DIY Chatroom Home Improvement Forum

DIY Chatroom Home Improvement Forum (
-   Electrical (
-   -   3way switch (

satsu 10-27-2006 12:46 PM

3way switch
I have a long closet shared by two bedrooms with a 3way light switch to control the closet light from either bedroom. I went to replace the light fixture and I made a strange discovery. Instead of throwing the breaker like a safety-oriented person I simply adjusted the switches so that the light was off, than I removed the fixture. As I was putting up the new fixture I felt a small shock. I grabbed my multimeter and checked the hot and found it had 18Volts AC. No matter how I adjusted the switches the hot wire to the fixture always had either 18V(off) or 120V(on)!! So I ran to the store and grabbed 2 new 3way switches, turned off the breaker, affirmed that the wiring was correct with my cable tracer (see picture below) and installed the 3 way switches. To be honest, when I first hooked them up I confused a common with a traveler, but I straightened that out and still the problem persists, I either have 18volts or 120volts to the fixture. I tested the travelers on the switch that is connected directly to the incoming hot wire and found that in either position 1 traveler would have 18V and the other would have 120V!!! I tested the switches for continuity and found that from the common, only 1 OR the other traveler would read ohmage indicating that the other should be completely disconnected which leaves me wondering how it would pass 18V when connected?

Shouldn't I be able to have 0V or 120V rather than 18V or 120V??!! Why are my new 3way switches 'bleeding' 18V into the 'disconnected' traveller? Is it possible I damaged them by hooking them up improperly at first so that they are shorting? Am I missing something fundamental here? Sorry for the long question and all the details-Thanks for any and all help.

redline 10-27-2006 12:57 PM

see website below for wiring.


It appears you have the traveler and hot misplaced.

satsu 10-27-2006 04:00 PM

Hmmm...As I said I traced the wiring and I've verified that it complies with the proper wiring diagrams. I'm definately not confusing a hot for a traveller, but even if I had, I would not expect the result I noticed. It should still either work or not work with 120V or 0V @ the fixture, not strangely bleed 18-25V under any conditions..(from my amateur expectiations) So I went to another 3 way switch controlled light fixture, made sure it was 'off' and removed and unwired the fixture. With the breaker back on I measure either ~25V or 120V across the black and white wires feeding the light depending on the position of the switch. I replaced the fixture with the switches ON (120V @ fixture), turned on the breaker and I had light. Flipped the switch OFF (~25V @ fixture) and light went off.

I'm now thinking maybe this is normal. Has anyone else ever measured the voltage of the circuit feeding a removed fixture on a 3 way light switch circuit with the switches in both positions? Perhaps there is a logical reason why I can measure ~25V or 120V but never 0V... Either way, I replaced the fixture and the light works properly, same as it has for the last 15 years+ I hope it's not drawing current :thumbsup:

darren 10-27-2006 09:38 PM

it could be voltage feeding back, known as backfeed in the business.

I can't explain but maybe somebody like md will be on and teach us more about backfeed.

sootybuttercup 10-28-2006 02:09 PM

I've seen this a lot of times, especially in industry. It is likely "induced" voltage that you are reading. When cables run parallel, in close proximity, the voltage can transfer from one cable to another (by magnetic fields) creating a measurable voltage. When an induced voltage is present on a wire, more often than not it is capable of producing almost no current and disappears when anything is connected to it. Depending on the quality and sensitivity of the meter being used to meaure this voltage, the readings can vary quite a bit as well.

joed 10-28-2006 02:50 PM

You are using a digital meter. They are known to give false readings like this. It is called phantom voltage. If you use a meter with a needle it will probably show 0 volts. or if you measure the voltage with the light fixture attached you will get 0 volts.

sootybuttercup 10-28-2006 02:59 PM

joed is correct....a digital (LCD) meter will show these induced (phantom) voltages when an analog (needle) meter usually won't...because there isn't enough "oooomph" there to even move a meter needle. The readings aren't actually "false" because a voltage is really there...but for all intents and purposes, especially in a house, an induced voltage is not a problem.:whistling2:

mdshunk 10-28-2006 07:06 PM

When you get these few volt "phantom voltage" readings, try holding the metal parts of your meter probes and the voltage will disappear. Don't try this with full 120, because you'll get the crap shocked out of you. When you're pretty sure that you just have a few volts, and suspect phantom voltage, putting the load of your body in that circuit will make your meter read zero. A little resistor will do the same thing. The old Wiggington Industries style "Wiggy" meters used a solenoid inside to make the meter indicator move, and would never falsely show "phantom voltage" because of the current required to move the wiggy.

satsu 10-28-2006 09:19 PM

Wow! If I had turned off the circuit I would have never learned all of this interesting information-well worth the tiny shock I received :) I think the phantom voltage is being magnetically induced in the switch itself rather than the parallel cables in the romex because there is no voltage on either traveller until it passes through the switch. After the switch one traveller has 120V and other has the induced ~25V. Either way, touching the metal probe parts (adding resistance) caused the voltage to dissapear (as noted by mdschunk) so the diagnosis is definately induced voltage w. 'no oomph'/negligible amperage.

I assume adding the resistance of a 120V light bulb also dropped the voltage to 0V so that the light bulb is not doing any work (heating up) nor wasting any electricity when it is not lit.

Thanks so much for all the replies and the good info. Obviously I have alot to learn and sometimes my curiousity leads me to second guess myself. Getting shocked definately had me second guessing, so I'm gratefull that you all were able to put my worries to rest. Maybe if I stand parallel to a bunch of electricians I can experience induced information and phantom intelligence and I won't have to bother you all so much in the future. Great site here and great participants!!


sootybuttercup 10-29-2006 11:56 AM

No bother...glad to help !!!!:thumbup:

roadhouse1301 11-10-2006 01:33 AM

I noticed the same phenomenon today while working on my 3 way switch. Mine was 25 volts and after a little pussyfooting around I HAD to lick my fingers and grab ahold of that sucker. No jolt. I am happy to have seen this post as I was suspecting bad switches.

sootybuttercup 11-10-2006 07:24 AM

A tip for anyone using their body as a voltage tester...use 2 fingers on the same hand and make sure no other part of your body is touching anything that may be grounded (wet floor, copper pipe, neutral wire, etc). The last thing you want is voltage passing through your heart (if you grabbed a wire in each hand, or were standing on a wet floor in regular shoes). I'm not recommending "human voltage testers" but, like any other dangerous thing, there is a right way and a wrong way...if you insist on doing it.

iGotNoTime 11-10-2006 12:14 PM

Sorry if you consider it spam, but I just wanted to add that it was a great link redline had posted. I learned alot from it, and appreciate it.

I have a problem with a 3-way switch in the master bedroom. When one switch is shut off the other will not work to turn it on. It seems to go in both directions. I will use the link to try and fix it when I am brave enough to mess with electricity some day. Water, close quarters and even heights don't scare me, but honestly I am scared of anything with a wire. I always check the breaker 2 or 3 times before touching anything LOL.

eric932 11-10-2006 04:12 PM

just for the record, from what i understand, induced voltage is going to be pretty well negligible in a dinky closet light circuit. plus the wires running parallel in a cable assembly will act capacatively as well, cancelling out some degree of inductive effects. i don't think i know anyone that would take induction and whatnot into account in a non-industrial setting. damn all phantom meter readings.

sootybuttercup 11-10-2006 05:23 PM

Eric...induced voltage is a factor anywhere it causes problems or confusion. In the home setting, it likely only leads to confusion with "false" voltage readings showing up on digital meters to confuse the DIY'er. In an industrial situation, it can cause shocks, loss of automated control, data corruption, and all sorts of problems. Capacitance is irrelevant when talking about voltage induced from one circuit into another.:whistling2:

All times are GMT -5. The time now is 01:31 PM.