A scary discovery, what's going on?

[roger wiegand]

I went out to my shop to turn the lights off just now; the programmable switch had lost its memory due to a prolonged power outage. To my surprise the switch box was blazing hot. The picture below is what I found inside. Needless to say this scared the wits out of me. I've powered off the circuit to think about what to do. I use the switch every day and have never noticed it being hot before.

In theory the circuit should not be overloaded. I have 12 two-bulb T5HO fluorescent fixtures for light in the shop. at 54 watts/bulb that should consume 12 X 2 X 54= 1296 watts. The 15A circuit should be able to support safely 80% of 15A or 0.8 X 1800 watts = 1440 watts continuous load. The overhead lights are the only load on that circuit.

In practice the wire nuts are melting!

Any insight as to what's happening here? Obviously I could split the lights onto two breakers, but that would be a pain and they really should be safe as-is.

IMG_6827.jpeg

[Cameron Wood]

Looks like a bad wire connection overheating- very apropos of the arc fault discussion.

[Lee Schierer]

I would be interested whether the wires are twisted together or just bunched together and held by the wire nut.

[Tom M King]

I was getting ready to ask the same question that Lee asked.

[Dave Zellers]

I would be interested whether the wires are twisted together or just bunched together and held by the wire nut.

Exactly what I was wondering. Are the wires loose and arcing?

[Bill Dufour]

I think Wagos are harder to mess up then wire nuts but they take up more room. When you redo those connections you will have to clean and expose the wire to bright copper. Soot is an insulator. I assume no aluminum was involved?
Bill D

[roger wiegand]

I will take a picture and post it when I take it apart later this morning. No aluminum. This was wired about a decade ago. They are almost certainly twisted, as that's how I was taught to do it.

Is twisting a good thing or a bad thing? The interweb seems to be divided pretty close to 50-50 on the subject. Some manufacturers say it's not necessary, but don't say not to do it.

[Jim Becker]

Pre-twisting provides a very positive connection, IMHO. I've also now moved to Wago connectors, but if I do a wire nut, I always pre-twist with lineman's pliers, clean/trim the end and then apply the wire nut followed by tape.

It will be interesting to see what you find when you peel that apart to see why it's arcing.

[Edward Weber]

Pre-twisting provides a very positive connection, IMHO. I've also now moved to Wago connectors, but if I do a wire nut, I always pre-twist with lineman's pliers, clean/trim the end and then apply the wire nut followed by tape.

It will be interesting to see what you find when you peel that apart to see why it's arcing.

I haven't used the Wago but that's the way I've been doing wire nuts for decades.

[Jim Becker]

I haven't used the Wago but that's the way I've been doing wire nuts for decades.

I "discovered" them when I was wiring the new shop, Edward. They were a game changer, especially with the heavy #10 wire on the machine circuits. They also need to be taped, IMHO, but there's a lot less screwing around with tools...cut, strip, insert and lock, tape. Done.

[roger wiegand]

It's going to be embarrassing in all likelihood, but here are my best guesses. I believe strongly in the importance of "near miss" reports for improving safety. I've sure had an important lesson.

The stranded wire from the timer switch is the principal point of failure. I see several things when inspecting the wire and the connection.
1) The stranded wire is melted along its length
2) it appears to be wrapped around the solid wire several times and in good contact with both the solid wire and the spring inside the wire nut.
3) It appears corroded inside the wire nut
4) it looks as though several strands (3-5) of the stranded wire were broken off short and possibly not in contact with the solid wire (poor wire stripping technique?)
5) the other two connections appear similar, but are uncorroded and not melted

Possible issues:

1) missing strands will clearly lower the ampacity of the wire, resulting in greater heat; why this only became apparent after a decade is unknown.
2) the wire nut was one of the lightweight ones that come for "free" with the switch, the others were of much higher quality. A possible source of corrosion? (Circuit was (unusually) out of use for 10 days with very high humidity, a contributing factor?)
3) stranded wire was not tinned

Suggested solutions:

1) Greater care in handling stranded wire
2) tin the ends of the stranded wire
3) Employ a clamp-type connector to join stranded to solid wire
4) Check all other similar installations in the house and shop, correcting as needed

My tentative conclusion is operator error on installation, and possibly a low quality wire nut.


IMG_6828.jpeg IMG_6829.jpeg IMG_6830.jpeg IMG_6831.jpeg

[Bert McMahan]

I'm no electrician but I don't think you want tinned wire in a compression connection. I've personally seen two equipment failures (one very expensive) due to a tinned, stranded wire being installed in a screw terminal (i.e., compressed under a screw head). The solder will creep over time and cause the joint to fail. I don't know if this applies to a wire nut also, but since it's a compression fitting as well I would err on the side of not tinning stranded wire.

No idea why your wire melted though!

[roger wiegand]

Oh, something else to learn about. My friend, who is an electrical engineer and an old-time TV/radio repair guy, has been telling me to always tin the ends of stranded wire.

I'm no electrician but I don't think you want tinned wire in a compression connection. I've personally seen two equipment failures (one very expensive) due to a tinned, stranded wire being installed in a screw terminal (i.e., compressed under a screw head). The solder will creep over time and cause the joint to fail. I don't know if this applies to a wire nut also, but since it's a compression fitting as well I would err on the side of not tinning stranded wire.

No idea why your wire melted though!

[Bert McMahan]

Do you mean for wire nuts or for other uses? You definitely don't want to tin wires for crimp connections, and AFAIK you don't want to tin wires in screw terminals (as I mentioned, I've seen them come loose myself). You can probably do it for very small/fine wires to keep them together, but not for the really thick stuff.

Plus, if the connection heats up and the solder softens then your connection will get even worse.

You absolutely should tin wires before soldering them, but I don't think you want solder in any "physical" connections (like a crimp, ferrule, or screw terminal). I do think there are some ferrules designed to be crimped on THEN filled with solder, but not the other way around.

[John Lanciani]

In theory the circuit should not be overloaded. I have 12 two-bulb T5HO fluorescent fixtures for light in the shop. at 54 watts/bulb that should consume 12 X 2 X 54= 1296 watts. The 15A circuit should be able to support safely 80% of 15A or 0.8 X 1800 watts = 1440 watts continuous load. The overhead lights are the only load on that circuit.

I suspect your circuit current is a fair bit higher than you think. You haven't accounted for the power consumed by the ballasts, nor have you accounted for the power factor of it. (Watts are never equal to volts x amps in the real world, you must know the power factor to calculate watts) I'd bet an internet buck or two that your circuit current is +/-14 amps.