well i was playing around with my digital multimeter the other day when i was bored and decided to test receptacles in my house. I realized I do not have a ground on any of my receptacles so i was testing from hot to neutral (which i couldnt be sure of because neither of the slots was bigger than the other, they both had the little slot for a horizontal blade like the ones for most 20 amp receptacles except it was on both sides, and I can't see the color of the screws because I'm not taking anything apart in this old house until I have some real experience with electrical.) Well from hot to neutral I was getting 122 volts so that was normal. than from what I assume was hot to ground ( using the screw on the metal faceplate as ground) i was getting 77 volts. than from what I assume was neutral to ground I was getting 37 volts. I asked my teacher and he said it was probably knob and tube wiring. I understand a little bit about knob and tube wiring but not why it would cause neutral to ground potential. Could anyone shed some light on this for me. Thanks =)
Neutral to ground voltage
- Thread starter Picclo
- Start date
- Status
cadpoint
Senior Member
- Location
- Durham, NC
Search and read up on stray voltage, also pay attention to the posters talking about certain DVM
and their own limited ability about these types readings based on these low meters. This one fluke was just posted here in the last few days/weeks,and presents an overview!
Not so sure about knob and tube, I've never seen any in working order to test something like it!
and their own limited ability about these types readings based on these low meters. This one fluke was just posted here in the last few days/weeks,and presents an overview!
Not so sure about knob and tube, I've never seen any in working order to test something like it!
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
090915-1001 EST
Picclo:
First some basics. Voltage is measured as the difference between two points. Current is measured at a point. Ohms law is a mathematical model relating voltage, current, and resistance for an idea resistor. In the real world practical devices differ from the theoretical models in various ways. Some are very close to the theoretical, and others may differ substantially. Various environmental factors may be significant.
A theoretical resistor has a constant resistance independent of frequency and applied voltage.
A theoretical capacitor is an insulator between two plates. On a steady state basis no current flows thru it with an unchanging applied voltage. On an AC basis the impedance (reactance for an ideal capacitor) is inversely proportional to frequency. Gets lower as the frequency increases.
A theoretical inductor is has zero resistance and its impedance increases linearly with frequency. Gets greater as the frequency increases.
You can view a capacitor as a conductor of current when AC is applied across it.
Ground refers to a connection to earth. In the radio or electronic area it may be the equipment chassis, and in some applications the chassis may not be connected to earth. This can be confusing but derives from the early days of radio. In the power electrical field ground is also a little confusing. To reduce this confusion somewhat the terms "ground electrode". "ground electrode conductor, GEC", and "equipment grounding conductor, EGC" have been created. You can look-up the definitions.
If you say something is grounded, then this implies that there is a low impedance path from that something to the earth. But also note that earth itself may not be a low impedance compared to the various grounding conductors to that earthing point.
Because you have a metal outlet box and metal cover does not mean that that metal is connected thru a low impedance to the earth.
To make the measurements with which you are experimenting I suggest that you get a long extension cord, and wire a temporary 3 prong receptacle close to the main panel for reference points. Take a 3 prong female connector from an extension cord with maybe 10 feet of the cord to make this test outlet.
You are in an old house. I suspect that the incoming water pipe is used as the ground electrode. If not then tell us what is the grounding setup. Assuming the water pipe is the electrode, then connect the ground pin of the reference receptacle to the water pipe where it enters the house. From the reference receptacle connect the neutral wire to the neutral bus in the main panel, and the hot wire to a breaker (fuse) that is an unloaded circuit or that all loads on that circuit can be turned off. Now you have three voltage reference points at or near the main panel. Plug the test extension cord into into the reference receptacle. Now you have long test leads from these three reference points for use in making voltage measurements anywhere in the house.
Assume you have a good high impedance voltmeter that can resolve 0.1 V on a range that can measure 120 V. Such as a Fluke 27 that I used.
First, I suggest that you connect the EGC pin of the extension cord to the common terminal of the voltmeter. With the other voltmeter lead measure the voltage to the main panel neutral. The voltage should be quite small. This morning mine reads 1.2 millivolts. If you get several volts, then you have some sort of problem. There should be very low resistance from the main panel neutral (neutral and ground should be connected together in the main panel or somewhere close by) to the entry water pipe, and there should not be a large current thru this bonding path. Thus, the voltage drop should be very small. Suppose the bonding path was 6 ft of #4 copper and there was 10 A of current to ground, then the resistance is 0.0015 ohms, and at 10 A the voltage drop is 15 millivolts.
Assuming you do not have a bad condition of ground to neutral, then switch to the the neutral connection on the extension cord as your voltage reference point. Go to some outlet and measure the neutral voltage at the outlet. If there is no load on this circuit, then the voltage should be near zero. Now connect a load to the other half of the same receptacle to which the meter is connected. I suggest a 1500 watt heater, about 12 A. If we assume the neutral is #14 copper and is 50 ft long, then resistance of neutral is about 0.13 ohms, and the voltage reading should be about 1.5 V.
Next hold the red lead free in space. I read 0.1 to 0.4 V. Put my fingers on the probe and it goes to maybe 0.5 V. Next I moved the probe along the side of an 8 ft Slimline tube. The max I read was about 1.7 V. All these are a result of capacitive coupling to the meter thru a very small amount of capacitance. Your wall plates on your receptacles are not grounded and you are reading some form of capacitive coupling, or you have insulation leakage.
Take another extension cord and connect one lead in the cord to a hot somewhere and leave the other conductor unconnected. The voltage between neutral or ground and the floating conductor will be quite large.
An experiment with a 50 ft extension cord.
Hot and neutral connected to the wall receptacle. EGC was floating. Hot to neutral 124 V. Neutral to EGC 58.2 V, and Hot to EGC 61.3 V. Basically a capacitive voltage divider.
Next only Hot was connected to the wall. Both neutral and EGC were floating. Cord EGC to cord neutral, 0.46 V. Hot to EGC or to neutral of the cord, 16 V.
Then measurements to wall neutral. Floating EGC of cord to real ground 101.3 V. And floating neutral to real ground 101.0 V.
Ponder these experiments and maybe create some of your own. Try a 250 roll of Romex.
There are other experiments to run relative to voltage drop to various outlets to evaluate the quality of the connections in these circuits.
.
Picclo:
First some basics. Voltage is measured as the difference between two points. Current is measured at a point. Ohms law is a mathematical model relating voltage, current, and resistance for an idea resistor. In the real world practical devices differ from the theoretical models in various ways. Some are very close to the theoretical, and others may differ substantially. Various environmental factors may be significant.
A theoretical resistor has a constant resistance independent of frequency and applied voltage.
A theoretical capacitor is an insulator between two plates. On a steady state basis no current flows thru it with an unchanging applied voltage. On an AC basis the impedance (reactance for an ideal capacitor) is inversely proportional to frequency. Gets lower as the frequency increases.
A theoretical inductor is has zero resistance and its impedance increases linearly with frequency. Gets greater as the frequency increases.
You can view a capacitor as a conductor of current when AC is applied across it.
Ground refers to a connection to earth. In the radio or electronic area it may be the equipment chassis, and in some applications the chassis may not be connected to earth. This can be confusing but derives from the early days of radio. In the power electrical field ground is also a little confusing. To reduce this confusion somewhat the terms "ground electrode". "ground electrode conductor, GEC", and "equipment grounding conductor, EGC" have been created. You can look-up the definitions.
If you say something is grounded, then this implies that there is a low impedance path from that something to the earth. But also note that earth itself may not be a low impedance compared to the various grounding conductors to that earthing point.
Because you have a metal outlet box and metal cover does not mean that that metal is connected thru a low impedance to the earth.
To make the measurements with which you are experimenting I suggest that you get a long extension cord, and wire a temporary 3 prong receptacle close to the main panel for reference points. Take a 3 prong female connector from an extension cord with maybe 10 feet of the cord to make this test outlet.
You are in an old house. I suspect that the incoming water pipe is used as the ground electrode. If not then tell us what is the grounding setup. Assuming the water pipe is the electrode, then connect the ground pin of the reference receptacle to the water pipe where it enters the house. From the reference receptacle connect the neutral wire to the neutral bus in the main panel, and the hot wire to a breaker (fuse) that is an unloaded circuit or that all loads on that circuit can be turned off. Now you have three voltage reference points at or near the main panel. Plug the test extension cord into into the reference receptacle. Now you have long test leads from these three reference points for use in making voltage measurements anywhere in the house.
Assume you have a good high impedance voltmeter that can resolve 0.1 V on a range that can measure 120 V. Such as a Fluke 27 that I used.
First, I suggest that you connect the EGC pin of the extension cord to the common terminal of the voltmeter. With the other voltmeter lead measure the voltage to the main panel neutral. The voltage should be quite small. This morning mine reads 1.2 millivolts. If you get several volts, then you have some sort of problem. There should be very low resistance from the main panel neutral (neutral and ground should be connected together in the main panel or somewhere close by) to the entry water pipe, and there should not be a large current thru this bonding path. Thus, the voltage drop should be very small. Suppose the bonding path was 6 ft of #4 copper and there was 10 A of current to ground, then the resistance is 0.0015 ohms, and at 10 A the voltage drop is 15 millivolts.
Assuming you do not have a bad condition of ground to neutral, then switch to the the neutral connection on the extension cord as your voltage reference point. Go to some outlet and measure the neutral voltage at the outlet. If there is no load on this circuit, then the voltage should be near zero. Now connect a load to the other half of the same receptacle to which the meter is connected. I suggest a 1500 watt heater, about 12 A. If we assume the neutral is #14 copper and is 50 ft long, then resistance of neutral is about 0.13 ohms, and the voltage reading should be about 1.5 V.
Next hold the red lead free in space. I read 0.1 to 0.4 V. Put my fingers on the probe and it goes to maybe 0.5 V. Next I moved the probe along the side of an 8 ft Slimline tube. The max I read was about 1.7 V. All these are a result of capacitive coupling to the meter thru a very small amount of capacitance. Your wall plates on your receptacles are not grounded and you are reading some form of capacitive coupling, or you have insulation leakage.
Take another extension cord and connect one lead in the cord to a hot somewhere and leave the other conductor unconnected. The voltage between neutral or ground and the floating conductor will be quite large.
An experiment with a 50 ft extension cord.
Hot and neutral connected to the wall receptacle. EGC was floating. Hot to neutral 124 V. Neutral to EGC 58.2 V, and Hot to EGC 61.3 V. Basically a capacitive voltage divider.
Next only Hot was connected to the wall. Both neutral and EGC were floating. Cord EGC to cord neutral, 0.46 V. Hot to EGC or to neutral of the cord, 16 V.
Then measurements to wall neutral. Floating EGC of cord to real ground 101.3 V. And floating neutral to real ground 101.0 V.
Ponder these experiments and maybe create some of your own. Try a 250 roll of Romex.
There are other experiments to run relative to voltage drop to various outlets to evaluate the quality of the connections in these circuits.
.
- Location
- Massachusetts
It sounds like you do not really have a grounding means at that receptacle.
if the grounding wires are made up, but floating, the voltage will read oddly due to the DMM issues combined with the capacitance created by the proximity of the open wires that are not grounded, as I found out in a post I recently made. I had said the electricity was absorbed somehow, which is actually the case (wrong word) when you consider capacitive charging,and my high impedence DMM added to the unpredictable voltage readings.
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
090915-2016 EST
danickstr:
If your cable lengths are moderately long, 50 ft maybe, then one might expect about 1180 pfd from hot or neutral to EGC. Based on a piece of romex with 3 #12 wires. The EGC is centered between hot and neutral and measures about 23.6 pfd per foot from either hot or neutral. The capacitance between hot and neutral with EGC floating is about 16 pfd per foot.
To a floating Romex EGC there is a very nice equal capacitance voltage divider.
The input capacitance of my Fluke 27 is about 30 pfd shunted by 10 megohms. This meter will have little effect on readings from a 50 foot length of Romex. Floating wire readings should not be too unpredictable if you have some knowledge of the circuit. The readings may actually tell you something about the circuit.
Knob and tube will have much lower capacitance between wires than romex.
If you have K&T or Romex without a ground and there is no resistive leakage to an outlet, then there is only a small capacitance between the hot components (wire and the receptacle) to the box. That capacitance and the input capacitance of the meter will determine the voltage reading. If the wiring to box was 10 pfd, the approximate voltage read would be about 25% of 120, 30 V. If the coupling was about 30 pfd, then 60 V. And for a coupling of 90 pfd, then about 90 V.
Your choice of the word absorbed is not a satisfactory description. It is not clear what you mean. Find a new word for what you are trying to say.
.
danickstr:
If your cable lengths are moderately long, 50 ft maybe, then one might expect about 1180 pfd from hot or neutral to EGC. Based on a piece of romex with 3 #12 wires. The EGC is centered between hot and neutral and measures about 23.6 pfd per foot from either hot or neutral. The capacitance between hot and neutral with EGC floating is about 16 pfd per foot.
To a floating Romex EGC there is a very nice equal capacitance voltage divider.
The input capacitance of my Fluke 27 is about 30 pfd shunted by 10 megohms. This meter will have little effect on readings from a 50 foot length of Romex. Floating wire readings should not be too unpredictable if you have some knowledge of the circuit. The readings may actually tell you something about the circuit.
Knob and tube will have much lower capacitance between wires than romex.
If you have K&T or Romex without a ground and there is no resistive leakage to an outlet, then there is only a small capacitance between the hot components (wire and the receptacle) to the box. That capacitance and the input capacitance of the meter will determine the voltage reading. If the wiring to box was 10 pfd, the approximate voltage read would be about 25% of 120, 30 V. If the coupling was about 30 pfd, then 60 V. And for a coupling of 90 pfd, then about 90 V.
Your choice of the word absorbed is not a satisfactory description. It is not clear what you mean. Find a new word for what you are trying to say.
.
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
090916-1301 EST
danickstr:
That is a good solution. I like it.
.
danickstr:
That is a good solution. I like it.
.
K8MHZ
Senior Member
- Occupation
- Electrician
hurk27
Senior Member
- Location
- Portage, Indiana NEC: 2008
So Incapacitated is the resistance side of the story, or is it the inductance side
:grin:
LarryFine
Master Electrician Electric Contractor Richmond VA
- Location
- Henrico County, VA
- Occupation
- Electrical Contractor
It is, and it's my opinion that the wiggy is the troubleshooting meter of choice most of the time. The voltmeter is necessary when you need to know the exact voltage, not just whether 'real' power is available.
hunt4679
Senior Member
- Location
- Perry, Ohio
Try a solenoid meter "wiggy" and see what you get for voltage
mivey
Senior Member
I vote for inductance.So Incapacitated is the resistance side of the story, or is it the inductance side
sparks2000
Senior Member
- Location
- Chicago,IL
I agree, use a wiggy!
sparks2000
Senior Member
- Location
- Chicago,IL
I just bought a condo, so the panel in my unit is a subpanel, I just noticed that they have the nuetral and ground bonded! I almost S#!T my pants! I s this why all my electronic dimmers keep burning up and malfunctioning?
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
091117-1419 EST
sparks2000:
Should not be a cause of dimmer failure.
Are the dimmers two wire devices? Take a defective dimmer apart and see what failed, and if it is or is not a three wire dimmer. Whether a two wire or three wire dimmer should not really make a difference, but it is still usefully information.
What were the symptoms of the failed dimmer? Always on? Never on? or Some sort of flashing problem?
Measure the voltage from any EGC point to a water pipe, assuming the water pipe system is metallic back to earth, or put a short rod in the earth for your reference point. Do this while making a large load change on any 120 V circuit. A 1500 W portable heater may be an adequate large load for this experiment.
Use a digital meter that resolves 0.1 V on a range that covers 120 V. I believe Fluke meters automatically range change at base 10 multiples of 3 V. So 120 V would read on a 300 V range.
If there is a bad neutral, then you may detect this by an excessive shift of the the neutral voltage relative to earth.
Or you can use two meters, one on each phase of the supply, to monitor for a neutral problem. Also 25 to 50 W, both the same wattage, bulbs can be use instead of the meters to look for a neutral problem.
For a dimmer to be damaged from over voltage I will guess it needs to go over 150 V unless very marginal components are being used.
I have two dimmers that have operated continuously at low output for 43 years and no failure.
You need to find out why your neutral and EGC are bonded at the subpanel.
.
sparks2000:
Should not be a cause of dimmer failure.
Are the dimmers two wire devices? Take a defective dimmer apart and see what failed, and if it is or is not a three wire dimmer. Whether a two wire or three wire dimmer should not really make a difference, but it is still usefully information.
What were the symptoms of the failed dimmer? Always on? Never on? or Some sort of flashing problem?
Measure the voltage from any EGC point to a water pipe, assuming the water pipe system is metallic back to earth, or put a short rod in the earth for your reference point. Do this while making a large load change on any 120 V circuit. A 1500 W portable heater may be an adequate large load for this experiment.
Use a digital meter that resolves 0.1 V on a range that covers 120 V. I believe Fluke meters automatically range change at base 10 multiples of 3 V. So 120 V would read on a 300 V range.
If there is a bad neutral, then you may detect this by an excessive shift of the the neutral voltage relative to earth.
Or you can use two meters, one on each phase of the supply, to monitor for a neutral problem. Also 25 to 50 W, both the same wattage, bulbs can be use instead of the meters to look for a neutral problem.
For a dimmer to be damaged from over voltage I will guess it needs to go over 150 V unless very marginal components are being used.
I have two dimmers that have operated continuously at low output for 43 years and no failure.
You need to find out why your neutral and EGC are bonded at the subpanel.
.
- Status