Current on waterpipe

[bjohnson]

This is a carryover from the old forum.

Quick rehash on the problem:

The waterline coming into my house has up to 5 amps on it. This continues through the ground wire to the panel to the grounded condutor on the service entrance. I've opened the main breaker to verify that the source was not from within the house or caused by an unbalanced load on the conductors and confirmed that the source was not from within my house. Also note that there was no current on the water pipe after the clamp for the ground wire and no current on any of the house piping. The current varies with it typically highest during the evening. My house is located closest to a pole mounted transformer.

My hypothesis is that one of my neighbors has a loose connection on their grounded conductor and thanks to Mr. Kirchoff, a portion of that current is returning to the transformer via the water pipe and my connection.

The city Electrical Inspector is at least showing an interest and contacted the POCO and relayed the information. The POCO came to the house to look. Unfortunately I was not there at the time and they only took readings on the conductors and chalked it up to the imbalance between the two. and came to my house to verify I knew what I was talking about. I also showed him the EMF readings from my gaussmeter that followed nicely along the waterline. The last word on the topic was that inspector wanted to study up on it a bit more and he'd get back to me.

My questions is about how to remedy the situation. The way I see it so far is one of four options: 1) Find the neighbor with the problem and tighten/troubleshoot the connection. Easiest and probably what I'll do. Any tips would be appreciated. 2) Install a dielectric couple on the waterline after the 10 foot point to satisfy the NEC requirements. The main drawback is this would require either cutting the concrete floor in my basement or digging 8+ feet in my front yard. 3) Disconnect the ground clamp. Not very good since there could potentially be a potential on the water pipe and under the right (wrong?) circumstances cause problems. 4) Do nothing.

EMF wise the basement has a linear field following the water pipe. On the floor the levels have been in excess of 50 mG (meter pegged). 3 feet off the ground will range from 10 to 30 mG. On the floor above readings were >5mg.

I will note that my gaussmeter is not known for accuracy. (Cellsensor)

And if Mr Riley is watching, I'd like to commend him on his book 'Tracing EMF's in building wiring and grounding'. I found it very informative.

[ February 26, 2003, 01:00 PM: Message edited by: bjohnson ]

 

[charlie]

Re: Current on waterpipe

I think you don't have a problem but have a perceived problem. What is the real problem with a few amperes on the water pipe? The inverse square law will take care of the EMF problem unless you are living on top of the water line.

The normal unbalanced current return path is the lowest impedance path and it will split with all parallel paths. This will include coming from all the neighbors who are served from the transformer where the current goes through the load, back to the service through the grounded conductor, through the grounding electrode conductor to the water pipe, through your water pipe, through your grounding electrode conductor, to your service to the grounded conductor, and finally to the transformer.

The bottom line is most of the unbalanced current will go back through the grounded conductor but some will go to other places. This includes your water pipe. If you wish to stop this current, you may install a non-metallic piece of pipe but, in my opinion, it would be to your own detriment.

 

[karl riley]

Re: Current on waterpipe

First a quick comment for Charlie: the magnetic field from a single conductor, as well as the net current field from the service conductors due to diversion of neutral to the water pipe, does not weaken with the inverse square of the distance. That would make it hardly a problem, as you assume. The field in these cases weakens only directly with distance, which means the field is huge compared to the inverse square situation, which holds true of power lines with no net current situation.

The fields measured in this case definitely qualify (in my opinion) as objectionable. If this field was in an office environment there would be a lot of jittery computer screens. I won't go into the health effects dialog, but 5 mG has now been convincingly associated with childhood leukemia by the WHO.

This is a typical situation encountered in residences connected by metallic water pipes. Your neighbor's neutral connections at his service and at the pole or transformer may or may not be loose or corroded, since some current will follow the parallel path through your house back to the transformer in any case.

The POCO cannot determine this with their standard measurement protocol since they do not disconnect the grounding connection to the water pipe when they measure the voltage differences which would identify a loose or corroded neutral connection. An ammeter will show the percent of neutral which is going to the water pipe compared to what is returning on the service neutral. But they don't use ammeters, at least in my experience.

If you put in a dielectric connector (or piece of plastic pipe) 10'+ outside the foundation (which I would do) you should notify your neighbor that since his service can no longer use your neutral, he should call the POCO to check his neutral connections after you put in the spacer. Of course he may be connected to other houses which use that transformer. In all such cases some connections are tighter than others, so the neutral is splitting all kinds of ways. But you don't have to be part of the weird chance distribution.

Digging in the basement? Must be 10' outside the house.
Disconnect the clamp on the water pipe? No way! It bonds the pipe throughout the house for basic safety.

Karl

 

[tom baker]

Re: Current on waterpipe

I used to be an electrician for a water utility, we often ran into "hot meters" and I would go investigate them. It was always caused by a loose neutral on the utilty side, from not being crimped or a tree rubbing thru the line. One meter was sparking to the metal meter box lid, the crews put in a non metallic nipple, the house then had 240 volts across all the 120 Volt loads, as a result the furnace motor burned up, and the owner was just home from open heart surgery.
Mike Holt has a fascinating video on this subject called "open neutrals" if you call the office they will send you a copy.

The american water works association has always been against grounding to water pipes for a while they had a non voting member on CMP5. I have a copy of a 1945 study done by the IAEI investigating water quality due to water pipe grounding, it stated no effects could be found.

It appears to me the biggest issue is with magnetic fields not the current on the water pipes.

 

[bjohnson]

Re: Current on waterpipe

Karl:

The digging in the basement approach is conceivable since the waterline physically enters the living area at the rear of the house. The city connection is at the front and the waterline travels under the basement floor. There is no easy access to the city waterline/house connection because it is buried at least 8 ft. The idea of opening the basement floor was concocted because I'll be cutting into the floor (at another location) to add a bathroom in the basement anyway. Since the waterline would be relatively close to the surface, digging would be minimized. Also, the waterline would be farther than 10 feet (..bonded to water pipe within 5 feet of entry and in contact with the earth for 10 feet...etc) required and I would be in compliance with the NEC requirements. Maybe kinda goofy, but in my book it beats the full excavation hassles and expense.

 

[don_resqcapt19]

Re: Current on waterpipe

bj,
If you do decide to cut the water pipe and install a nonmetallic section under the basement floor, please be very careful when you do this. If the pipe is acting as the grounded conductor for another house, you will have voltage between the two sections of pipe after you cut it. A plumber in a nearby town received a sever enough electrical shock to require hospitalization while repairing a water leak in someone's yard. He got between the two sections of pipe after he had cut them and the pipe was acting as the neutral because of an open neutral at the pole.
Don

 

[karl riley]

Re: Current on waterpipe

BJ, Jumper cables between the two parts of the pipe before cutting is the usual safety precaution.
I had assumed that the 10' in contact with earth had to be outside the foundation, but I don't know why I assumed that. If the non-bonded section of the water pipe under the floor is covered up and not accessible, I see no reason why that would not pass.
Karl

 

[bjohnson]

Re: Current on waterpipe

That's the way I read the requirement at any rate. Could be wrong, but the 10 feet in contact with earth should be met.

I plan to give the powers that be (city Electrical Inspector and POCO) some more time to see if they have any other solutions to offer. If not, come spring I'll make my changes if necessary.

Based on answers from the old thread, how long does the seperation need to be? Someone had mentioned 10feet or more, but that seems a bit excessive. I was thinking a simple 6" or so, but if there is a reason for long, please let me know.

Thanks.

 

[jxofaltrds]

Re: Current on waterpipe

BJ

I am going to assume that you have an overhead service. Have the local inspector and the poco meet you at your home. Have them perform this test.

NOTE: Before they do this explain to them the hazards involve, the return of others power on your neutral.

Have them cut your neutral (with your main off). Now have them take jumper cables and connect the cut neutral.

NOTE: Doing this may put anyone who might come into contact with your waterlines in danger. Tell them that this could cause harm to others (maybe) on an other service.

Having said this I would not do this. This will allow you to show them something as servicing your home could harm people.

Do you understand my point? Wish I had better writting skills.

Mike P.

 

[karl riley]

Re: Current on waterpipe

BJ, in basically dry soil the plastic threads of a stock-room dielectric coupler should be enough. If the soil is kept wet (soaked by sprinklers, etc) some like to use a foot or two of plastic pipe. I wish I had time to do some simple hands-on experiments with different soils, but so far I have had no appreciable neutral current flow after the above actions.
Karl

 

[karl riley]

Re: Current on waterpipe

Mike, before involving the POCO there is a simple measurement that can be made at the neighbor's service point if he/she agrees. With some loads on in the house,with a clamp-on ammeter measure the neutral current on the service neutral; then on the grounding electrode conductor to the water pipe. Compare the two. The water pipe amps are usually 1/3 or less of the service neutral amps. This is just a rule of thumb from experience. In this case you obviously don't have an open neutral. But if the water pipe was carrying say more than 50% of the neutral amps, maybe there is enough impedance on the neutral to cause a voltage imbalence when you cut off the water pipe path. So then I would advise the homeowner to call the POCO to check the neutral connections. Sometimes they are very cooperative. Depends...
Karl

 

[don_resqcapt19]

Re: Current on waterpipe

Karl,
Can you comment on Andre's post from the old forum? Thanks.

posted February 22, 2003 at 01:32 PM
--------------------------------------------------------------------------------
Don:
Isolation of water pipe system for one house with required bonding at service point(or "natural" bonding as copper pipe to refrigerator-ice maker), but not isolate at other house, can create galvanic action (with potential difference as small as 1v). It can contaminate drinking water with copper ions. "Green children" were reported few times in homes with isolated copper pipes.
It very hard to add requirement for simultaneous action (in every neighbour house) to the Code. Only isolation with long plastic pipe (long as tens yards ) would minimize the galvanic action because poor conductivity of water.
Current thru metalic pipes does not make the effect as long as all pipes are on the same potential (properly bonded).
Andre

 

[jxofaltrds]

Re: Current on waterpipe

Karl

Thank you for the info.

If a ground rod is installed, why couldn't you isolate the incoming water line? Am I read the code wrong? Does it say you must use the waterline for the electrode?

Mike P.

Karl check your MH email.

 

[bjohnson]

Re: Current on waterpipe

Ok...I've been doing some reading code wise....see what you think:

250.104(1) tells me I SHALL use the metal waterline as a grounding electrode.

250.53(D)(2) says I need a supplemental ground (e.g. ground rod)

250.52(A)(1) is the within 5ft 10 ft or more in contact with earth part.

However......If you take a look at 250.6 (Objectionable Current over Grounding Conductors) seems to give an out:

250.6 Objectionable Current over Grounding Conductors.
(A) Arrangement to Prevent Objectionable Current. The grounding of electrical systems, circuit conductors, surge arresters, and conductive non?current-carrying materials and equipment shall be installed and arranged in a manner that will prevent objectionable current over the grounding conductors or grounding paths.
(B) Alterations to Stop Objectionable Current. If the use of multiple grounding connections results in objectionable current, one or more of the following alterations shall be permitted to be made, provided that the requirements of 250.4(A)(5) or 250.4(B)(4) are met:
(1) Discontinue one or more but not all of such grounding connections.
(2) Change the locations of the grounding connections.
(3) Interrupt the continuity of the conductor or conductive path interconnecting the grounding connections.
(4) Take other suitable remedial and approved action.

250.4(A)(5) states:
(5) Effective Ground-Fault Current Path. Electrical equipment and wiring and other electrically conductive material likely to become energized shall be installed in a manner that creates a permanent, low-impedance circuit capable of safely carrying the maximum ground-fault current likely to be imposed on it from any point on the wiring system where a ground fault may occur to the electrical supply source. The earth shall not be used as the sole equipment grounding conductor or effective ground-fault current path.

This seems to negate the use of the waterpipe as a ground electrode (still bonded, but no cutting the basement floor required.)

Well....is it a possiblity?

 

[jxofaltrds]

Re: Current on waterpipe

BJ

That is how I read it also.

Mike P.

 

[don_resqcapt19]

Re: Current on waterpipe

Bj,
250.4(A)(5) has nothing to do with service grounding electrode systems. Service grounding electrode systems should not be playing any real part in the fault clearing process. To clear the fault, the current must go back to its source, not the earth. The code section you cited is to prevent the use of a grounding electrode in place of an equipment grounding conductor. If you use the grounding electrode system and the earth as the fault clearing path, you have a very dangerous installation. The earth's resistance is almost always too high to open even a 15 amp overcurrent protective device in the event of a fault. If a fault occurs on equipment where the earth has been used in place of the EGC, the metallic parts of the equipment will be raised to a potential much above the earth. The voltage drop across the earth is so high that if you are standing on the earth more than a foot or so from the grounding electrode and at the same time touching the faulted electrical equipment you will receive a serious or fatal shock. Look at Ground Rod Does Not Reduce Touch Potential . The grounding electrode system is to help provide protection from lightning and accidental contact with higher voltage system, and to stabilize the voltage to earth. See 250.4(A)(1). This is different from the required bonding of the interior metal water piping system. The bonding is intended to prevent the interior water piping system from becoming energized by a fault. It does this by providing a fault clearing path back to the grounded conductor via the bonding conductor and the main bonding jumper.

250.6 does not give you an "out" to eliminate the water pipe as a grounding electrode. It permits the elimination of "one or more" grounding conductors, but not all grounding conductors. It says nothing about eliminating a required grounding electrode.
Don

 

[bjohnson]

Re: Current on waterpipe

Don: I agree with you, 250.4(A)(5) has nothing to do with service grounding electrode systems. However, I mentioned 250.4(A)(5) because 250.6 listed it as a requirement that must be met to satisfy the requirements of 250.6(B).

The existing system has a connection to the waterpipe as the primary grounding electrode (250.104(1)). I also have a supplimental grounding electrode per 250.53(D)(2). I have an objectionable current caused by the use of the waterpipe as a grounding electrode. (Ok, it's actually caused somewhere else, but for the envelope of my house the waterpipe is the cause.) Does 250.6 allow me to isolate the grounding system from the waterpipe entering the house? This would still leave the grounding rod as the path the earth in the event of surge, lightning stike, etc which is the same purpose as the waterpipe to the street serves and meets the definition of grounding electrode (250.52(A)(5)). The water pipes within the house will still be bonded to the panel. Everything will be the same except my waterpipe will no longer be connected to my neighbors and the waterpipe will not be the primary grounding electrode.

The only major change will be the installation of the dielectric between the waterpipe as it enters the house and the water meter. Not to mention not having to cut a hole in my basement floor 10ft from where the waterpipe enters the living space.

 

[don_resqcapt19]

Re: Current on waterpipe

BJ,
In my opinion, you cannot at this time isolate the underground water pipe from the interior water piping and not use the underground water pipe as an electrode. We will know for sure in July when the ROP for the 2005 code comes out as I have submitted a proposal to permit you to do just what you propose. I see nothing in 250.6 that permits you to disconnect a required grounding electrode.
Don

 

[karl riley]

Re: Current on waterpipe

Seems like we have this topic pretty well worked out. First, Don: I can't comment on the quote from Andre because I honestly don't understand what he is saying, and I don't want to guess.

The question of removing a connection to a grounding electrode was hashed out years ago by CodeWatch, a Code newsletter, as allowing the water pipe to be removed as a grounding electrode, provided it was replaced by another qualifying grounding electrode (a second rod would do). However, the interior portion of the piping still must be bonded to the service neutral/ground point, so how do you prevent the current from continuing out on the water pipe? It would seem that you could then cut the pipe in the basement inside of the water meter (very convenient). But no, can't do, since you now have a short length of metallic pipe within the house that is not bonded to the system ground. Could become energized; thus prohibited.
So once again we get back to the necessity of cutting the pipe outside the house (or under the floor, if that flies). So if you cut it outside, why not do it 10' out so you don't have to add another electrode? If it's 8 feet down, that's difficult or expensive, but that's the Code. OK?
Karl

 

[bjohnson]

Re: Current on waterpipe

That makes sense Karl.

The city Inspector called me again today. We talked about different possiblities (apparently he's gone to the state level for assistance. I've created quite the stir around here.)

I'll check your book, but do you have any suggestions for determining if the current is coming from the transformer or from another home? The city Inspector would like to prove it's a POCO problem and vice versa. Short of making measurements at every house served by that transformer that is.