's wired wrong.I have wired and re wired these outlets and breakers three times. Is there something I'm missing??
 
Have you megged the wires? Remember that the AFCI has a 30mA ground fault circuit and a grounded to grounding short can cause it to trip, however that normally only happens when there is a load but can happen without a load if the panel where the AFCI is installed is not the panel with the main or system bonding jumper.
 
craig65 said:
They have been reset and have stayed on anywhere from several hours to several days before triping again.(Still with nothing being pluged in).

How can you be sure nothing is being plugged in during that time. My guess is something is being plugged in that is causing the problem. Is the lighting on the circuit???
 
Dennis Alwon said:
How can you be sure nothing is being plugged in during that time. My guess is something is being plugged in that is causing the problem. Is the lighting on the circuit???

After all, you state this is in an assisted living center. So that means there's....

1. Staff. And 99% of them don't realize you're there looking for a problem with this circuit, they just know it doesn't work. Nor will they communicate with anyone involved with the problem.
2. The residents. No offence to the elderly, buy some can't remember what they were doing 5 minutes ago.
3. Visitors. See #1, and change number to 99.999%
 
Have you disconnected the feed to the first outlet in the circuit to see if it holds then???? If so work into the circuit until you find the problem in the circuit.
 
If they have a capacitive leakage limit, how close are you? Distributed R's and C's could add up.

You could also ask the manu. to be more specific about, "Of the infinite number of ways to wire things incorrectly, which ones should you be looking for?"
 
If the arc faults are next to each other in the panel try seperating them. I have had a tripping problem with Siemans arc faults and this fixed the problem. Apparently AFCIs generate enough heat when next to each other in a panel to cause tripping.
 
From this link
http://www.cpsc.gov/CPSCPUB/PUBS/afcifac8.pdf
I see that these things are supposed ignore "normal" arcing. This implies that the AFCI circuitry decides that a "long duration" arc is abnormal and therefore should be stopped.
There may be other arc qualifying factors as listed in UL1436.

Switches are supposed to have a snap-action to limit the arcing time and preserve contact life. I guess a switch with a weak spring could be teased into excessive arcing and so fool the AFCI.

Possibly any appliance/motor that is failing and not having the snap-action contact opening could cause long duration arcs. Maybe high intensity discharge lamps, also.
 
don_resqcapt19 said:
Have you megged the wires? Remember that the AFCI has a 30mA ground fault circuit and a grounded to grounding short can cause it to trip, however that normally only happens when there is a load but can happen without a load if the panel where the AFCI is installed is not the panel with the main or system bonding jumper.

this is the step I would now take
 
I had a very similar problem recently. AFCI tripping with
nothing on the circuit. Found neutral was shorted to
ground. Started at the middle receptacle, and disconnected
wirenuts to isolate it to a neutral ground short in a cable
between the 7th and 8th receptacles on the circuit.
Replaced the cable, no problem. I am not 100% sure
why the ground fault was sensed with no load - my guess
is there were other heavily loaded circuits on the same
panel, and there was some voltage on the neutral because
of current flow elsewhere.

If this happened again, I would look for a ground fault.
 
AFCI's, product recalls, salespeople

AFCI's, product recalls, salespeople

Two stories:
OK, three. . .

I worked at a place that bought a machine that was supposed to mimic a microprocessor. Many problems, many calls to the company, and the salesman denied, denied, denied.
Then one day they came out with version B which they said solved all the problems in the first version (the same one which the salesman denied had any problems)!!!
Some people are sometimes bothered by reality, but not these guys.

A housewife/customer with a heavy French accent had drain lines put in, in the backyard, and they didn't work properly. She left many messages, to no avail. As long as I was already there, I offered to call for her.
I left one message on the landscaper's machine: "the drain line you put in was suppposed to be pitched downward 1" in 10'; instead it is either level or pitched upward in places."
They called her back before I left, and I wasn't there that long.

Some car makers have unannounced recalls, for their "special" customers. I saved a neighbor lady $1800 on an engine rebuild by hearing about an engine design defect through a "grapevine", an auto grapevine.
Maybe customers who buy lots of a thousand of these AFCIs are already enjoying call-back-free workdays.

OK.

I would like to call an AFCI manufacturer and see how many barriers I have to jump over before I talk to the guy on the factory floor who designed the thing (if he still works there).
Let's see if he can dazzle me with brilliance (or the other option).
Some companies play hardball, and you'd have to subpoena the guy.

Let's say I want to design an AFCI.
First I must take many scope photos of arcing and group them into "normal" and "abnormal". If there are clear differences then my job is somewhat easier. If not, if there is considerable overlap, then I must make a statistical or arbitrary decision as to what is "normal."
Then to test this gadget I would need to build a versatile arc generator, not a trivial job, and expose my design to all these arcs.
If the AFCI could flip a coin, it would already be right 50% of the time, but I'd hope it would approach 100% in correctly classifying normal/abnormal arcs.

Once a month someone would push the test button and the AFCI, which is by now the size of a breadbox, would regenerate all these arcs and see if it classifies them correctly some minimum percent of the time.
NOT!

So the test button MUST take some shortcuts. It might just hook up a resistor. If I get a chance I'll take one of these things apart someday.

IMHO:
I'd think almost any company who makes an AFCI tester can afford to more nearly simulate an arc according to the UL standards than the manufacturer of some little box mounted in the wall.
If the AFCIs had no problems, these AFCI tester companies would go out of business.
And it is in the financial interest of the AFCI makers to tout the wondrous things that the test button will tell you (but see first story above).

I know! They need an AFCI with a synthesized voice and a tiny loudspeaker that says "look 10.4 feet north-northwest, in back of that ugly wall painting."
Yeah, that's it. . .

Here's an excerpt off the Internet from someone's lawsuit over implied merchantability. I'm not a lawyer, but I think the AFCIs slip right past this without a scratch.

\/

"Whether a breach of the implied warranty of merchantability has occurred is a question of fact. However, before the issue of a breach can be addressed, there must be a finding that an implied warranty of merchantability applies to the item sold. That is a question of law.
2. There is an implied merchantability for goods sold if the seller is a merchant with respect to goods of that kind. To be merchantable, the goods must be fit for the ordinary purpose for which they are used.
3. To establish a breach of the implied warranty of merchantability, the buyer must prove the ordinary purpose of the type of goods involved and that the goods sold were not fit for that purpose. The buyer must show that the goods were defective, that the defect was present when the goods left the seller's control, and that the defect caused injury.
4. A buyer of a used car has a right to expect that it will not turn out to be completely worthless but cannot reasonably expect that the car will be the finest of all possible goods of that kind. The implied warranty of merchantability varies with the particular car. A late model, low mileage car, sold at a premium price, is expected to be in far better condition and to last longer than an old, high mileage car.
5. An implied warranty of merchantability warrants the operation of major components that are necessary for the car to operate, such as the engine and transmission, and it is the responsibility of the buyer to ensure that the components incidental to operation are in working condition."
 
Dennis Alwon said:
How can you be sure nothing is being plugged in during that time. My guess is something is being plugged in that is causing the problem. Is the lighting on the circuit???
The rooms are unoccupied
the only thing pluged in the room is a rechargable flash light same as in all the other wings. It is only in 5 rooms all in different panels feed from the same MDP. the flashlignts all charge when I reset the breaker (s) for several hours (days). Some trip before the others but at one point in time or another they all end up triping
The lighting is on it's own arch fault with no problems.
 
So you have 5 rooms on 5 circuits that are tripping after a period of time with the flashlight charger plugged in.

I would unplug one flashlight and see if that solves the problem for that room. If it does then your flashlights are the problem. I would also double check every connection but it seems you did that.
 
Status
Not open for further replies.
Top