Poor Ground?

The main ground wire from the electrical panel in my house is grounded to the plumbing near my water tank. I have a well, however, the piping coming into the house from the well is plastic. Therefore, the only ground I have is the plumbing in the house. I recently noticed that I am reading low voltage off my neutral at several outlets. Could this be the result of the poor grounding of the main panel? Thank you! Ed

 

A low voltage reading off the neutral is normal, although I'd be interested in your numerical definition of "low". In any event, this is unrelated to grounding.

But on the subject of grounding, you need to add some grounding rods for proper grounding. The plumbing inside your house does absolutely nothing for grounding of your panel. Note however that the equipment grounding system serving your receptacles is a separate system. You can have great equipment grounding without any grounding of your panel. The two grounding systems serve separate purposes.

 

I was reading any where from 12 - 50 volts. If not normal. what would cause this? how about lose neutral some where?

 

The voltages that you referred to seem to be in the normal range depending on whether you are measuring a true neutral carrying the unbalanced load or a grounded leg. Again as John said the equipment grounding system is a totally different animal than a grounded or neutral system that carryies the unbalanced load back to the transformer. A neutral or grounded leg must be white or gray and insulated because it is a current carrying conductor. An equipment grounding conductor is supposed to have voltage on that green or bare wire only if there is a short to ground and only momentary voltage should be on that green or bare wire at any one given time.

The NEC requires you to bond all metal water lines that are not in direct contact with the earth to be bonded as one entity with the equipment grounding system of the main service rated panel [grounding bar] YOu must then also provide a grounding electrode such as a ground rod or any other approved grounding electrode conductors listed in 250.50.

Hope this helps

Wg

 

WG,
12 to 50 volts between an EGC and the grounded conductor is normal???? These two conductors are bonded together at the service disconnect. The only difference in potential between these two points would be the voltage drop due to current flow on the grounded conductor. If you have 12 to 50 volts between the grounding and grounded conductor that would mean that you have a total voltage drop of 24 to 100 volts on the circuit (the voltage drop on the grounded conductor is only half of the total circuit voltage drop). If the equipment on this circuit seems to be working OK I would expect an EGC that is not connected to the system and that you are reading a capacitive coupled voltage. If the equipment is not working correctly, I would expect to find a bad connection in the grounded conductor.
Don(resqcapt19)

 

Edgar you have no "ground".
Look outside for a ground rod. (bare copper out of the panel to the outside then down to the ground).
If your well has been replaced this could be why it is plastic.
What you are describing is a "bond".

WG I believe you must always bond all metal pipe.

 

Thanks everyone. My electrician is coming next week to check this out. I have posted more information on my other topic on this board "low voltage" if you are interested. I will be sure to post back and let you know what we find out. Thanks again! Edgar

 

Your right in having an Electrician check out your Grounding.....There should never be the kind of voltage your measuring on your neutral12-50 volts. In theory the voltage between the grounding conductor<the one attached to earth via a water pipe or ground rod> and the grounded conductor <the neutral, colored white> should be zero because they are firmly mechanically connected together at the service panel and again in the meter socket. You may get 1/2 to 1 volt readings between them at locations farthest from the panel because of voltage drop.

Good luck Mike BTW I'm a master licensed Electricain In Local 134 of the IBEW an it justs Ticks me off to read what some "experts" write.

 

I suggest a "Ground-Fault current" test to determine if their is a problem with the conductive paths. First. take an accurate voltage reading across the terminals of one of the receptacles where you read voltage Neutral-to-Ground.Next, plug a 1400 watt load-could be a toaster and iron-into the receptacle and read the voltage across the receptacle with the "Load" applied-it should be constant. Next repeat the test with the load conncected to the Grounding path which could be the Grounding conductor of non-metallic cable or the metallic path of armored cable. The Grounding path should comply with Art. 250.4 (5), effective Ground-fault current path. A drop in line voltage across the load indicates resistance in the conducting path.

 

Mike134, welcome to the forum. There are no experts here. Just a bunch of people who know quite a bit but not everything. Just about everybody says something incorrect here from time to time. Luckily, we each have different areas of knowledge. By working with cooperation and watching out for each other, we manage to get to the right answer eventually. Glad to have you as part of that.

 

John Nelson, I second your last statement, very well said.

Jxofaltrds, I agree that all metal plumbing must be bonded. I said what I said in an attempt to point out that if a water pipe is in contact with the earth that water pipe must be tapped within 5' of the entry of that contact with earth and use it as a grounding electrode with a back up grounding electrode listed in 250.50 installed in case the metal pipe in contact with earth is removed. Many people get confused that when you contact grounding to a metal water pipe you are using the water pipe as grounding source when most often you are just meeting the requirement of bonding the metal water pipe that is not in contact with the earth as one with the grounding electrical system. I guess I caused confusion in an attempt to avoid confusion by poor wording.

As for the argument that a maximum limit of 3 volts or less will appear in a voltage test between a neutral or grounded leg can not be accurate.

Try the following statement.

You have a multiwire circuit with one circuit pulling 20 amps and the other circuit pulling 2 amps. This condition in my opinion can easily be found in dwelling where multiwire circuits are installed especially in a kitchen. No my schooling tells me that when calculated you have 18 amps on that neutrar carrying the unbalanced load. If that is correct and you have no problem with that satement then tell me the volt amps present on that neutral conductor. Now remenber you are only going to have 2 or 3 volts present on that neutral per everyones opinion.


Next experiment, would be a 120 volt resistive load with 16 amps applied to that 120 volt branch circuit on the opposite side of your dwelling. Can you take an amp meter and clip around the white grounded leg wire and read amperage on that grounded leg serving that branch circuit with 16 amp load applied? If so how many amps did you read and what would the volt amps be considering that amp reading. Remember you only can have 2 or 3 volts per everyone's opinions per this post. What is the volt amps equal to those amps read on that white grouded leg.

If you have a neutral carrying 18 amps unbalanced load between two phases if serves as discribed above then how do you have amps without having voltage? If the voltage is only 2 or 3 volts max on that neutral white wire seems like the volt amps would be outragous.

If you are on the opposite side of a dwelling and you have a grounding conductor connected to the same point in the panel neutral bar, and you strike a jumper from your 18 amp loaded neutral carrying the unbalanced load of a multiwire branch circuit then would you get a spark. Wouldn't some of that voltage want to take an easier path on the empty unloaded no voltage grounding conductor. Doesn't the current seek the easiest path?

While I agree that inside the main panel you have the neutral and grounding conductor at one potential but that is the only place that single potential exists. when you are dealing with a structure that has 120 volt and 240 volts mixed within the same structure at different loads and different locations and different potential. Then you find the electricians connecting all the whites from all the branch circuits together within the 4 gang switch box then you are going to find the voltages discribed within that dwelling between the white and the bare or green.

While in a lab or in a factory where the loads are more evenly balanced and the white wires are not intermixed within the junction boxes and installed by a pro then you are probably right. IDEALLY speaking there should be no more than 2 or 3 volts read between the metal conduit or green insulated grounding conductor and the insulated white or gray wires.

Sorry to tell you but a dwelling is by far an ideal wiring system. We are mixing 120 and 240 in many areas in a certain appliance that can be activated at any given time. No plan nor reason as to what is going to be running at any given time. Changes constantly.

Ideally speaking you guys are partially right. In the real world you may be missing a few entities involved in a very dirty unstable load system of any dwelling.

Hope this helps

Wg

 

WG,
While I'll agree that you can have a lot of current on the grounded conducer, there shouldn't be much voltage when you read between a grounded conductor and an EGC. These two conductors are, as you stated, at the same voltage in the panel. When you read between these two points any where else you are measuring the voltage drop on the conductors. Since in a correctly installed system, there is not normally any current on the EGC, we are really measuring the voltage drop on the grounded conductor. Are you asking me to believe that having 12 to 50 volts of voltage drop on the grounded conductor of a 20 amp circuit is normal????? No way!! Think of it this way: We are connecting one lead of our volt meter on the grounded conductor at the receptacle and the second lead at the grounded bus in the service panel. This is really what we are doing, only using the EGC as an extension fro one lead of our meter. The only voltage between these two points is based in E = I/R. In this case R is the resistance of the grounded conductor,and its splices. Excessive voltage here indicates an overloaded grounded conductor or a poor connection.
Don(resqcapt19)

 

resqcapt19, you said "indicates an overloaded grounded conductor". My question is how can you have a grounded conductor that can overload being 20 amps or more yet only have 1 or 2 volts on that system.

Theory. 30 amp electric dryer with a 120 - 1/3 horse power motor. what is the volt amps on the neutral? Is there more than 2 volts present?

While I agree that the potential is zero at the panel, are you saying that the voltage goes back up the grounding conductor to the other side of the home on a 12 awg branch circuit rather than going back to the center tap of the transformer on a much larger wire.

It is my contention that the potential is not equal between the white and hte bare at end of line of a multiwire branch circuit such as that dryer.

My next contention is that you multiply the differences of potential throughout a multi outlet branch circuit heavily loaded will show quite a spark between the white and the bare a lot like electronic circuits with resisters in parallel. The current wiil rush down that second conductor if it can. The tester will register as an average of 14 volts in a dwelling in these conditions and will peak as high as 50 volts if the balance is off that much or if you have a highly inefficient appliance.

Doesn't look like anyone is going to come close to agreeing with my theory that I have seen for many years. Just have to keep my theory to myself.

Curiosity I thought would have someone experimenting in a messy house wiring design somewhere. But guess not.

resqcapt19, while I agree that your theory is correct that there should be little voltage on a neutral that is well balanced. There should be no amperage on that neutral either. In fact why do we call it a current carrying conductor and just wire like europe using to hot lines and few if any neutrals? Makes sense to me.

Sorry for my beliefs and experiences.

Wg

 

We don't need to speculate or go on hunches. The resistance of copper wire is very precisely known. In order to see 12 volts between the grounded wire and the grounding wire, you'd need to have a full 20 amps running for more than 300 feet (i.e., the 20-amp appliance would need to be electrically more than 300 feet from the panel). This is not a guess. It is a fact. This situation would be very, very rare indeed in a residence. So I think the simple math rules out the likelihood of ever seeing 12 volts on this wire. And to see 50 volts would require the full 20 amps to be running about a quarter mile -- a preposterous distance.

I'm not sure how we tangented off onto multiwire circuits in this discussion, but to see 12 volts on the neutral of a multiwire circuit would require the above lengths or more, and is therefore even more unlikely. I'm also not sure how a 30-amp circuit entered this discusson.

The bottom line is that 12 volts between the grounded and grounding conductors on a 20-amp circuit could never be considered normal on a general-use receptacle in a residence.

 

Wg said:
As for the argument that a maximum limit of 3 volts or less will appear in a voltage test between a neutral or grounded leg can not be accurate.

Try the following statement.

You have a multiwire circuit with one circuit pulling 20 amps and the other circuit pulling 2 amps. This condition in my opinion can easily be found in dwelling where multiwire circuits are installed especially in a kitchen. No my schooling tells me that when calculated you have 18 amps on that neutrar carrying the unbalanced load. If that is correct and you have no problem with that satement then tell me the volt amps present on that neutral conductor. Now remenber you are only going to have 2 or 3 volts present on that neutral per everyones opinion.

I agree you have 18 amps on the neutral. To answer your question lets assume we have a shared neutral circuit consisting of 100 feet of 12/3 w/gnd feeding two purely resistive loads, one on each leg. The resistance of No. 12 solid copper is about 2.0 ohms per thousand feet, so 100 feet is 0.2 ohms. Further assume two 120 volt transformer windings in series, with the neutral,or common point grounded as the source. So we have a two loop circuit with each loop consisting of a 120 volt source and a 0.2 ohm impedance representing the wire with some resistive load and the common between the loads tied back to the panel neutral through a 0.2 ohm impedance. We first have to solve for both loads to produce 20 amps in one loop and 2 amps in the other. Using node voltage analysis you will find that the 20 amp loop would require a 5.62 ohm load, while the 2 amp loop would require a 61.6 ohm load. The voltage drop on the white wire, measured from the panel neutral to the load common is 18 amps x 0.2 ohms= 3.6 volts. Watts (and in this case volt amps) loss on the white neutral wire is 18 squared x 0.2 = 64.8 watts. I'm fairly certain these numbers are correct because the sum of the losses and loads equals the sum of the inputs. Since the neutral, white wire is tied to the ground bus at the panel and assuming the bare ground is connected to the same bus and assuming the bare wire has no current, then you should read 3.6 volts between the white wire and the bare wire in this situation. If you want to try this at home make sure the wire is stretched out fairly straight, otherwise you will have some fairly significant inductive losses that I didn't count on.

Next experiment, would be a 120 volt resistive load with 16 amps applied to that 120 volt branch circuit on the opposite side of your dwelling. Can you take an amp meter and clip around the white grounded leg wire and read amperage on that grounded leg serving that branch circuit with 16 amp load applied? If so how many amps did you read and what would the volt amps be considering that amp reading. Remember you only can have 2 or 3 volts per everyone's opinions per this post. What is the volt amps equal to those amps read on that white grouded leg.


Yes, you will read 16 amps at all points on the white, grounded conductor, unless you have an accidental ground somewhere. Again, assuming 100 feet of #12 copper, the voltage at the end of the circuit between the white, grounded conductor carrying 16 amps and the bare grounded conductor carrying no current will be 16 amps x 0.2 ohms = 3.2 volts. Watts loss in the white grounded wire (assuming resistive load) will be 16 x 16 x 0.2 = 51.2 watts.


If you have a neutral carrying 18 amps unbalanced load between two phases if serves as discribed above then how do you have amps without having voltage? If the voltage is only 2 or 3 volts max on that neutral white wire seems like the volt amps would be outragous.

If current flows through wire there will be a voltage drop along the length of the wire. So says ohms law, unless you can find some wire with no resistance . Volt amps equals volts times amps, so 3 x 18 = 54. Not so bad for that kind of load.


If you are on the opposite side of a dwelling and you have a grounding conductor connected to the same point in the panel neutral bar, and you strike a jumper from your 18 amp loaded neutral carrying the unbalanced load of a multiwire branch circuit then would you get a spark. Wouldn't some of that voltage want to take an easier path on the empty unloaded no voltage grounding conductor. Doesn't the current seek the easiest path?

If the feeder is 100 feet of #12 copper, 18 amps load will produce 3.6 volts drop along the length of the neutral grounded conductor. Attaching a jumper between the neutral and the bare ground wire carrying no current may or may not produce a visible arc. But the neutral current would then be shared with the ground wire, cutting the voltage drop from the neutral at that point back to the panel in half, assuming good connections.

As for the remainder, it doesn't matter whether it's 120 or 240, ohm's law still applies. You just have to remember that when dealing with ac you have to consider phase angle. For example, in the very first two loop circuit, the 61.6 ohm load will have more than 120 volts on it. In fact, it's 123.2 volts. You can't just add ac voltages algebraically (sp?).

I'll bet the cursor off my slide rule that Edgar's ground is open.

Ichabod

 

I doubt we will solve this one to everybody's satisfaction, but here is an attempt to at least narrow the scope of the disagreement: Why are we focusing on multiwire circuits with shared neutrals, Warren? I submit that conditions on the neutral conductor of a 3-wire circuit with 20 amps on one phase and 2 amps on the other phase (ie: 18 amps on the neutral) are absolutely electrically identical to a 2-wire circuit drawing 18 amps (ie: once again 18 amps on the neutral).

 

If Edgarf is reading 20volts Neutral-to-Ground at a receptacle we will refer to as the "test-point", allow me to present this hypothesis; the resistance of the Ground path from the test-point to the service where the Ground path is bonded to the Neutral is 20 ohms. At some other point in the circuit there is a 100 ohm load connected "Hot"-to-Ground with an open Neutral connection .The 100 ohm load is in series with the 20 ohm Ground path and the current in this circuit is 1 amp. The voltage drop across the 20 ohm Ground path between the test point and the Neutral-Ground bond is 20 volts which is the meter reading at the test point. If there is a 2-volt drop in the Neutral between the test point and the service than the reading at the test point will be 22 volts- the VM is in series with 2 voltage-drops.

 

The question was asked why I brought in multiwire circuits. If you consider the subject that we are talking of "dwelling" most often you are not dealing with Master electricians. Just for giggles go to 5 different houses and open all the 2, 3 , and 4 gang switch boxes and look to see if all the whites in those switch boxes are connected under one wirenut. Then consider how many branch circuits both of the same phases and of opposite phases have their white wires connected together intermixed within those boxes. This is one point of dirty electrical design commonly found in dwellings. Then look at a multioutlet branch circuit and consider with several different loads intermittently coming on and off. Consider the constant changes in the conditions of load in a dwelling as equipment is run and shut off and lights turned on and turned off. YOu will almost never get the same amp reading on any one given branch circuit two days in a row if tested. Now consider all the whites intermixed in these dwellings by the unknowing people who didn't know of the meaning of a neutral and a grounded leg. Just how many people have you tried to explain the there is no neutral serving a 120 volt receptacle by itself. Kind of scary thought but more true than you know.

Then consider if there is two volts on a grounded leg and the equipment grouding conductor at that same point has 0 potential disregarding resistance of wire. Where do you suppose that voltage went. If you were an electron and you looked down the highway of that white wire that is full of other electrons and you have a bare wire connected to the same place that is of 0 potential then which road would you take maybe that empty grounding wire with 0 potential? Remember you are testing from a white that is loaded to a bare that is empty. YOu will pick up the readings discribed until that bare wire becomes equally loaded. Then you would read none. Remember all the metal in the house is connected to that unloaded bare wire.

While it is true that both the white wire and the bare wire is in contact with each other in the main panel. Just because you have voltage at end of line on that white wire does not mean that voltage will travel up that bare wire to end of line bacause they are connected together in the panel. That voltage as a choice to go down a much larger wire that has much less resistance that that little bare wire, it is called the service conductor and service lateral to the center tap of the transformer. The small voltage will pick the easiest route being the big wire going to the transformer center tap not back up the bare wire to end of line beside the loaded white wire. When you test to the 0 potential bare to the loaded white the electron will take off down that empty bare till that bare loads to equal value.

Then consider all the whites in all the junction boxes wrongly connected together and your theories just went down the drain. No rules then exist because you have too many unknown variables that are constantly changing.

This is why you can read this voltage same as you have all read posts saying they read voltage between two screws of the same switch. Ghost readings that constantly change. You do not really have the voltage they are reading you have loading and unloading due to different potentials.

Resistance and voltage drop has little to do with these readings that we are talking of.

Best explaination is dirty design. There is no stability in load within a dwelling due to diversity, difference of pototentials, etc.

What I have been referring to is not normal readings but often found in the dwellings.

Sorry I know most won't pick up what I am saying but you have all read my replies and over time have picked up my level of knowledge, no matter how weak it is. Try and give me the benifit of the doubt just a little and give me the credit that I have read exactly what the poster discribed many times over the years. The readings are ghost readings and have little to do with voltage drop but a lot to do with transient resistance and a lot more to do with differences of potentials.

Sorry I am on the wrong side of the fence again !

Wg

 

sorry for the long wait on this. and i want to thank everyone who replied. ok, checked the whole house and the mystery voltage was everywhere. checked all suspect circuits for bad connection and everything was fine. so... on to the panel. checked out ground and everything was ok. so... checked each circuit at the breaker. turning each one off one at a time and looking for mystery voltage. BINGO! found two breakers with voltage in off position, and they where feeding each other! had to shut both off to lose voltage. go to find out, had two circuits leaving panel to 2nd floor and connected directly to each other (in a junction box) in one big loop. extra home run feeds where mistaken for a junction to a circuit. oh the joys of owning a modular home. i feel like i asked you guys a trick question. i guess a bizarre problem like this, that has everyone divided with good theories... calls for a bizarre answer. thanks again. - ED

ps: dimer switches will cause a simular voltage in off position at the light.

 

Thanks Ed. Your bizarre answer was worth the wait. I can now understand why we didn't figure it out, and I'm amazed that you did. Good job!

 

What he found is exactly what I was trying to say. To use the word normal may have caused a reaction where I should have used the words commonly found. A very high percentages of homes are wired so that the white grounded conductors are intermixed and often times two breakers of the same phase are connected making a circle. This is what I was trying to point out that dwellings have a tendancy to have the nastiest wiring designs concerning wiring integrity there is. I do not condone it but am aware that this intermixing of circuits both hot and white are commonly inadvertantly done in dwellings by less skilled electricians. Hate to see it and it should be fixed when found but often gone undetected.

You did a good research and glad you found the culprit.

Good Luck

Wg

 

Edgar, Hopefully the rest of your house isn't wired this way.....I'd have guessed an open neutral causing the high neutrals voltages......your situation shows why diyer's should allways turn off all the power......good luck with your projects

Mike

 

I don't understand. Just because you had two breakers feeding the same cable(s) should not give you 10 to 50 volts from the neutral to the grounding conductor. There must have been something else going on.
Don(resqcapt19)

 

Don
I have to agree with you that what was found does not explain the stray voltage on the grounded conductor.

I have found overloaded neutrals that resulted from applying the current from two ungrounded conductors to a single neutral when there is NO difference of potential between the two ungrounded conductors. If the ungrounded conductor was back fed then there was far more current available to it's loads then the neutral could carry at an acceptable voltage drop.
--
Tom

 

Tom,
Almost all cases of excessive voltage (more than 2 or 3) between the grounded conductor and the grounding conductor are the result of excessive volatge drop in the grounded conductor as you have stated, or the EGC is not connected to the system.
If the voltage went away, what other changes did he make when he found the interconnected circuit?
Don(resqcapt19)

 

Don, I agree completely and I'm glad you didn't just let the discussion die when Edgar found 2 hot legs tied together and everyone was ready to accept that as the source of the mysterious voltage on the neutral. I can't see how multiple sources of power (or anything to do with shared neutrals, incorrectly tied into other circuits or not) can cause a significant voltage reading from neutral to ground. As many keep pointing out, the neutral and ground are bonded together in the main service panel (supposedly), so the resistance between those two points at the outlet where Edgar measured should be less than one-tenth of an ohm. The neutral wasn't melting, so we can asssume there wasn't a ridiculously high amperage flowing through it. Therefore a voltage reading of 20v or 50v or whatever means one of two things to me: 1) a poor neutral connection somewhere such that a decent percentage of the voltage drop in the circuit occurs on the neutral leg, or 2) a completely open neutral and the voltage Edgar saw was just a phantom reading (induced voltage on the neutral just from paralleling an energized hot conductor for 50 feet, or whatever). If, in the very beginning, Edgar had de-energized the circuit and used an ohmmeter to test neutral to ground, I'll bet he would have found either no continuity at all, or a significant resistance and we would have all agreed right then what the problem was.

(Edgar, if you tell me you did that test and found almost zero ohms, then we gotta call Ghostbusters).

PS: Wgoodrich, I have read hundreds of your posts and I have great respect for your advice. You are WAY more knowledgable and experienced than I am, but I have to disagree with your explanation on this one. (Hey, even Einstein had trouble with the cosmologial constant).