Hey guys... while helping a fellow apprentice I encountered what I believe to be this infamous phantom voltage effect and I just would like to make sure I didnt leave anything out.
He was replacing a faulty light fixture in the dining room controlled by a single switch. When he tested the switch he was getting 120 volts on and 48 volts off... I found that the cable leaving the switch to go to some unknown device was a 3 conductor with red, black and white... the red was the switched leg to the light and black a constant power going an outlet perhaps. I thought about phantom voltage and this seems like a perfect scenario where it could happen... I didnt try and measure if the black had a load on it.
I thought that as the travel in the wall for some distance that this could induce a voltage on the switched leg.
I connected a temporary porcelain fixture and the bulb didnt stay on while the switch was off(and initially read 49 volts) also didnt measure a current with it off so I concluded it was phantom voltage..
Am I right?
What other tests can I do to see if it is phantom voltage?
Does an energized conductor with no load induce a phantom voltage? or does it have to have a load?
The voltage reading is the result of a combination of inductive and capacitive coupling of the wiring in series with the high input impedance of the meter. The gut feeling about how much coupling you have can come from the percentage of the source voltage that you read and a look at your meter specs. In other words, if you have a 10 Mohm input z and are reading about 40 volts out of a 120 volt supply, you are probably looking at about 20 Mohms of coupling impedance. The voltages aren't purely additive though because of the voltage phase shifts across reactive components. This just gives you a ballpark concept of how much coupling you have.
Now folks here will start suggesting that you use a wiggy or Simpson 260 or some tester. I still think you would learn more by inserting known, parallel resistances, with your meter leads. How hard would it be to throw a couple resistors into your meter pouch. Consider pinching your meter leads and noting your skin resistance.(NO MEGGERS PLEASE!) Throw a resistor of about that value into your pouch as a test value. Now you have a ballpark idea of how much voltage might be dropped across you if you were added to the mix. But we're talking learning here, not job productivity. Joe
The first source is capacitive coupling. That's when an energized wire runs near (usually parallel) to the wire you're testing. The electrical charge on the energized wire causes electrons to move in the wire you're testing.
The second source is inductive coupling. That's when an energized wire runs near (usually parallel) to the wire your'e testing and the magnetic field from the energized wire induces a voltage in the wire you're testing.
The difference between the two is that one is caused by the electrical charge in the other wire and one is caused by the magnetic field from the other wire. The effect is the same.
In either case, the voltage you read is real. We call them "phantom" voltages because once you try to get any current from the wire being tested, the voltages go away. The tiny induced voltage is more like static electricity on the carpet than the voltage we're used to working with. You can measure thousands of volts of static electricity but there's just no current available. Once you touch your spouse and she gets a little shock, the voltage is gone.
So when you test these things with a digital volt meter, you can measure the voltage. That's because a digital meter does not draw any current. If you tested this with an analog meter, you'd measure zero volts. That's because an analog meter draws a tiny bit of current to move the needle. The wire you're testing can't provide enough current to run the meter and the voltage drops to zero. It's' like trying to start your car from a 9V smoke detector battery. If you measured the battery voltage, it would read zero since the battery sinmply couldn't keep up with the demand.
And that's what Joe's talking about when he says to use resistors. By placing a resistor between your meter terminals, you'll place a small load on the wire you're testing and will get zero volts.
I am really surprised meters for electricians don't have something on the order of 20,000 ohms per volt built in. (20 meg on a 1000v max meter) If you were probing on a core storage sense line circuit that might be a problem but it is not going to affect anything a regular electrician does.
I've always thought that the perfect meter would have 4 features that I don't have now:
1) a shunt button that does exactly what you suggest 2) a set of amp jaws that snaps off connected with wires that retract so I can hook the clamp on an awkward wire and still hold the display in my hand. 3) retracting leads so I don't have to wind them up and jam a wad of wire in my tool bag. 4) a small storage compartment to keep different lead tips, alligator clips, piercing probes, etc.
My Fluke 8060 came with a carrying case that held probes and tips. I made up a set about a foot long I left connected. The "clamp on" was separate, connected into the lead holes. Still no shunt. I just learned that a "floating wire" was a random number generator. Sometime that is a very useful thing to know. If waving your hand over the tips (no touch) changes the number you have a floater. Usually just grabbing the insulated handle will couple enough to change the display.
A very easy way of dealing with this is to keep a small analog voltmeter in the toolbag, one with a magnetic coil that has a much lower resitance than a modern solid state voltmeter. If you ever measure a voltage on the fluke that you suspect may be a phantom voltage, test it with the analog meter and see if if drops to 0.
In real life if you are just checking voltage and continuity a $5 HF meter is all you ever need. I use my "good" meter when I need to check current or a capacitor value. You can see if the capacitor works at all with the analog meter on ohms. If you have a known good one to compare it with you can get a pretty good idea if it has changed value. At least close enough to get a motor going.