On a multi wire branch circuit [from a 208Y/120V panel fed from a 30kVA xfmr] with one breaker / phase turned off, there is still a reading of 24V, measured at the receptacle? Sounds to me like a neutral / grounding bonding issue but have not see this before? Thoughts??
MWBC - Stray Voltage
- Thread starter Mike01
- Start date
- Status
chris kennedy
Senior Member
- Location
- Miami Fla.
- Occupation
- 60 yr old tool twisting electrician
Gotta tell us meter used when you post this type of question.
- Location
- Placerville, CA, USA
- Occupation
- Retired PV System Designer
And, of course, what two points you measured between!
If there is no load on the phases with open breakers, a reading of 24 volts phase to neutral could just be capacitive or inductive coupling, seen by a high impedance meter. That kind of neutral to ground voltage, on the other hand, is never good regardless of what kind of meter you use.
Last edited:
cadpoint
Senior Member
- Location
- Durham, NC
You need to start turning off other breakers till the voltage goes away.
One could well have a incorrect shared neutral. One also might be getting a
reading from stray voltage, fluke sells a product SV225 thAt can be used
with their meters, this elimates this capacitance of which goldsigger is posting about.
One could well have a incorrect shared neutral. One also might be getting a
reading from stray voltage, fluke sells a product SV225 thAt can be used
with their meters, this elimates this capacitance of which goldsigger is posting about.
- Location
- Tennessee NEC:2017
- Occupation
- Semi-Retired Electrician
GoldDigger has an alias!:dunce::lol:
ATSman
ATSman
- Location
- San Francisco Bay Area
- Occupation
- Electrical Engineer/ Electrical Testing & Controls
Ha! Ha! :roll:
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
130818-1519 EDT
Mike01:
A multiwire branch circuit is one with a neutral shared with other circuits. If no current is flowing on that neutral, and the neutral and EGCs are properly terminated together at the main panel, then anywhere between the neutral and EGC there should be essentially 0 V. With a sensitive voltmeter you may see a small difference voltage. Normally unlikely to be anywhere near 1 V. If there was a large alternating magnetic field close by, then voltage would be somewhat higher. The intensity of the magnetic field, frequency, and coupling coefficient to the one turn loop consisting of the EGC and neutral determines this induced voltage. It could be in the low millivolt range.
I believe the circuit conditions you describe are:
1. Four wires in one cable. These being two hots on two different phases and separate breakers, a common neutral, and an EGC. These wires don't all have to be in a cable, but are moderately close to one another.
2. One breaker is turned off, and there is no connected load on that breaker.
3. The voltmeter is a high impedance DVM like a Fluke 27 or 87, and voltage on the floating wire is measured relative to neutral or EGC.
On a circuit going to my work bench with no load on either the hot or neutral I measure 38 mV between neutral and EGC with a Fluke 27. Add a 100 W bulb for a load and this increases to 76 mV. This increase is expected and results from the 0.83 A added to the neutral and the neutral impedance. I do not know the source of the 38 mV, could be inductive, but not capacitive. To get much signal on a loop in normal stray magnetic fields requires lots of turns and a large loop window.
Your relatively large stray voltage of 24 V is most likely capacitively coupled. In the real world to measure an induced (magnetic) voltage in a conductor requires a closed loop consisting of the meter and the wire. Something has to close the loop of the wire. In very simplified terms this might be the floating wire and capacitance from the wire to the EGC. This is a distributed closing of the loop all along the floating hot wire. The inductive effect will be small.
A 4 ft wire in free space in my basement reads about 0.9 V relative to EGC. This is a capacitively coupled signal. In my backyard the reading is 4 to 10 mV about 5 ft above earth relative to earth. In my breezeway where there are some power wires that go the garage and 5 ft above ground the reading is 50 or so mV and increases as I raise the test wire toward the ceiling. All these experiments are very uncontrolled.
The test wire is 4 ft of flat Romex with EGC between neutral and hot. The is in the range of 80 to 100 pfd of capacitance between the wires in the 4 ft Romex. The capacitive reactance at 60 Hz is about 30 megohms.
If the meter is referenced to the bench EGC, and the other meter lead is connected to one outer wire, the 4 ft EGC is floating, and 120 V is applied to the other outer wire the meter reads 42 V. Same setup but now unfloat the 4 ft EGC and connect it to the bench EGC, and the reading is 20 V. The grounded EGC in the cable becomes a partial shield between the two outer wires.
The further the wire with the 120 V is from the floating wire the lower will be the coupled voltage. For a fixed spacing of the two wires the longer the wires the greater is the capacitively coupled voltage.
If you have this type of capacitive coupling, and place a 100 W incandescent bulb across the meter terminals, then voltage should drop to near zero.
Measure your neutral to EGC voltage. If this is small, then you don't have a neutral problem.
.
Mike01:
Comments that I present below will be somewhat redundant to comments above.
A multiwire branch circuit is one with a neutral shared with other circuits. If no current is flowing on that neutral, and the neutral and EGCs are properly terminated together at the main panel, then anywhere between the neutral and EGC there should be essentially 0 V. With a sensitive voltmeter you may see a small difference voltage. Normally unlikely to be anywhere near 1 V. If there was a large alternating magnetic field close by, then voltage would be somewhat higher. The intensity of the magnetic field, frequency, and coupling coefficient to the one turn loop consisting of the EGC and neutral determines this induced voltage. It could be in the low millivolt range.
I believe the circuit conditions you describe are:
1. Four wires in one cable. These being two hots on two different phases and separate breakers, a common neutral, and an EGC. These wires don't all have to be in a cable, but are moderately close to one another.
2. One breaker is turned off, and there is no connected load on that breaker.
3. The voltmeter is a high impedance DVM like a Fluke 27 or 87, and voltage on the floating wire is measured relative to neutral or EGC.
On a circuit going to my work bench with no load on either the hot or neutral I measure 38 mV between neutral and EGC with a Fluke 27. Add a 100 W bulb for a load and this increases to 76 mV. This increase is expected and results from the 0.83 A added to the neutral and the neutral impedance. I do not know the source of the 38 mV, could be inductive, but not capacitive. To get much signal on a loop in normal stray magnetic fields requires lots of turns and a large loop window.
Your relatively large stray voltage of 24 V is most likely capacitively coupled. In the real world to measure an induced (magnetic) voltage in a conductor requires a closed loop consisting of the meter and the wire. Something has to close the loop of the wire. In very simplified terms this might be the floating wire and capacitance from the wire to the EGC. This is a distributed closing of the loop all along the floating hot wire. The inductive effect will be small.
A 4 ft wire in free space in my basement reads about 0.9 V relative to EGC. This is a capacitively coupled signal. In my backyard the reading is 4 to 10 mV about 5 ft above earth relative to earth. In my breezeway where there are some power wires that go the garage and 5 ft above ground the reading is 50 or so mV and increases as I raise the test wire toward the ceiling. All these experiments are very uncontrolled.
The test wire is 4 ft of flat Romex with EGC between neutral and hot. The is in the range of 80 to 100 pfd of capacitance between the wires in the 4 ft Romex. The capacitive reactance at 60 Hz is about 30 megohms.
If the meter is referenced to the bench EGC, and the other meter lead is connected to one outer wire, the 4 ft EGC is floating, and 120 V is applied to the other outer wire the meter reads 42 V. Same setup but now unfloat the 4 ft EGC and connect it to the bench EGC, and the reading is 20 V. The grounded EGC in the cable becomes a partial shield between the two outer wires.
The further the wire with the 120 V is from the floating wire the lower will be the coupled voltage. For a fixed spacing of the two wires the longer the wires the greater is the capacitively coupled voltage.
If you have this type of capacitive coupling, and place a 100 W incandescent bulb across the meter terminals, then voltage should drop to near zero.
Measure your neutral to EGC voltage. If this is small, then you don't have a neutral problem.
.
- Status