Can any of you sparkies explain why this GFCI melted at the white post of the load side? It’s hard to see on this picture, but the plastic separator between the two connection holes is melted and the wire insulation was burned.
Homeowner’s previous handyman moved two existing kitchen outlets (on separate circuits) up into new boxes inside the cabinets and installed a GFCI in each one. The reason was so their new tile backsplash would have no unsightly outlets. He then installed plugmold strips with an outlet every 6 inches on the underside of the cabinets and wired them to the load side of the GFCIs.
So far so good, but here’s the bad news:
1. The previous idiot used 14g Romex into and out of the GFCIs.
2. I did not find junction boxes, so he may have spliced inside the old boxes that now have tile over them.
3. The two circuits come off a two pole 20 amp breaker.
4. Tenant has been running a microwave oven [nameplate rated 13a], a 4-slice toaster [rated 1700W], a hamburger grill [rated 1440W], a kettle [rated 1000W] and a coffeemaker [rated 900W] off the circuit that failed. Now the toaster and microwave have been moved to the other circuit.
When I told the tenant that she is overloading the circuit she said “But we’ve been living here for 4 months and there’s never been a problem before.”
My Questions:
Why did the GFCI melt down at the load-side white terminal, but no damage on the hot-side load terminal nor either of the line terminals?
Why would these two kitchen circuits be on a two-pole breaker meant for a 220v circuit? Seems to me that when one pole tries to trip, the other one that its toggle lever is tied to would tend to delay tripping. Assuming I can replace the 14g wire with 12g, should I decouple these breakers?
If the HO won’t go for the cost of opening walls to re-wire with 12g, I have told him that we should at least replace the 20a breakers with 15a in order to reduce fire hazard of over-heated 14g wires. Any problem with doing that?
Thanks in advance. I’m sure you will all want to answer between 3:30 and 6:30 Pacific Time tomorrow.
BruceT
Edited 2/3/2007 8:56 pm ET by BruceT999
Edited 2/3/2007 8:59 pm ET by BruceT999
Edited 2/3/2007 9:00 pm ET by BruceT999
Replies
Screw was not tight enough.
Ozlander
"Screw was not tight enough."I thought of that but when I checked it the screw seemed tight; couldn't tighten it any more. Now that I think about it, what seemed like a tight screw might be one glued in place by melted plastic.BruceT
It's always the weakest link. Once a connection heats up a little bit, the resistance goes up and it gets even hotter.I could also see a neutral overheating if it was used for the return on two circuits that weren't on opposite poles, but such a miswiring would trip the GFCI.Nothing wrong with using the double-pole breaker for two separate circuits, though.
So convenient a thing it is to be a reasonable Creature, since it enables one to find or make a Reason for everything one has a mind to do. --Benjamin Franklin
You've got a 14 gauge wire on a 20 amp breaker running heating elements and microwaves. That there is your problem. I'd at least drop it to a 15 amp breaker. You have to at least do that.
Looks like the connection was in the wrong place (see the little hole in the back) loose screw would also do this with a high load.
#14 is rated for 15 amp breaker so thats what it should be changed to. If its two circuits then two independent breakers. #14 in and #14 out is fine and correct. But not on 20amps. Kitchens require two circuits.
Thats a lot of high load stuff in that kitchen, any way to add another circuit?
The wire is in the right place, although so much exposed copper lends credence to the earlier post about a loose screw. That "little hole in the back" is a bubble in the melted plastic. "Thats a lot of high load stuff in that kitchen..."As I started adding up the loads, I half expected to find an arc welder in there. :)BruceT
I don't think that it failed at the terminal. Hard to tell from the picture, but it looks like the failure might have been in the wire about 3/8" from the screw.My guess is that the wire was nicked when stripped. That on top of it being #14.But #14 is actually rated for 20 amps. And if you have to do derating calculations then you can use the 20 amp rating. Just can't have it protected by a larger than 15 amp breaker. Not sure of why that limitation, other than maybe to allow for some margin of safety for things like nicked wires.But a kitchen is very hard load.A 20 amp breaker can allow 2-3 time rated current for manybe a minute. They are near instaneous for short currents. But a time inverse curve for overloads.And many things in the kitchen only operated for a couple of minutes. So it can be overloaded without tripping the breaker.About the 2 pole breaker, my guess is that it is a multi-wired circuit (shared neutral), with it broken out in one of the burried boxes.Codes does not require a 2 pole breaker for multi-wire circuits unless one device has both hots on it. But some people still like them because that limites the possibility the breakers be moved to the same leg or that someone things that a box that has both hots in it is dead becaue the receptacle on one leg is dead. Of course that is not an issue in this case.You can verify if it is a multiwire circuit by looking at the cable connecting to the breaker..
.
A-holes. Hey every group has to have one. And I have been elected to be the one. I should make that my tagline.
Loose terminal, 100%.
Look closely at the red circle on the pix, the edge of the terminal is oxidized, sure sign of high heat for some time (>150C for hours) or very high heat for a short time (400C for seconds). Would guess the hours scenario.
View Image
"...sure sign of high heat for some time (>150C for hours) or very high heat for a short time (400C for seconds). Would guess the hours scenario. "By "hours scenario", do you mean all in one incident, or cumulative effect of a few minutes at a time over 4 months?BruceT
Actually most new GFCI outlets are back wire. That's a single hole on the load side and 2 holes on the line side. You are supposed to insert the wire in the hole and the screw tightens a backing plate to holed the wire in place.
I don't think the screw is supposed to have the wire wrapped around it. Did you ever notice that the screw is loose on the side, not tight like on a regular outlet. Check it out.
Headstrong, I'll take on anyone!
"I don't think the screw is supposed to have the wire wrapped around it. Did you ever notice that the screw is loose on the side, not tight like on a regular outlet. Check it out."What, in my slightly-out-of-focus pic, looks like wire wrapped around the screw is actually part of the terminal. The apparent curve at the top is caused by plastic from the case beginning to flow downward and covering part of it. The wire was in the correct place."That's a single hole on the load side and 2 holes on the line side."All the GFCIs I have seen have two backwire holes for load as well as for line. Maybe some brands are different.
BruceT
I'm not buying the "nicked wired" answer because the break is up in the insulation.
I think that that spot got a little work hardened, which increased the resistance at that spot. As this spot got hotter, the increased heat increased the resistance even more until the effect spiraled up to melt down. Not to say that the whole wire wasn't getting a bit warm at the same time.
At 20 amps, a resistance of 1/100ohm will develop a 4 watt hot spot. Over 4 months time. . . add some oxidization shrinking the cross sectional area of conductance. . . Spot resistance increasing all the time. . .Heat deformation of the electrical insulation moving it away from the wire, thus crerating a small heat insulating air gap. . .then make some coffee, toast a Pop Tart, and nuke a bagel.. . .Holy Arcs and Sparks, Batman!
SamT
Anyone who doesn't take truth seriously in small matters cannot be trusted in large ones either. [Einstein] Tks, BossHogg.
"I'm not buying the "nicked wired" answer because the break is up in the insulation."Sorry for the poor picture quality. The wire did not break, I cut it in order to remove the GFCI. ". . .then make some coffee, toast a Pop Tart, and nuke a bagel.. . .Holy Arcs and Sparks, Batman!"The tenant did describe a "sizzling sound" coming from the GFCI before it shut down, but the arcing was not coming from a break in the wire.
BruceT
Sorry for the poor picture quality. The wire did not break, I cut it in order to remove the GFCI.
In that case, I refer you to Junkhound's post.
edit: if you dismantle the GFCI case, you may find a smaller than 14ga conductor inside connected to the toasted terminal. This would be the primary source of heat. These smaller conductors are usually short enough to work under normal circumstances.
SamT
Anyone who doesn't take truth seriously in small matters cannot be trusted in large ones either. [Einstein] Tks, BossHogg.
Edited 2/4/2007 3:05 pm by SamT