Another (US) electrical wiring problem.

Posted by: wfaulk

Another (US) electrical wiring problem. - 18/03/2007 20:14

For a while now, I've had a switched receptacle where only one of the outlets worked, so I dropped by the hardware store to pick up a replacement.

The first thing I did was measure the voltage at the receptacle. With the switch off, the upper receptacle measured 121V. The bottom receptacle measured 28V. WTF? With the switch on, it measured 121V in both receptacles.

At this point, it seemed that the formerly bad outlet was now working. Later on, it appeared that there was an intermittent connection in a wire nut, which I fixed. So, technically, I solved my problem, but I'm really concerned about 30V running through that outlet.

Anyway, here's how it's wired: The receptacle was evidently intended to be wired with one outlet switched, the other always on. The always-on outlet was the one not working at all. There are two cables coming into the box: one 2-conductor plus ground (B, W, G), and one 3-conductor plus ground (B, W, R, G). The Gs are all tied together and then connected to the ground screw on the outlet. Each W is tied to a neutral screw on the receptacle. The B from the 3-conductor cable is tied to the switched outlet's hot screw. The R from the 3-conductor cable and the B from the 2-conductor cable are tied together, and from there go to the hot for the always-on outlet.

At the switch panel, there is the outlet's switch, plus a switch that controls the porch light. In that box, there are four cables, three 2-conductor cables (let's call them A, B, and C) and one 3-conductor cable (D). The grounds are all tied together and attached to both switches. The whites are also all tied together. AB goes to the bottom of the porch switch. BB, CB, and DR are tied together and then go into the bottom of the outlet switch and the top of the porch switch. DB goes into the top of the outlet switch. This seems completely screwed up to me.

I'm not sure what is connected to what, though it seems that the 3-conductor cable is the same one in both locations.

Does anyone have any ideas what's going on? I'm starting to suspect a bad switch.

I'm going to see if I can post a wiring diagram in a bit, since the textual explanation is a bit hard to visualize.

Edit: changed to reflect common terminology
Posted by: mlord

Re: Another (US) electrical wiring problem. - 18/03/2007 20:23

My mind is having trouble reading your posting, because what you actually have are more commonly known as a two-conductor (plus ground) and a three-conductor (plus ground).

Arghh.. it's taking me forever to parse your posting and make that correction in my mind..
Posted by: mlord

Re: Another (US) electrical wiring problem. - 18/03/2007 20:33

Okay, I don't know about the 30V (but that's very bad).

However, for your first (outlet) box:

The 2-wire cable is delivering electricity from the panel. The bare ground wire should be directly bonded to the metal box, and go from there to a pigtail with the bare ground wire from the other cable. Black is hot, White is neutral. All is good.

The 3-wire cable goes from there to the switch. For some odd reason, they chose the red wire to continue the hot run, rather than black. White is still neutral. The Black wire acts as a hot return from the switch to the switched outlet.

At the switch box, the 3-wire cable (above) is delivering electricity from the panel on the Red wire, via the outlet box discussed above.

From here, one 2-wire cable goes to the porch light -- the black for this is connected to the incoming Red (hot) wire. Okay.

The other two 2-wire cables continue the circuit onwards to additional outlets elsewhere.

All perfectly normal, except I would have swapped the red and black conductors (at both ends) for the 3-wire cable. No biggie, the code usually allows either method.

EDIT: As elperepat pointed out, the circuit could be the other way around, with the outlet box being downstream from the porch switch, and not the other way around. This doesn't really affect anything, though, other than which bare ground wire gets hooked directly to the screw on each metal box.

-ml
Posted by: mlord

Re: Another (US) electrical wiring problem. - 18/03/2007 20:37

Mmm.. I suppose the 30V at the not-so-off outlet could be caused by induction, since the switched black wire runs alongside the hot red wire for (how far?) a decent length. Mmm.. I oughta be able to measure that somewhere here just to see if induction is a real possibility or not with this.

-ml
Posted by: elperepat

Re: Another (US) electrical wiring problem. - 18/03/2007 20:43

[edit](mlord wrote his replies while I was writing this)[/edit]

Main power comes from either BB or CB.

Red is always live. It brings power from the switch box to the outlet box, then goes somewhere else.

Black is switched. It is powered by the switch. The fact that line is on the bottom of the switch is not usual, but it should work.


The 30V you're seeing may be caused by induction (the two wires parallel are long enough to act as primary/secondary of a transformer). Try plugging something (night light, lamp) in the switched receptacle and just pull it back enough to measure voltage across prongs. If it's induction, voltage should drop to 0V. It should not cause problem.

A drawing would probably help though.
Posted by: wfaulk

Re: Another (US) electrical wiring problem. - 18/03/2007 20:46

Okay, so you think it's all wired properly. Fair enough. Seems kinda weird to me. Maybe I was initially thinking that there would be two hot cables coming into the switch box, and tying all those neutrals together seemed odd. But I guess that's probably not the case.

The distance from the switch to the outlet is about 15 feet. There's currently nothing plugged into the outlet, so there shouldn't be any, uh, flux(?), to cause induction, right? I mean, effectively, it's all open circuits.

I think I'm going to remove the outlet switch and test its resistances.
Posted by: mlord

Re: Another (US) electrical wiring problem. - 18/03/2007 20:46

Quote:
My mind is having trouble reading your posting, because what you actually have are more commonly known as a two-conductor (plus ground) and a three-conductor (plus ground).


The reason for this, by the way, is that electical codes seldom consider the bare ground wires as "conductors". They're just present for safety, and under normal circumstances don't conduct anything. And they usually don't count for "box fill", whereas real insulated conductors do count for "box fill".

The term "box fill" refers to electrical code limits on how many conductors are allowed to enter/exit a given-size electrical box, based on the shape and cubic-inch volume of the box. The bigger the box, the higher the "box fill" limit.

A typical deep wall outlet box has a fill count maximum of 6, which allows for a circuit to enter on 2-wires (plus bare ground), exit on another 2-wire cable, and have a third 2-wire cable venture off to a switch and back again. Total of 6 "conductors" in the box, plus the three bare ground wires.

Fill count gets more complicated than that, as it also can take into account how many plastic wire connectors are present, plus other details. EDIT: I actually fibbed above: the box fill for that example was really 8, because two plastic wire connectors are needed to hook it all together.

Cheers
Posted by: mlord

Re: Another (US) electrical wiring problem. - 18/03/2007 20:48

Quote:

The distance from the switch to the outlet is about 15 feet. There's currently nothing plugged into the outlet, so there shouldn't be any, uh, flux(?), to cause induction, right? I mean, effectively, it's all open circuits.


Keep in mind that all of this is happening in the middle of a potentially long run of a single 15A (or 20A) circuit from the main panel. So something several outlets downstream could still be drawing current, which might then allow for some induction between conductors.

But I don't really believe that theory. And I'm just a software geek who once rewired his house.

Cheers
Posted by: wfaulk

Re: Another (US) electrical wiring problem. - 18/03/2007 20:55

Okay, I plugged a lamp back into the 30V outlet. It didn't light at all and there is no voltage across the prongs or between the hot and neutral screws. Unplug it and the voltage is back.

So you're telling me that this is "normal"?
Posted by: mlord

Re: Another (US) electrical wiring problem. - 18/03/2007 20:58

Quote:

So you're telling me that this is "normal"?


Only when wired with 3-wire cable, where the switched conductor lies next to a hot conductor for more than a few feet.

Cheers
Posted by: wfaulk

Re: Another (US) electrical wiring problem. - 18/03/2007 21:11

(All the boxes are plastic, so none of them are tied to ground.)
Posted by: wfaulk

Re: Another (US) electrical wiring problem. - 18/03/2007 21:52

Okay. All seems good now, then. I'm still a little freaked out at 30V on a supposedly dead outlet, but I guess there's not enough actual, um, electricity there to do anything with since it's just induced.

This raises an interesting point, though. How does this all work. Introductory electricity classes will tell you that I=V/R. I have a definite V and a definite R. What, exactly, is missing that prevents there from being enough energy to create the I? I mean, I have 30V of potential, and a 60W bulb (which, using my multimeter, is about 17 Ohms, and which I would assume is solely resistive). Why don't I have about 2A running through there?
Posted by: mlord

Re: Another (US) electrical wiring problem. - 18/03/2007 22:05

Quote:
Why don't I have about 2A running through there?


Because the inductive transfer isn't supplying anything more than a few micro amperes. Tiny supply. It takes practically nothing to show a voltage, so long as the impedance is high (impedance is based mostly on resistance, plus reactance).

Cheers
Posted by: wfaulk

Re: Another (US) electrical wiring problem. - 18/03/2007 22:07

I guess I've never understood how it is that sometimes current is a variable determined by the voltage and resistance and how sometimes it's a constant, as you're saying here.
Posted by: Neutrino

Re: Another (US) electrical wiring problem. - 18/03/2007 23:00

What kind of a meter did you use to measure the voltage? I'll bet it was a digital meter right? A digital meter has a very high input impedance. This is why your meter is reading a voltage on an open wire. Look closely at the switch recpt. You should see that one of the little copper tabs that ties the top and bottom of the recept has been removed. This is to allow only one of the plugs to be switched while the other is hot all of the time. Again this is normal. The switch in the wall should only effect one of the plug-ins in the switched recpt.
Posted by: mlord

Re: Another (US) electrical wiring problem. - 19/03/2007 01:00

Quote:
I guess I've never understood how it is that sometimes current is a variable determined by the voltage and resistance and how sometimes it's a constant, as you're saying here.


Well, in this case it really is a variable, but the important bit is that the voltage will also vary. As soon as a load is connected, allowing current to flow, the voltage drops to zero (or darned close..). When you plug that back into ohms law, you get zero (or close enough to it) current flow.

Things interact, there's no static fixed value here, other than the resistance.

Cheers
Posted by: wfaulk

Re: Another (US) electrical wiring problem. - 19/03/2007 01:24

I guess that explains how Ohm's law doesn't fail, but I've never been explained to how one knows what the voltage is going to be. Ah well. I'll stick to systems admin.
Posted by: wfaulk

Re: Another (US) electrical wiring problem. - 19/03/2007 01:30

Quote:
What kind of a meter did you use to measure the voltage? I'll bet it was a digital meter right? A digital meter has a very high input impedance. This is why your meter is reading a voltage on an open wire.

Well, if it's truly open, there should be no voltage at all. I guess the high impedance (yes, it is a digital multimeter) plus the inducted voltage means a much larger reading than I'd expect.

Quote:
Look closely at the switch recpt. You should see that one of the little copper tabs that ties the top and bottom of the recept has been removed. This is to allow only one of the plugs to be switched while the other is hot all of the time. Again this is normal. The switch in the wall should only effect one of the plug-ins in the switched recpt.

Yeah, I understood that part. I thought I'd mentioned the tab, but it probably got lost during the initial post. I tried to write it about five times, and I kept trying to go to the end of the line via Ctrl-E, which brought up Firefox's search page, and when I went back, all my text was gone. Fun!
Posted by: mlord

Re: Another (US) electrical wiring problem. - 19/03/2007 01:34

Quote:
Quote:
What kind of a meter did you use to measure the voltage? I'll bet it was a digital meter right? A digital meter has a very high input impedance. This is why your meter is reading a voltage on an open wire.

Well, if it's truly open, there should be no voltage at all.


No, think about an empty outlet -- it measures 120V even though it's truly "open" (inifinite impedance).

Then if you were to short it, connecting hot-to-neutral with something suitably heavy-duty, the voltage will fall to zero due to the zero impedance. Albiet rather briefly, before a breaker/fuse blows and/or something melts.

Cheers
Posted by: wfaulk

Re: Another (US) electrical wiring problem. - 19/03/2007 01:37

Well, I meant if there was no (intended) potential created due to connection to an electrical source between the two wires at all, which is what I think Neutrino meant, too. Clearly, if there is an intended potential, one will be read.

But I'm out of my depth enough that what I just said is likely to have made no sense at all.
Posted by: The Central Guy

Re: Another (US) electrical wiring problem. - 19/03/2007 02:10

This may not apply to your situation, but here's some advice I received from an electrician awhile back...

There's a couple of different ways to attach wires to a typical electrical outlet. The first way is to wrap the wires around the screw terminals (clockwise is best). The second way is to push the wire (with the end stripped a bit) into a small hole in the back of the outlet. The wire will then be gripped by a spring contact inside the outlet. Many times "pros" will use the second way, especially since it is easier and faster to wire up an outlet that way. Time is money, etc.

The problem is that these connections can become loose over time, disconnect, or connect intermittently and then get hot with current flow. This may depend on the quality of the outlet and conditions inside the box (wires bent - putting pressure on the connection, etc.).

My electrician friend told me that you'll never have any trouble with bad connections if you wrap the wires around the screw terminals...Every time I open up an outlet, I always re-do the wires and wrap them around the screws. So far, so good...I did have one that was a little intermittent before I fixed it

This may not be your issue, but I thought the info may be of value in the future....

Randy
Posted by: wfaulk

Re: Another (US) electrical wiring problem. - 19/03/2007 02:56

I wasn't clear about the cause of the intermittent connection. That was a wire that had come loose from a wire nut in the outlet's box. The insulation was kind of shiny, indicating to me that it may have been sparking in there. I think everything's okay now. Mostly I was just freaked out by the 30V current on the switched outlet while off, but people seem to think that that's okay, so I'm willing to go with it.
Posted by: wfaulk

Re: Another (US) electrical wiring problem. - 19/03/2007 02:58

By the way, thanks everybody. As usual, the BBS is the best, quickest source of information for just about anything.
Posted by: Neutrino

Re: Another (US) electrical wiring problem. - 19/03/2007 04:43

Good point Randy, for all of my home electrical work I choose to use the type of receptacle that has the clamp type of connection. You put the wire in the large hole and tighten the screw until the wire is clamped. These are more expensive but give the best connection of all. I’m not a residential electrician so time and money are not as important as safety.

This might help explain why there is voltage on the open wire. It is a very general description of how a transformer works. A transformer is basically two coils of wire wrapped around an iron core. The coil of wire that is connected to a voltage source produces a magnetic field. As the lines of magnetic force move across the second coil of wire they produce a voltage in it. The amount of voltage produced in the second wire is dependant on the number of turns of wire in each coil. Two wires laying next to each other will react in the same way. Because of the short lengths of the wires in your home and the distance between them the amount of energy transformed is very small. Your digital meter, while having a very high impedance, still places enough of a load on the open wire to pull the voltage down to 30 volts or so. While you are doing the measurement the open wire is really not open anymore. Your meter completes the circuit. The resistance in your body is much lower than that of your meter. If you were to touch the open wire and ground at the same time the voltage would drop down so far that you would not even feel it. All this only applies to AC systems. Also, you might have noticed that the 30 volts was not always there. For the first wire to pass a voltage to the second it must have a load on it to produce current. Without a load there would be no current and so no magnetic field. As the load goes up the strength of the magnetic field also goes up and the amount of energy transferred to the open wire goes up. Since your meter has a set impedance this would show up as the voltage measured going up as the load on the live wire goes up and vise versa. I hope this helps!
Posted by: elperepat

Re: Another (US) electrical wiring problem. - 19/03/2007 20:55

Quote:
I'll bet it was a digital meter right?

Old school electricians tend to use solenoid voltage testers like this one. These testers have low impedance which eliminate the "phantom voltages" caused by induction. Sure they are not precise to the sub-digit, but most of the time, they just want to detect if voltage is present or not and they do that job better than anything else.

The high impendace(resistance) of digital voltmeters is useful to prevent introducing errors in readings by limiting the "alternate path" of electrons through the instrument. Low resistance of the instrument would reduce the overall circuit resistance and thus reduce voltage. When testing home wiring, there's usually no need to bother with precision, so low impedance instruments is all right.


Concerning the "phantom voltage": Imagine the open wire as a current source instead of a conventional voltage source. It is able to supply no more than maybe 3 uA. When you close the circuit with a 10 megOhm instrument, this gives a 30V drop across that instrument. If you measure voltage on that wire with a solenoid voltage tester which have a 4 kOhm resistance, voltage would only be 0.012 V.

I hope I'm clear. It seems obvious in my mind using French. It's a bit more cloudy in English ;-)
Posted by: wfaulk

Re: Another (US) electrical wiring problem. - 20/03/2007 11:39

Quote:
Imagine the open wire as a current source instead of a conventional voltage source.

That makes sense, other than no one has ever indicated to me before that it was possible to think of current as the base value.