I got this extension cord with a used electric oil radiator. I assume the extension cord had been used with the radiator in the past. It looks like a 14g or 16g extension cord.

The extension cord has the factory end pieces on it. I went to use the cord from an exterior gfci outlet. I plugged the cord into the house outlet, walked 50' passing the female end on the way to get an electric lawn mower that I wanted to test, turned around and the female end of the cord was arcing and a 3-4" flame was coming out of the end. I had left the cord end near some tall grass so I ran and unplugged it from the house. I then stomped out still burning fortunately green somewhat lush grass.

I assume the female end of the cord must have been damaged and the insulation between the hot and neutral prong holders must have somehow degraded enough that it started arcing. This really surprised and scared me - if the grass had been dead and dry I could have started a massive fire.

I read an article recently about a man who used an "outdoor" extension cord inside to run his space heater where the cord passed under a rug under his bed. An electric fire ended up burning down his mobile home. The article blamed it on using an outdoor rated cable inside and that you should never do that. Wouldn't an outdoor extension cord be a way safer option than an indoor rated cord?

I suppose the person who got really lucky was the previous owner of the oil filled radiator as they were one use away from burning down their bedroom but maybe the arcing wouldn't have been as bad if something had been plugged in.

I am now paranoid about leaving any kind of extension cord plugged in and unattended. I've often made shortened extension cords out of 100' 12 awg outdoor extension cords and extension cord repair end pieces and I use these all the time inside (typically for diy projects that involve power tools). Similarly I sometimes leave extension cords plugged in outside that aren't immediately in use. How paranoid should I be about an extension cord turning into a flame thrower?

On inspection of the GFCI outlet it was tripped so either it tripped when I pulled the cord or had already tripped after the fire was in action. I guess that is some relief that maybe I didn't need to pull the cord.

melted extension cord


5 Answers 5


I read an article recently about a man who used an "outdoor" extension cord inside to run his space heater where the cord passed under a rug under his bed. An electric fire ended up burning down his mobile home. The article blamed it on using an outdoor rated cable inside and that you should never do that.

That's not what happened there. The problem was, the extension cord was trod upon. Months or years of foot traffic, or a furniture leg happening to land square on it, crushed the cable. Each wire has many strands. Either enough strands broke for the remaining strands to get very hot, or the strands separated altogether and began series arcing which creates a tremendous amount of heat (but current is limited to what the appliances take, so the breaker never trips).

A "space heater" is the worst case load, since it draws the absolute maximum amps.

This exact problem is why AFCI (arc-fault detecting) circuit breakers were required for bedrooms in 2005 - because bedrooms are always having plugs mashed against furniture, run under rugs, and of course electric blankets. This was widened to most living space in 2014.

However FEMA canceled the AFCI requirement on "FEMA Trailers" aka the mobile homes they stockpile to house Americans after disasters like Katrina. Those don't have AFCI and nor do pre-2007 homes. (but it's an easy optional upgrade, and since it's optional, it can even be done with a $20 AFCI receptacle at the first receptacle on the string, instead of a $50 AFCI breaker).

I plugged the cord into the house outlet, walked 50' passing the female end on the way to get an electric lawn mower that I wanted to test, turned around and the female end of the cord was arcing and a 3-4" flame was coming out of the end.

That didn't happen by accident to a new socket. That socket was already degraded inside from previous overheating. And I'm sure there was evidence of that, as warping or discoloring. Those molded-case sockets are pretty good at giving such signs.

That it happened after use with a heater is no surprise; a heater is the max power allowed on a 15A circuit.

I am now paranoid about leaving any kind of extension cord plugged in and unattended. I've often made shorted extension cords out of 100' outdoor extension cords and extension cord repair pieces and I use these all the time inside.

Well, stop!

The entire point of NEC 210.52 (receptacles at frequent spacing along walls and countertops) is to get people to stop using extension cords indoors.

If you need more receptacles - and clearly, you need more receptacles - then you are better off adding more receptacles. The simplest way to do that without drywall work is using Legrand Wiremold surface conduit to come off an existing receptacle box and extend to new receptacle locations of your choice. Use 12 AWG THHN wire inside the Wiremold. Terminate to receptacle side screws and torque to spec.

Do this shipshape and Bristol fashion, don't use some hokey alternate system. You're not after "hokey".

While this won't be cheap, it'll be cheaper than heavy grade extension cords like I use (12 AWG and intended for large appliances).

Now as far as repairing extension cords, the right way to do that is to replace the plugs and sockets as needed. Trying to splice the cable inline is a waste of time IMO, and I just stick a plug and socket there.

Choose replacement plugs and sockets whose strain relief actually fits the cord material.

Regularly inspect your plugs and sockets, and if they look in any way degraded, replace them immediately. If the jacket is cracking, into the trash it goes (save the replacement plugs and sockets for reuse).

  • Thanks. I edited the question to reflect that the extension cord usage is not for standard "house" items. My house adheres to modern code with AFCI for bedrooms (not in all areas) there are outlets every 6' or so. I use 12 awg extension cords for power tool use when doing projects or in the area of my house that i've dedicated as my shop. Should I be paranoid about a 12awg 3' new extension cord being left pluged into the wall and a mitresaw? I could possibly mount an outlet on the ceiling and plug the mitresaw in directly but this seems like overkill ? Jun 17, 2022 at 21:18
  • 3
    @Fresh I would keep a close eye on the extension cord is all. That fire-cord had a problem before you plugged it in, I suspect that's why the other guy got rid of it. It's always better to use NEC wiring methods to extend receptacles to where you want them. It is just more inspectable. Jun 17, 2022 at 21:38
  • I'm guessing the requirement for AFCI breakers didn't happen everywhere in 2005/2014? I've never heard of worldwide coordinated legislation like that ever happening. Jun 18, 2022 at 10:00
  • 2
    Sorry, but as a Bristolian, it's 'shipshape and Bristol fashion'. This answer is extremely good. Jun 19, 2022 at 9:44
  • 1
    @Michael foreheadslap I knew that, I just didn't remember I knew it. Jun 19, 2022 at 18:38

There are springy contacts inside the outlet end that grab firmly onto the metal pieces of the plug. The firm grab makes for a low-resistance connection and very low power dissipation across the connection.

Eventually extension cord ends like these lose their springiness in the contacts (even wall outlets do this). As the springiness gets less, the contact pressure on the plug gets less, and the resistance gets higher. Power gets dissipated over this resistance as the load continues to draw its current, which generates heat. As less springiness and more power dissipation over increasing resistance at high current continues, the heat can get to the point that the material in the plug or outlet spontaneously combusts. The lose connection may also arc, which generates really super-high heat, which also starts fires.

I have had a few extension cords like yours that lost their contact strength, and I have felt the outlet end get hot under high current usage. A high-quality replacement outlet end saves the cord.

Edit: I noticed that nothing was plugged in yet. Something would have been bridging the contacts in your cord outlet to pass electricity between the slots. What it was would only be a guess, but the result is the same: increased power dissipation leads to heat, then to fire.

  • 1
    Plus, even if the extension cord started with just as good a socket as a standard wall receptacle, the movement of extension cords during normal use makes the sockets wear out faster. Jun 17, 2022 at 20:14
  • 4
    Noted that OP said the cable end was near enough grass to be a fire concern, and also that the grass was wet. I wonder if the cable end landed in enough dirty outside moisture to bridge the contacts ?
    – Criggie
    Jun 18, 2022 at 13:26
  • There's part of me that thinks grass blades would burn away quickly and not make enough heat over enough time to melt the end like that. I surmise a manufacturing defect between the contacts inside the plastic that finally wiggled through and went critical. Which means this could have happened inside the house.... A hand router at my old workplace went flame-thrower out the strain relief because the wire inside fatigued. Jun 19, 2022 at 21:22

For any outdoor work with any normal tools, a minimum 12 Gauge extension cord is recommended.

Newer use more than one cord.

You are right, a overload (or lots of water) in the cord plug caused melting and consequent Arching.

A GFCI or AFCI circuit breaker could have helped.

You can get a wet protection for the cord plug.

cord protection

The cord was damaged over time (not at once), and eventually started shorting).

You do not know where that cord was in its previous life, and what torture was it exposed to.

  • 3
    GFCI was protecting the circuit. The cord was only plugged into the outlet, the tool/lawn mower was not plugged in yet.
    – crip659
    Jun 17, 2022 at 18:35
  • 2
    If the problem was "wet grass caused a short, GFCI could/should/would have helped. In this case it didn't, so suspect prior damage that now became a real problem. Jun 17, 2022 at 18:38
  • 3
    AFCI would have helped. Jun 17, 2022 at 19:29
  • 2
    Had the cord (that I use for a fckn radiator!) laying in wet grass and it started on fire. Yeah, and? +1.
    – Mazura
    Jun 18, 2022 at 0:51
  • 6
    A coworker and I deliberately overloaded an extension cord and watched it through an IR camera. It was enlightening how much the cord heated up and how unevenly. Weak spots would definitely be evident, speaking of not knowing where the cord has been.
    – donjuedo
    Jun 18, 2022 at 12:25

This is my speculation: When the cord was used with a heater, the socket got hot enough to make the rubbery plastic soft. Maybe, the heater plug kept the contacts in place during use, and when the heater was unplugged the socket was still hot and soft, which allowed the contacts to move and touch each other. This implies the cord was also unplugged at the same moment.

That is to say, a short circuit developed just moments after the cord was used the last time. When you later plugged the cord, the short circuit was already there and started getting hot until the plastic ignited.


There are a lot of answers here that while they describe bad things that can happen I do not believe are relevant to your case. Your description says there was no load on the cord at all.

If there were truly no load there would not have been a fire, thus there must have been a load even if you didn't intend there to be one--I think you got some grass inside the hot and at least one of the neutral/ground holes. It was damp enough to provide a decent current path and dissipated a fair amount of energy--thus becoming the ignition source.

I see no reason to suspect prior damage in this scenario.

  • 2
    I actually doubt that. If enough wet grass got in to do all that, some current would likely have gone to the physical ground, which would have meant an imbalance hot/neutral and a GFCI trip. Jun 19, 2022 at 2:52
  • @manassehkatz-Moving2Codidact Only if grass went in the ground hole and it's a GFCI outlet. I find it much more believable that grass caused a short than that it just had a spectacular failure simply sitting there with no current flow. Jun 20, 2022 at 3:10
  • 2
    Nothing to do with the ground hole. Wet grass conducting from hot or from neutral to the physical ground - actual dirt/soil/etc. would cause a GFCI trip. Yes, a short to the ground pin would too, but the GFCI is not monitoring for "something on the ground pin", rather it is monitoring for "hot <> neutral". Jun 20, 2022 at 3:13
  • When I said wet grass what I meant was that the grass was lush. Even if the grass was wet I have a really hard time believing that you could in 1000 drops, drop a cord in such a way that you get grass to connect up like this. The area I put the female cord end was actually mostly concrete with unkept grass growing between the segments. Jun 20, 2022 at 7:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.