Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

House almost burnt down from electrical fire, wondering why

  1. Apr 22, 2017 #1
    Hello. I entered my home to find that a light was on. I definitely did not leave the light on when I left.

    When I looked at the switch, it was burnt and there was smoke residue everywhere on the mirrors and a burnt smell in the room. I went to turn it on and there was a hot heat coming from the switch area. I tried to turn it off but the light would not turn off. The only way to turn off the light was by delicately balancing the switch inbetween the on/off position.

    I am trying to figure out why this occured. I had a mouse in the house last year and I am wondering if a mouse could have done this, or if this was somekind of odd mechanical failure of some sort. The Fuse to the room been turned off and the switch has been disabled. No heat is coming from it anymore.
  2. jcsd
  3. Apr 22, 2017 #2
    I had a somewhat similar situation with a dimmer switch that heated up badly when I began using compact florescent lights in the fixture. A new dimmer switch designed for the situation was the solution.
  4. Apr 22, 2017 #3


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Pull the fuses or switch off the circuit breaker for that circuit and get it checked by an electrician.

    It's possible that frequent use of the switch caused a wire to become disconnected and is shorting to another terminal on the switch.
  5. Apr 22, 2017 #4


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Not a great idea :-)
  6. Apr 22, 2017 #5


    User Avatar

    Staff: Mentor

    I recently had a customer whose control data network had suddenly gone flaky on them. It had worked well for years, and all of a sudden they were getting lots of communication errors between the networked control devices. After a bit of troubleshooting, they found that a rat or mouse had chewed up one of the two termination resistors on their doubly-terminated bus network, which caused reflections and comm errors. They replaced the termination resistor, and the network went back to normal.

    So yeah, I think your mouse theory may have merit. I second the suggestion to have an electrician take care of it.
  7. Apr 22, 2017 #6
    Electrician is coming on Monday, if all things go well.

    Typo I made. What I meant was, "I meant to turn it off."
  8. Apr 22, 2017 #7


    User Avatar
    Gold Member

    A possible problem is that the mouse made the circuit wet, and this then drew a current enough to cause fire but not enough to blow a fuse. I have had similar situation and suggest installing a Residual Current Breaker so the circuit trips out.
  9. Apr 22, 2017 #8

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    My GUESS is
    a failure internal to the switch
    or a loose wire on the switch.

    Reason for those two guesses is that
    (1) you said the heat is coming from the switch area.
    (2) and not enough current flowed to trip the breaker

    Limited current will make heat only where there's substantial resistance. That should be only in the light bulb.
    The current to the lamp should flow freely through the switch making no heat.
    If the switch's internal contact develops resistance it will get hot and make things worse as it corrodes or melts.
    If one of the screws holding the wires on the switch comes loose it may get hot and wreck the switch.
    If it's one of those cheap switches where the wires aren't under a screw but push in the back, well, i just don't trust that method of connection.

    I found a loose screw on my water heater plug, the insulation on the wire was charred. Just lucky i noticed it before it got as bad as yours.

    So - will you let us know what your electrician finds ? It might help the next guy.

    old jim
  10. Apr 24, 2017 #9
    Ok. My electrician came and he said it was an overload problem. I looked at the box and it was a secure metal box. So that rules out a mouse.

    But I dont know how the switch could have overloaded, it had a shaver plugged in to the socket but it was turned off, and the light was turned off.
    In another room, I had a refrigerator and a space-heater turned on. But he said the other rooms should not have overloaded that room's switch.

    The metal box rules out the possibility of a mouse chewing the wires, and nothing was turned on in the socket. So I don't know what occurred.
  11. Apr 24, 2017 #10

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Hmm a switch AND a socket in that box?

    I guess the electrician didn't say whether the wiring daisy-chains through that switch to someplace else . That would give a possible explanation, other loads just pass through the switch .

    If he left you with the burnt up switch can you post a picture someplace ?
  12. Apr 25, 2017 #11
    The sort of problems described above can be indicative of routine wear and tear absent of any other failure.

    Courtesy of Wikipedia:
    The above picture is blatantly NA-biased but chances are the snap switch in question is relatively similar to the above pictured. These control devices are mechanical arrangements with a limited service life. For the average wall-mounted toggle age is a big factor to consider in service life as the materials used in manufacture have come leaps and bounds since the days of bakelite ect. A common problem with this lever and detent system for holding an electrical contact together involves a situation where the (usually) nonmetallic housing cracks or separates away from the normally metal mounting strap to which it is jointed. The lever system loses its fulcrum point and typically the springs that a snap switch normally has can fall out of arrangement, leaving the contacts to flop around willy-nilly, creating all sorts of fun problems one would normally find with a bad connection, most notably intermittence and heat.

    This does not completely rule out an overload or other fault situation. Another way in which a switch of this type can easily fail is for the current-carrying metal parts to become very hot due to carrying more demand for current than its material design can cope with causing parts to melt and once again throwing the arrangement of the system out of whack. This sort of runaway failure is a common theme in electricity, for example a slightly cracked case may put insufficient pressure on a spring to fully engage the contact point it presses on; this can create heat which alters the metal and transfers to the plastic, which may melt and lose mechanical strength, which further propagates the faulty contact.... and finally, cue the magic smoke!
  13. Apr 25, 2017 #12


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Perhaps live to the other room with the heater was supplied via the switch terminal, eg the switch terminal was used as a junction box. If that the became loose it could burn up the switch.

    As others have said.. a photo of the switch would be nice to see.
  14. Apr 25, 2017 #13

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Indeed , and krater's great picture shows what looks like springs for those 'push in' connections from the back.

    "Live" might come into the switch via the top screw terminal and continue on to next room via the push-in. Looseness at either would make heat whenever the heater(or anything else) in next room is ON drawing current.

    Speculation of course but exactly the sort of thing electricians find.
    Myself i won't use those push-ins. They're not even allowed on 20 amp rated fixtures anymore, just 15 amp.

    Thanks krater and CWatters .

    Was either one just recently connected?
  15. Apr 25, 2017 #14
    No they had been running for a while.

    The electrician left the outlet but I think he either threw out the switch or used the same switch.
    I took a peek at what he was doing and the wires all seemed Fried on the side closet to the outlet. The coating of the wires that were near the wall were not fried. It was a gradient, the closer to the outlet they were the more fried they looked. I am going to upload a picture soon.
  16. Apr 25, 2017 #15


    User Avatar
    Science Advisor
    Gold Member

    Shouldn't the electrician have addressed the reason for an overload by adding or moving circuits or whatever? Sounds to me like the same thing could happen again.
  17. Apr 25, 2017 #16
    I don't like to euphamize arcing faults particularly by using tame-sounding descriptaves like sparks or flashes. I prefer to discuss them more accurately as fire shooting out of things. Very hot gas, liquid metal, and everything it's carrying end up in places. It doesn't always make sense. It is the reason the moving parts that can pass such flatulence normally exist in a box.
  18. Apr 26, 2017 #17

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    It's a safe bet that is where the heat source was. An insufficiently tightened screw terminal can cause that , the wire eventually relaxes maybe from thermal cycling and the joint becomes resistive. That exaggerates the thermal cycling with load.

    A wire that's charred at a connection and becomes less charred as you look further from the connection is a giveaway. The connection was the source of heat.
    Were the circuit overloaded the wire would heat along its whole length. That's what circuit breakers protect against.
    When a "bad" connection gets 'bad enough", normal load current will overheat it.

    So, while we lack enough details to absolutely nail down your problem, you have presented circumstantial evidence of a loose connection.

    I hope you find the "smoking gun" , so as to put your mind at ease.

    In old house wiring i have fixed a lot of 'loose connections'. Only one actually got to the point of smoke and flames, but several charred wires.
    A good habit to form is feeling around your fixtures for undue heat or warmth in your everyday using them.
    See this thread for a non-electrical example. a refrigerator door unduly warm:
  19. May 30, 2017 #18
  20. May 30, 2017 #19

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Thanks for the feedback ! Is all well now ?
    Snipped from your pictures

    Without handling it , just going by your photos
    i would say it looks like you had a loose neutral connection through which current from a significant load ( like a heater or airconditioner) was returning..
    If that's right the receptacle had to be getting warm before it went up in smoke. That's why it's good to feel of them now and then especially in heating season.

    But troubleshooting from a distance is very tentative .

    Thanks for the pictures and for letting us know !.

    old jim
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted