strange voltages in circuit


Old 06-12-07, 09:41 PM
Thread Starter
Join Date: Jun 2007
Location: Acton (near Boston)
Posts: 13
strange voltages in circuit

When I bought this house, almost 30 years ago, the builder pointed out one light switch,
and said it did not do anything, but was available in case I wanted to add outdoor lighting
sometime in the future.

The future is now, so I pulled off the cover plate to look, and something looked strange.
There are two red wires attached to the switch. One goes out of the box as part of a BWR
(black, white, red) cable, and the other goes to a wire nut to join 3 blacks and a white.
I figured I'd need a circuit diagram to understand what was what, so I got my meter and had
my first shock (surprise, not electrocution). One side was 120v, as expected. The other side
was 35v or 120v depending on the position of the switch.

So I looked closer and things got worse. There are three switches and four cables in the box.
The left switch controls a hall light. There are two other switches that also control the
same light. Leaving those switches alone, the red wire from cable 1 is hot, 120v, as one of
the two connections at the end away from the ground lug of the switch. The connection near
the ground lug is 120v or 0v, depending on the switch, and the wire is the red of cable 2.
The black of cable one is attached to the other connection point on the end away from the
ground lug. But it is at 0v or 50V depending on the position of the switch.

The white of cable one, the black of cable two, the black of cable four, the red to the hot
side of the supposedly unused switch, and a black wire to the hot side of the center switch,
are all joined by a wire nut. Another wire nut joins the white of cable 2, cable 3, and cable 4.
The center switch adds no surprises. The switched side is 120v or 0v.

Things got worse again, when I opened the other two boxes that control the same light as
the left switch. Both of them had a contact that read 0v or 50v depending on the position
of the switch. But the worst part is all three switches had only three connection points.
I thought the "middle" switch had to have four contacts. Does this seem like the cause of
the strange voltages? If so, is there any easy and clever way to identify the "middle"
switch? I can map the circuit eventually by disconnecting wires at various places. But that
will be a pain. The circuit operates a ceiling lamp in one room that is controlled
by two more switches and two half outlets in another room that are also controlled by two
switches, as well as one smoke detector (not sure about the others), the CO detector,
and several more assorted outlets.

While reading this forum to see if a similar problem had been discussed, I discovered
"phantom voltage". I was using a digital clamp meter, and calling fractional volts the same
as 0V, but I dug out an old analog VOM. It has perhaps one significant digit accuracy,
but the 50V was about half the 120v reading, and the 35v was about one third, so I think
these are not phantom readings.

There are no external signs of any problems. The lights operate as expected when switches
are moved. Some of the outlets on the circuit power a cordless phone, a receiver, and a
computer. All work as expected.

Thanks in advance for any solutions or suggestions.
Sponsored Links
Old 06-13-07, 05:46 AM
Join Date: Sep 2003
Location: Central New York State
Posts: 13,973
There is nothing wrong with your wiring. You do not understand how to use the meter you have, and worse yet you are using the wrong type of meter.

Put away the digital meter. It is the wrong tool for the job, especially if in your case. Use either an analog meter or a two wire tester.

You are reading through a switch. The 50 volts is phantom voltage. When the switch is closed you should see 120 volts, but when the switch is open there is no voltage on the other side.

What is it you wish to do?
Old 06-13-07, 08:17 AM
Thread Starter
Join Date: Jun 2007
Location: Acton (near Boston)
Posts: 13
readings confirmed with analog meter

First, thanks for the reply. I hope your conclusion is correct. I'd rather be a
worry-wort than have a problem.

But, as noted in my original post, I read about phantom voltage in this forum
and confirmed the readings with an analog meter.

This morning, the three switches that control one light are not in the same
position as when I took the original readings, but there is still a stray
voltage of about 30v on one of the terminals, using an analog meter.

The overall project is to install a light near the driveway. I decided to go
the low voltage way so I do not have to make such a deep ditch. I think
I know how to get through the wall and mount the weatherproof box
that the transformer will be plugged into. The hardest part of the project
seemed to be getting the wiring from the switch that was provided for
the purpose down to the new outlet, so that's what I looked at first.
I expected to find the switched side of the switch to be unused. Since
it is used I need to understand the circuit. I need to know what the
switch does, since it seems to have no effect.

I was using the voltage meter to trace where the wires went. It is a
lot more convenient than checking continuity with the power off.

If necessary, I can power the new outside light from a switch that
already powers the outside light near the entrance door. Even if the
strange voltages were bogus, I still need to trace the circuit to know
if I can use the switch that was supposedly left for my use. I need
to learn what the switch does.

As part of my effort to trace the wiring, I opened a junction box in the
basement. All the colors match. Two BW cables go to two outlets.
A BWR cable goes to a smoke detector where the R is capped. That
makes me wonder why run the red to the box. Perhaps the original
wirer ran out of BW cable. Or perhaps that is a of sign possible errors
elsewhere in the system.

I will systematically check all the combinations of switch settings and
the voltage readings, using an analog meter. That is sixteen settings
and many readings at each setting.
Old 06-13-07, 08:48 AM
Join Date: Sep 2000
Location: United States
Posts: 18,497
Smoke detectors are commonly wired with black-white-red cable. The red wire is uses as the interconnect so that when one detector goes off, they all do. If your red wire is capped, perhaps it is a wiring mistake, or perhaps the smoke detector has no interconnect, or perhaps you only have one smoke detector, or perhaps this smoke detector was wired in an addition with no feasible way to interconnect it to the others (but the red wire was run so that any future smoke detectors could be interconnected).

Even an analog meter can be subject to phantom voltage, but is less likely to be so. For an accurate test, you need some sort of test device that deliberately puts a load. That can be as simple as a $2 neon circuit tester.
Old 06-13-07, 09:03 AM
Join Date: Sep 2003
Location: Central New York State
Posts: 13,973
You cannot accurately test for continuity on a circuit unless all loads are removed. That means all light bulbs need to be removed, nothing can be plugged in, and any other hard wired loads need to disconnected (such as fans). For this reason, I recommend against testing for continuity.

Whenever you have a switch turned off and the other side is hanging you are suspect to stray voltages. These could be caused by other items on the same circuit. When a switch is off, the reading on the switched side means very little or nothing.

Without more information and actually seeing your setup, I cannot comment on the unused switch. Perhaps it might go that smoke detector, or perhaps it goes to the receptacles in the room but is unused. You will have to do some sleuthing or at least provide us COMPLETE information on the wiring in the box.
Old 06-14-07, 09:14 PM
Thread Starter
Join Date: Jun 2007
Location: Acton (near Boston)
Posts: 13

I will map out the entire circuit and report back.
This will get a few more switches and outlets converted from
back stabbed to screwed. It will probably take a few days.
Some of the boxes are behind furniture.
Old 06-17-07, 10:54 AM
Thread Starter
Join Date: Jun 2007
Location: Acton (near Boston)
Posts: 13
split outlet

While tracing the wiring, I noticed something I had not seen before.
This only affects the convenience of tracing the circuit, as it is easier
to remove and replace a backstabbed wire than to unhook and reconnect
a junction under a wire nut.

At an outlet box containing a normal dual outlet, one outlet is switched
and the other is unswitched. The connection between the two terminals
for the white wires has not been removed, and both white wires are connected
to the outlet. Is this good practice, bad practice, does not matter, or what?
If I should remove the connection, I will.

Old 06-17-07, 11:30 AM
Join Date: Aug 2005
Location: USA
Posts: 995
If they are backstabbed, they should be moved to the screws, but it is acceptable (when not a multiwire circuit with a shared neutral) to use the receptacle itself to make the connection of two wires.

Sometimes it's more convenient, sometimes not.
Old 06-17-07, 01:43 PM
Join Date: Sep 2003
Location: Central New York State
Posts: 13,973
Using the receptacle, and the connection between the screw terminals is one way to avoid using a pigtail to connect two (and only two) wires to the receptacle. It avoids the need to make a pigtail, and it is less to pack into the box.
Old 06-19-07, 11:06 PM
Thread Starter
Join Date: Jun 2007
Location: Acton (near Boston)
Posts: 13
problem solved. thanks.

Thanks for the assistance and the information. I've read about phantom voltages and I think
I'm prepared for the next time I encounter them. To get good readings on the "output" side
of a switch I can disconnect the wire and measure at the terminal. Tracing the circuit was an
adventure. It includes smoke detectors on three levels, a three way switch in the LR, a three
way switch in the DR, several lights in the basement, lights outside the LR door, several
outlets in the LR and basement, and the four way setup for the light in the entry hall with
switches there, in the kitchen, and upstairs. I found what I was looking for. The red wire
in one cable is capped in a box in the basement, so it will be easy to make a run to the
transformer for the low voltage outdoor lights. Even with all the different loads I think I'll
still be under 10 Amps on the 15 Amp circuit while the smoke detectors are wailing.
Backstabbing was eliminated. Thanks again.

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off

Thread Tools
Search this Thread
Display Modes