Pool ground grid??? and aux. ground rods

[Lightman]

After reading other thread about ground rods at a subpanel,, thinking about pool grounding. (working life at utility), three wire service to house, neutral and all cable, phone shields tied to one point at service entrance / meter base, per NEC.

Had a few call outs for measurable voltage from pool water to pool deck in the past.

Always had detectable voltage from neutral / ground wire at meter base, guy wire, fence, to a screwdriver stuck in the ground few feet away. Even between two ground rods three feet in dirt few feet apart??? various location even in middle of field miles from houses or powerlines.

Question is after the sub panel at pool controls, four wire feed from main service panel, does the motor frame and the equipotential zone around the pool including the bond to the water, tie back to the fourth ground wire, EGC, in the control subpanel?

Most problems seemed to start when fiberglass pools and fiber concrete started being used, few problems with either concrete shell, or liner pools with metal forming walls. One location, same transformer side by side houses, older metal walled pool with plastic liner, rebar in concrete no voltage, new pool fiberglass no rebar in concrete, with ground wire under concrete around pool, about 12 volts pool water to wet concrete 2 foot from water.

I have built and worked on grounding systems for substations, cadwelds, ten foot grid with 40- 50 foot sectional ground rods at driven and cadwelded each intersection.

Most pool service referred to the discussion about diffrence between equipotential zone around pools being diffrent than the normal ground and they are sometimes isolated. from the EGC?

 

[jkwilson]

You aren't dealing with a grounding system, but a bonding system. A bonding system will function whether it is part of the service ground network or not. In the US the bonding system of the pool is not necessarily connected to the service ground at all, but in Canada a connection between them is required. In reality a bonding system that connects to the metal case of the pump and all of the other things is incidentally connected to the service ground.

In your example, the problem is the lack of rebar in the deck. The rebar is supposed to be connected to the rest of the bonding system. Without rebar or a metal grid, there is no way to bond a concrete deck.

In order for there to be a detectable voltage in the pool, you need two things. First is current flowing through the earth or through the pool. The second is an incomplete bonding system. Without a bonding system, you will not have voltage if there is no current. Without current, you won't have a voltage whether there is a bonding system or not.

Many times the source of the current in the earth is inadequate grounding on the distribution network and a kind of disconnect between the NESC and the NEC. From your perspective on the distribution side, ground and neutral are the same thing. When you hook up service, you are required to connect the distribution side neutral to the secondary neutral at the transformer. From the electrician's perspective, ground and neutral are only the same as far as the service entrance, but since they are connected, the utility distribution neutral ends up being connected to the building service grounding electrode. That wouldn't be so bad, except in rural areas the quarter mile grounding practice ends up allowing the neutral current to run through the soil creating stray current and for the neutral conductor to ride a few volts above true ground by the time it gets to a residence. Not usually a problem in more populated areas where there are more grounds.

So the current producing stray voltage is often neutral current from the utility distribution network. (Phone systems, bad well pumps and many other things can also cause it) The second piece of the puzzle is that most electricians don't understand pool bonding, so they miss parts of it, and if it isn't complete it won't work.

From my perspective as an engineer, the fact that the distribution neutral and customer neutral aren't isolated seems crazy.

 

[C96]

does the motor frame and the equipotential zone around the pool including the bond to the water, tie back to the fourth ground wire, EGC, in the control subpanel?

Believe it or not, the answer is no. The #8 or larger solid copper bonding conductor used to reduce voltage gradients in the pool area “Equipotential Bonding” is not required to be attached to remote panelboards, service equipment, or electrodes.

 

[kd3pc]

JK is on target, binding vs grounding is key here. Pools are a unique beast when constructed and are treated as such in the NEC.

If the pool is liner based, with a metal perimeter and frame to support the walls and liner, that metal "should" have been bonded at installation, with all/any metal tied together. You also need that bonding to prevent galvanic action which will destroy the metal over the years.

 

[sberry]

As a note. A pool should have an insulated ground wire. There are several rules for other circuits and their proximity etc to reduce the potential for it to become energized from a difference source.

 

[Lightman]

At some locations I found the motor case on the pump was isolated (no continuity) from the equipment ground terminal at the line connection point and the outside caseof the pump, others were tied??? any uniform standard???

Believe it or not, the answer is no. The #8 or larger solid copper bonding conductor used to reduce voltage gradients in the pool area “Equipotential Bonding” is not required to be attached to remote panelboards, service equipment, or electrodes.

JKWilson
lack of rebar was addressed by the wording to place a bonding wire within 18-24 inches edge of pool under or in the surrounding deck? looked for specific wording to isolate bonded area around pool from EGC.

What is the normal pratice of bonding the water? our spec was to install a ground wire from neutral to **** plate at each pole + ground rods at each junction, transformer, arrestor, or fuse or recloser. only 1/4 mile rule was arrestor no more than 1/4 mile apart.

Location of the two pools side by side was a development small lots 100+ houses.
Other location we disconnected whole neighborhood, phase and neutral, suspected water heater or problem from other houses, still had problem.

Current terminlogy is earth to neutral voltage, no more stray voltage, too expensive to let it stray! LOL

Thanks all for comments.

 

[jkwilson]

Isolation isn't required. There was a note added to the 2008 NEC that said "Equipotential bonding is not required to extend to or be attached to any panelboard, service equipment, or grounding electrode." with the key word being "required". It's not prohibited, encouraged or discouraged, just not required. In Canada, it's just the opposite. The bonding system has to be grounded. My understanding is that the intentional grounding provides better protection against lightning strikes. The U.S. Code writing group chose to move toward better protection against unpredictable events like an excavator contacting a power line assuming people will get out of the pool when they see lightning.

Above ground pool pumps are usually double insulated and are specifically exempted from the bonding requirement in the NEC.

There are pipe fittings with stainless plates to bond the water. Heaters can be used to accomplish water bonding too.

2011 NEC tried to push for a copper grid under the pool shell and first 5ft? of deck, but it was backed off because of the lack of availability of such a product.