Got a service (warranty) call from an old customer. Breaker was tripping, they swore up and down that nothing electrical had been moved, opened, changed, etc.

Breaker won't stay on, and dims the lights for a split second before it trips. Sounds like a bolted fault.

An hour and a half later, I narrow it down to this:



How was 12 Ohms acting like a bolted fault?

Was the resistance decreasing with line voltage (rather than the 3VDC test voltage in the pic)?

I thought UL insisted that devices die in a safe manner...

Would you consider this safe?

Corrosion from having a small hot-tub plugged in (no in-use cover... Job was before 2002, and they didn't tell me about the hot-tub) was the main culprit. They're getting an in-use cover today!

Hey 66:
Sometimes the good old GFI's fail like you said. THe internal corrosion prevents the "trip" mechanism from opening; that's what I heard from the Hubbell sales rep. The WP in use covers don't stop "Condensation". If you can, see if the wall the box is set in is insulated; whatever you can do to stop condensation may save you another service call. Hey, how long is the warranty period???
John

Could be, Virgil.

The GFI electronics will most likely be fed via a full-wave silicon rectifier assembly, each of the two diodes in the path needing about 0.6V to fully saturate them. Add the extra PN junctions in the circuitry, and 3V may not be enough to give you a true indication of what the resistance will be when 120V is applied.

Another point: Did you try reversing the meter probes to see if you got the same reading? Don't forget that with the DMM you're testing with DC.

I'll side with Paul on this. At least, if you find a problem at 3 volts, it's safe to bet that it will likely be at least a bad at 120V.

Troubleshooting can give some insight into electrical systems that can't always be found in the books or with construction duties.

John,
I realize that the law only requires 1 year of warranty, but the GFCI Receptacle was installed less than 19 months ago (4-10-01), and I'd really like to think my work should last a lot longer than that! Besides, the job wasn't punched-out until 3-12-02, so then it would still be a warrantee call by that. Kind of a grey area, so I give the benefit of the doubt to the customer.

It was one of the rare instances that I used an Eagle GFCI... Leviton is my main brand, mainly due to the fact that my supplier distributes Leviton. I've never had this problem with a Leviton brand GFCI receptacle, or any other device for that matter.

The wall is insulated, would caulking help to seal around the box? It is flush mounted.

The plug on the hottub is too big to fit under the in-use-cover. If it ain't one thing... it's another!

Pauluk, good call, I always check with red on hot and black on neutral or ground (or the other hot)... I'll reverse the leads and see what I get...

*time passes*

...Trying various different combinations, line, load, and reversing the leads, I still get a reading settling down quickly to 12 or 13 Ohms. The DMM has read briefly from 120 to 250 Ohms for a moment before settling to 12 or 13 Ohms on each test.

How little resistance is this thing giving to trip the breaker instantly? Is that safe?

What if it were in an old home with an over-sized OCPD? Would it heat up internally and be a fire hazard?

Is UL aware that they can fail in this manner?

Am I just freaking out again?



[This message has been edited by sparky66wv (edited 11-06-2002).]

Virgil:
It's a real nice move to stand behind your work... there should be more like you & me!!

The thing I'm trying to say is....condensation forms from two different temperature extreemes. If the GFI/box is on a cool or cold wall, the condensation will build up in and on the device. I don't know how else to explain it. If you can get the WP cover with the gasketed door, and close it, and seal around the cord entry it should help.

Did you think about sending Eagle an e-mail at there web site??? They may send you something ......

John

Hmmm... Didn't think of that. I guess Eagle should be the first to know!

Condensation is a consideration, but I think being subject to driving rain while the cover was open did it in over the last few months.

I put on a cheap ($10.30) in-use cover rather than the Steel-City one with the gasketed cover... Guess I shoulda went with the better one... I will if (when) it happens again!

I've seen a lot of bubbles covers lying in pieces within a year, and this one is right in a walkway. I'll probably be replacing it within a few months.

As far as standing behind my work, I once worked for an EC whose excuse for shoddy work was "if it lasts a year, it'll be a service call". That left a bad taste in my mouth. I only worked for him about three weeks. Heck, a few months ago, Joe and I tossed the idea around for a lifetime warranty plan, including routine maintenace to check terminals and splices. For an extra price, of course. He even asked the people in the contruction management office if they'd go for such a plan... No takers.

66:
Now all this time I thought this was INSIDE (Jacuzzi tub). Damn!

John

John, after reading the thread again, I realized I didn't actually point out that the receptacle was on an outside wall for a deck.

Sorry!

Follow-up:

Cooper Wiring Devices called today regarding my email. They are very interested in the nature of this unusual and unique failure, and would like me to send the device to them for further study.

My first instinct is to do my civic duty and send it to them...

But...

With narrowed eyes and a snear on my lips, I think, what is this thing worth to them if they are this interested?

Only half kidding... I'd hate to throw away money by doing a good deed! You know, good guys finish last...

Nah, I'll just send it to them and they can do what ever good deed their hearts desire... a case of GFCIs would be nice!



What would you do?

Virgil:
I'd send it to them with a memo stating the facts that you put in your e-mail....and the fact that you made a warranty (good faith) replacement. Most manufacturers WILL reimburse you for a reasonable labor rate for WARRANTY of there items. They may send you "goodies", samples, etc. The best sure fire method is to get the local rep involved.
If he knows you buy his products, they usually do what they have to, to keep you happy.

A case of GFI's should not be a problem. If the GFI caused bodily injury, think about how many truckloads of GFI's a "good" PI lawyer could buy with the lawsuit proceeds!!!

You provided them a service, hell it should be worth something! Still say you're a good guy for the warranty thing.

Later.
John

John,

Did just as you said. Mailed it in with a copy of my email and a brief description explaining that it was a warranty call.

I'll see what giving up a $9 GFCI with $1.50 in postage and $1 for a packing box will get me!

I can always gloat that a life could be spared because of what Cooper learns from it.

As a final check before mailing it off, the resistance increased to 14 ohms! It was a damp, cold and rainy day, but so was the other day!

Do you think it might help to use a faceless on an inside wall to keep the electronics within conditioned air spaces?