Connect with us

House Electrical Circuit: Leakage?

Discussion in 'Home Power and Microgeneration' started by (PeteCresswell), Jan 7, 2013.

Scroll to continue with content
  1. Recently installed a generator cutover switch
    ( which tells me how much each circuit
    is drawing.

    I have one circuit which, when I *think* everything is unplugged,
    still draws 18 watts.

    First thing that comes to mind is that the circuit is serving
    something that I do not know about; but I am hard-pressed to find

    That seems to leave "Leakage".

    Is there any such thing as an electrical leak in home wiring?

    If so, typical causes? Location strategies?
  2. Neon John

    Neon John Guest

    There can be leakage but not the type you're thinking of. Romex has
    capacitance between the hot and ground wire and to a lesser extent, to
    the neutral. It is insulated with a high dielectric loss (PVC)
    insulation. Thus, some small amount of real power dissipation will be
    consumed per foot heating the cable even when there is no load. If
    the runs are long that could amount to a few watts.

    Do you have a doorbell transformer? It probably dissipated 4 or 5
    watts just sitting there.

    Is your furnace/heat pump on the circuit? If so, there's a control
    transformer that also will consume a few watts unloaded.

    Finally, the is the question of near-zero-indication accuracy of the
    meter. I see from the specs that the input spec is 30 amps.
    Presumably the current transducer is rated for at least 30 amps and
    probably 50. 14 watts would be 14/120 volts = 0.116 amps. That is
    0.116/50 = 0.233 percent of full scale on the transducer. That's not
    figuring in the tolerance of the voltage transducer. The CT is likely
    to be a 1 or maybe 0.5% part. So that 14 watt reading could be
    literally nothing.

    One way to check is to open all the breakers. What does the wattmeter
    say? Now close in the breakers, one at a time and see what happens.
    If the reading develops when you close a breaker then you know that
    it's real. Remember that an indication that low is way outside the
    tolerance of the metering element so it might be 1 watt or 20 watts or
    ??? If it really is 14 watts then that's enough power dissipation
    that you should be able to find something warm to the touch on that


    John DeArmond
    Tellico Plains, Occupied TN
    See website for email address
  3. Jim Wilkins

    Jim Wilkins Guest

    Doorbell or old Princess telephone transformer? It would likely be on
    a junction box cover rather than plugged into an outlet.
  4. Jim Wilkins

    Jim Wilkins Guest

    That's a very good point, so I checked two connected 100' extension
    cords with neon bulbs in their sockets on a P4460 KAWez. The readings
    were 0.00A, 0.0W.
  5. mike

    mike Guest

    Are we sure that the unit measures watts and not volt-amps?
    Still 14 VA is a lot for a disconnected circuit.

    Did you actually disconnect the wire? AKA "where does
    the transfer switch get power?

    Doorbell? Burglar Alarm? Garage door opener?
    turn off the suspect breaker and see if they all still work.

    You can break the circuit at any wall plate.
    Half-interval search method should get you there in few tries.
  6. [snip]

    any motion detector lights? cable/sateliite tv box?
  7. Vaughn

    Vaughn Guest

    They stick doorbell transformers in the strangest places. I've been
    living in this old house for 25 years and still haven't stumbled across
    mine. Yet the doorbell works.

    When that hidden transformer finally goes bad, I hope it's an open
    primary and not shorted turns.

  8. T. Keating

    T. Keating Guest

    Each GFI(outlet or breaker) or Arc-fault breaker will consume a couple of watts per unit.
    (tripped or active will make no difference.)

    There is not very much you can do about it, since building code requires them for certain

    Other leakage... Clocks on appliances, doorbell, transformer powering thermostat, power
    line modules(x-10), etc..
  9. Jim Wilkins

    Jim Wilkins Guest

    Another good point. A single-outlet plug-in GFCI here draws 1.4W.
  10. j

    j Guest

    Interesting. With a cost of electricity at just under $.12 kWHr, it's ~
    1$/W or $1.40 for that single GFCI. I have 5 of them, so $7/yr.

    How did you come up with the draw so quick?

    (copy sent to OP by accident)

  11. Jim Wilkins

    Jim Wilkins Guest

    I'll spare you a quick-draw joke.
  12. Per mike:

    That circuit was one of the two circuits that power the transfer

    Swapped it over to another circuit and the "leakage" went away.

    Thanks!... I never would have thought of that.
  13. Jim Wilkins

    Jim Wilkins Guest

    Emailed question answered here:

    No, the KAW doesn't measure the draw of the socket it's plugged into,
    GFCI or standard. I plugged a stand-alone, plug-in GFCI into the KAW.
  14. Pete C.

    Pete C. Guest

    Probably a doorbell transformer buried in the attic or similar.
  15. g

    g Guest

    In my old house they fixed the transformer onto the service panel. You
    cannot see it unless you take off the front cover...

Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day