Open neutral?

[customelectric87]

A customer just called about a problem concerning her electrical service. She claims sometimes when exiting her pool, she feels a minor shock from touching metal railing. Also claims inside the home she has been shocked from touching the metal screws on switch covers. It's sounds like a neutral problem. I will be going there soon, just wondering if anyone has seen this before.

 

[CONDUIT]

Touching the screws on switch plates may be from static electricity. I have had this happen at my house from walking across carpet. The swimming pool should be checked out by someone that is knowledgeable and component with the equiopotential bonding requirements. They need to find out the source of the problem and determine if it is on the customers side or the utilities side. I would advise customer not to allow anyone in the pool until the issue is resolved.

 

[LarryFine]

A shock occurs when touching points with a voltage difference between them, but, when it happens between two surfaces that should be at the same potential, don't presume which one is the one being energized.

In other words, while it's most likely to be the service neutral conductor as suggested by the plate-screw shocks, it's not impossible for the earth itself to become energized, as opposed to bonded conductive parts.

I suggest obtaining an "independent" ground reference, such as the EGC of an extension cord p-lugged into a neighbor's house, against which to measure with a voltmeter. Also, check the L-to-N voltages for balance.

 

[customelectric87]

I talked with the customer more today and found out the plate screw shock only happend a few times....so its probably static. As for any other problems, she has a few outlets not working. That could be just a bad connection on receptacles. I'm thinking now its strictly a grounding issue with her pool. Will know something next week.

 

[CFL]

First place I would check is the pool panel if she has one. Check to see if the neutrals and grounds are bonded (common mistake for unqualified pool builders).
Then I would check the neutral connections at the main and at the meter. I always check with a meter first before tightening anything. That way you know if that ever was the problem. Once you tighten, you've destroyed the evidence.

If neutrals are bonded at the pool panel and there's a high resistance neutral connection somewhere, alot of current can take a shortcut through the pool. Of course there's many places it could be coming from, but this is one example that I've actually seen. A customer woke up one morning and found both of his full grown dobermans floating in his pool.

 

[customelectric87]

I went to the home yesterday and was unable to determine a solution. I checked the pool sub panel and found all connections in good condition. The grounds and neutral are isolated at pool panel. 100 amp feed to pool panel in pvc using underground aluminum conductors. The pool has a light and motorized pool cover, which are on gfci breakers. Pool also has a heater. I used an extension cord to check the difference in potential between the pool water and her pool panel and found no voltage from water to ground or neutral. She claims the shock comes from standing in water and touching the concrete along any side edge. The service neutral and grounding electrode conductor checked good using a clamp on grounding tester. The utility company has checked the service conductors and her pad mount transformer. I'm thinking the problem may be in her grounds to pool frame. Any ideas?

 

[Dennis Alwon]

How old is the pool? Does it have equipotential bonding? It is very possible there is an underground wire that is creating stray currents. Often times this issue is a power company problem not the ec's. Of course there can be other issues.

 

[customelectric87]

How old is the pool? Does it have equipotential bonding? It is very possible there is an underground wire that is creating stray currents. Often times this issue is a power company problem not the ec's. Of course there can be other issues.

The pool is 5 years old and it does have equipotential bonding. The homeowner installed the pool wiring, she claims the inspector approved everything. She has pictures of all bonding before concrete was installed. That's a good point, it could be utility issue.

 

[gar]

110504-1026 EDT

customelectric87:

When you have an extension cord from the main panel in the house to provide a ground reference from the the panel you need to use a high impedance meter, Fluke 27 for example, that can read millivolts.

You need to measure the earth around the pool at various points. Use a 10" or so screwdriver as a probe. Check points 20 ft away from the pool in a couple locations, then 5 ft away, and directly adjacent to the cement pad. Then the water voltage.

There will be some voltage differences. I would expect at least some 100 millivolt readings. I have not run experiments at what voltage I might expect tingling with totally wet hands and body, but it is at least less than 6 V.

From the description of your measurements I did not see where you measured the voltage from the concrete to the water.

.

 

[cdcengineer]

Women have a lower threshold to electric shock. They can feel lower levels than men. Perhaps there's a faulty low voltage pool light?

 

[customelectric87]

Women have a lower threshold to electric shock. They can feel lower levels than men. Perhaps there's a faulty low voltage pool light?

The shock occurred during the day, so the light wasn't energized. The light and pool cover motor are gfci protected, so I wouldn't think they would be the problem. What's the best approach to check the pool bonding grid?

 

[don_resqcapt19]

Does this pool have the perimeter bonding that is now required? Do some voltage checks beween the service grounding electrode and earth at least 20' away. Do this testing with both the main on and the main off. If there is a big change when you turn the main off you need to look for some problem on the load side of the service.
If there is no change it is on the utility side, but if it is just a few volts it is likely the voltage you are reading is the voltage drop on the utility primary grounded conductor. There is little you can do to eliminate that other than a neutral isolator, but most utilities do not want to install them. If this is the problem you will have to find a way to bond the perimeter surface around the pool

The issue is that we are connecting the pool bonding system to the electrical grounding system and we are actually energizing the pool bonding system with a voltage that is the voltage drop on the primary and secondary grounded conductors. When we bond the perimiter surface we also energize that to the same voltage so there is no potential between the water and the pool deck. Some studies show that a few volts between the water and the deck can almost prevent you from getting out of the pool.

 

[customelectric87]

Would installing a "water bond" elimate the difference in potential between the equipotential ground plane and the water? Maybe install a brass or stainless fitting in the pool water supply pipe and attach grounding conductor using water pipe clamp. Any suggestions? I understand this is a requirement for above ground pools as per NEC 2008.

 

[ptonsparky]

110504-1026 EDT

customelectric87:

When you have an extension cord from the main panel in the house to provide a ground reference from the the panel you need to use a high impedance meter, Fluke 27 for example, that can read millivolts.

You need to measure the earth around the pool at various points. Use a 10" or so screwdriver as a probe. Check points 20 ft away from the pool in a couple locations, then 5 ft away, and directly adjacent to the cement pad. Then the water voltage.

There will be some voltage differences. I would expect at least some 100 millivolt readings. I have not run experiments at what voltage I might expect tingling with totally wet hands and body, but it is at least less than 6 V.

From the description of your measurements I did not see where you measured the voltage from the concrete to the water.

.

Working at dairy barns years ago, if I remember correctly, with wet hands workers could feel less then 2v at times. Don't know about subemerged bodies.

 

[Dennis Alwon]

Would installing a "water bond" elimate the difference in potential between the equipotential ground plane and the water? Maybe install a brass or stainless fitting in the pool water supply pipe and attach grounding conductor using water pipe clamp. Any suggestions? I understand this is a requirement for above ground pools as per NEC 2008.

If there is an underwater light that is bonded then That will generally bond the water. Of course any metal stairs etc will also.

 

[customelectric87]

If there is an underwater light that is bonded then That will generally bond the water. Of course any metal stairs etc will also.

Guys, the customer just told me that no steel rebar was used in the concrete slab around her pool. She claims to have pictures during the construction showing all bonding attachments. She didn't have them in front of her when we talked, but she didn't remember any rebar. I asked if any copper wire was installed for an equipotential grid and she informed me it wasn't done....So that would explain the difference in potential between the pool and concrete....Now, whats the best approach to establish a equipotential grid? Install ground rods horizontality under her concrete slab and bond to her pool equipment? Or cut into the slab and install a #8 bare copper wire 3' from the pool edge for the equipotential grid? Thanks for all the input so far.

 

[Dennis Alwon]

You don't need rebar or steel in the concrete but you still need an EPB. Not sure how you can do that with the concrete poured. Tear it up I guess. You need the perimeter bond so.....

 

[customelectric87]

You don't need rebar or steel in the concrete but you still need an EPB. Not sure how you can do that with the concrete poured. Tear it up I guess. You need the perimeter bond so.....

Ok, so let's say I cut the concrete along entire edge of pool, insert a bare copper wire, bond to pool panel and fill cut with concrete patch... Do you think this will eliminate the difference in potential?

 

[gar]

110505-1449 EDT

Put a screwdriver in the ground next to the cement slab and measure the AC voltage from the screwdriver to the water. Use a high impedance voltmeter.

I do not know if you can easily measure the voltage to the cement itself.

On my basement floor if I just touch the test lead tip to the floor,then most of the signal I get is from my capacitive coupling to the test lead. This was in the 180 millivolt range. The reference is the EGC.

Then I put a slab of aluminum 8" x 24" x 1.25" on the floor and the reading dropped to about 45 millivolts. The thickness of the conductive plate is not too important other than to stay flat and make close contact with the cement.

Measuring the earth next to the cement slab should be a reasonable indicator of the voltage on the cement because the cement has a large area contact with the earth.

You indicated no voltage between the EGC and the pool water. What was the millivolt reading? I doubt 0 millivolts.

.

 

[don_resqcapt19]

Ok, so let's say I cut the concrete along entire edge of pool, insert a bare copper wire, bond to pool panel and fill cut with concrete patch... Do you think this will eliminate the difference in potential?

The code permits a single run of #8 solid copper between 18" and 24" from the edge of the pool. This is permitted to be run under or within the perimeter surface. This should do the job of rasing the voltage of the perimeter surface to match than of the pool water.