• Welcome to The Building Code Forum

    Your premier resource for building code knowledge.

    This forum remains free to the public thanks to the generous support of our Sawhorse Members and Corporate Sponsors. Their contributions help keep this community thriving and accessible.

    Want enhanced access to expert discussions and exclusive features? Learn more about the benefits here.

    Ready to upgrade? Log in and upgrade now.

Equipotential bonding for spa perimeter surfaces

Mr. Inspector

Mr. Inspector

SAWHORSE
Joined
Nov 28, 2009
Messages
4,677
Location
Poconos/eastern PA
2018 IRC requires a #8 copper wire around a spa that is not listed as self-contained in the unpaved or paved perimeter surface 18" to 24" from the inside wall of the spa.

1. If a slab with no steel in it on stone on top of the ground under a spa that is not listed as self-contained goes 20" past the inside walls of the spa and the rest of the perimeter surface is just dirt or grass should the #8 Copper wire be in the dirt or the slab?

2. What if the same spa is on the same slab but the slab only goes out 12" past the inside wall of the spa?

I'm asking because I don't think there would be any continuity between the dirt and the slab and maybe the wire should be in both the ground and the slab.
 
Other criteria will help to dictate, height of rim of hot tub greater or less than 28".
Potential difference in current between the earth and the water is the conscern
 
680.26(B)(2)(b) Copper Ring. Where structural reinforcing steel is not available or is encapsulated in a nonconductive compound, a copper conductor(s) shall be utilized where the following requirements are met:

(1) At least one minimum 8 AWG bare solid copper conductor shall be provided.

(2) The conductors shall follow the contour of the perimeter surface.

(3) Only listed splicing devices or exothermic welding shall be permitted.

(4) The required conductor shall be 450 mm to 600 mm (18 in. to 24 in.) from the inside walls of the pool.

(5) The required conductor shall be secured within or under the perimeter surface 100 mm to 150 mm (4 in. to 6 in.) below the subgrade.
 
I would make you bond both the concrete and the dirt in both cases listed - the entire 3' perimeter needs bonded.

As to the single copper wire around the pool, they got a TIA into the 2023 NEC that removes that single copper wire as a bonding option because it was deemed that it was not effective enough due to conditions that were encountered in the field. So, even though it may still be in the codes you have adopted, it isn't the safe way to go. It's TIA 70 - 23 – 9.

See video link for more detailed explanation on the TIA from Ryan Jackson, who was on the NEC code panel that voted on it, relevant video segment begins at about the 19 minutes and 10 second mark:

 
steveray said:
Does the NEC?
Click to expand...
Depends on the version of NEC 680.26 that the jurisdiction has adopted.

The OP is apparently in PA, and if up.codes is to be believed, PA uses the 2017 NEC. That version of 680.26(B)(2) for Perimeter Surfaces offers two ways to do the perimeter bonding: (a) Structural Reinforcing Steel or (b) which is just a single #8 copper wire connected with listed splices, installed within a given distance range from the inside walls of the pool and a given depth range below grade. There is no discussion of different surface types within the perimeter region; a single #8 copper installed in accordance with 680.26(B)(2)(b) suffices.

The TIA to the 2023 NEC changes 680.26(B)(2) quite a bit. It now differentiates between (a) conductive pavement (b) unpaved areas and (c) non-conductive areas (which do not require bonding if they are separated from earth). Subpart (d) specifies that the different areas must be bonded together.

So yes, under the 2023 NEC with TIA (if the jurisdiction has adopted both), or the future 2026 NEC, if the 3' permiter surface to be bonded has different types of surfaces, each must be bonded individually. And while a single #8 copper wire is no longer sufficient for conductive pavement, it is still allowed as the method for unpaved areas.

Since the OP was about a "spa that is not listed as self-contained" it is worth noting that 680.42(B) waives perimeter bonding requirements for certain self-contained spas on or above grade.

Cheers, Wayne
 
680.26(B)(2) Perimeter Surfaces. The perimeter surface to be bonded shall be considered to extend for 1 m (3 ft) horizontally beyond the inside walls of the pool and shall include unpaved surfaces and other types of paving. Perimeter surfaces separated from the pool by a permanent wall or building 1.5 m (5 ft) in height or more shall require equipotential bonding only on the pool side of the permanent wall or building. Bonding to perimeter surfaces shall be provided as specified in 680.26(B)(2)(a), (B)(2)(b), or (B)(2)(c) and shall be attached to the pool reinforcing steel or copper conductor grid at a minimum of four points uniformly spaced around the perimeter of the pool. For nonconductive pool shells, bonding at four points shall not be required.
steveray said:
Does the NEC?
Click to expand...
Mr. Inspector said:
1. If a slab with no steel in it on stone on top of the ground under a spa that is not listed as self-contained goes 20" past the inside walls of the spa and the rest of the perimeter surface is just dirt or grass should the #8 Copper wire be in the dirt or the slab?

2. What if the same spa is on the same slab but the slab only goes out 12" past the inside wall of the spa?
Click to expand...
The bonded perimeter must extend 3' from the inside walls of the pool.

If the wire is in the dirt and not in the concrete, the dirt is bonded and the concrete is not. If the wire is in the concrete and not in the dirt, the concrete is bonded and dirt is not. Everything in that 3' must be bonded per 680.26(B)(2). Therefore, you have to bond both.

If you want to consider it to be a grey area by getting into the weeds about how bonding the concrete may kinda bond the dirt around that concrete (or vice versa) and how bonded you actually need to be to meet code, I would say the fact that there is a TIA concerning #8 wire's effectiveness even without a break between the dirt and the concrete makes that a tenuous argument for an inspector to be making, considering that they are supposed to be protecting the safety and welfare.
 
BN4537 said:
If the wire is in the dirt and not in the concrete, the dirt is bonded and the concrete is not. If the wire is in the concrete and not in the dirt, the concrete is bonded and dirt is not. Everything in that 3' must be bonded per 680.26(B)(2). Therefore, you have to bond both.
Click to expand...
Prior to the 2023 NEC TIA, the wording of 680.26(B)(2) does not support that conclusion. There is no discussion of different surface types and bonding each separately. 680.26(B)(2)(b) requires only a single #8 within the prescribed region.

In fact, if you have 36" wide x 4" thick of plain concrete around the swimming pool walls, the older wording of 680.26(B)(2)(b) would permit the the single copper conductor to be within the soil 4" to 6" below the plain concrete, rather than within the plain concrete: 680.26(B)(2)(b)(5) "The required conductor shall be secured within or under the perimeter surface 100 mm to 150 mm (4 in. to 6 in.) below the subgrade."

Cheers, Wayne
 
Doesn't the dirt bond itself through the concrete (or vice versa) like a CEE? And the 2023-2026 is just selling more wire as that is what the NEC is about these days?

On the upside, I think I heard that they were going to move away from this for on/ aboveground pools with a definition change or being considered storable...
 
Last edited:
Mr. Inspector said:
looks like we have different opinions here.


I could see that this would normally work because the concrete is buried in the dirt, but not sure if the concrete slab just sits on top of the dirt.
Click to expand...
The order of magnitude is entirely different. A Ufer is dealing with very, very high voltage induced by lightning that can overcome a lot of resistance. The pool perimeter is trying to limit the voltage gradient so that if the pool gets energized at 120v while you are in it, you don't get killed when you get out and touch 0v grade. So the same amount of resistance that lightning induced voltage can overcome could kill any usefulness of having a perimeter bond at all.

So the concrete with the wire in it is touching the dirt. How much resistance is between them? Is the bond wire doing anything at that point? If you have them bonded just enough that you have a 110v difference of potential instead of 120v... you would still be dead. You get the difference of potential down to 90v... you are still likely to get killed. The same amount of resistance would be totally irrelevant in a Ufer application because the voltage would be large enough to overcome that resistance.

I'm glad they will exclude storables though.
 
Last edited:
BN4537 said:
trying to limit the voltage gradient so that if the pool gets energized at 120v while you are in it, you don't get killed when you get out and touch 0v grade.
Click to expand...
I thought this was the other way around for stray voltage for the utility using earth as a return because they can and we won't stop that...?
 
steveray said:
I thought this was the other way around for stray voltage for the utility using earth as a return because they can and we won't stop that...?
Click to expand...
I hadn't considered that. The Mike Holt videos I watched talked about a fault in the pool pump or a light fixture energizing the pool.
 
BN4537 said:
I hadn't considered that. The Mike Holt videos I watched talked about a fault in the pool pump or a light fixture energizing the pool.
Click to expand...
Isn't the pool water already bonded or protected in other ways? I know when I dropped my saw in my pool it tripped the GFCI instantly.....But what do I know, I am not an electrician and I never stayed at a holiday inn...

1743019677666.png
 
OK, let me add a little twist to this based on direct application in my area.

Many pools have a concrete deck around them, but below the coping. A bed of sand is then placed on top of the concrete, which is then topped off with pavers that are flush to the coping. This concrete is typically around 3-1/2" thick, give or take. Sometimes, there is no rebar used at all. The layer of sand can be a 1/2' to 2" thick, then the pavers or travertine.

How do you comply with that? If you are at the bottom of the concrete, you are then way below the 6" max if using the most popular #8awg wire loop. If you put it on top of the concrete, in the sand, you are eliminating the concrete and only addressing the sand, and hopefully the pavers.

What are your thoughts on this?
 
jar546 said:
OK, let me add a little twist to this based on direct application in my area.

Many pools have a concrete deck around them, but below the coping. A bed of sand is then placed on top of the concrete, which is then topped off with pavers that are flush to the coping. This concrete is typically around 3-1/2" thick, give or take. Sometimes, there is no rebar used at all. The layer of sand can be a 1/2' to 2" thick, then the pavers or travertine.

How do you comply with that? If you are at the bottom of the concrete, you are then way below the 6" max if using the most popular #8awg wire loop.
Click to expand...
Putting some rebar into the concrete (on a new build) would be a simple way to comply, even if it isn't not required for structural reasons.

Otherwise, 680.26(B)(2)(b)(5) says "within or under the perimeter surface 100 mm to 150 mm (4 in. to 6 in.) below the subgrade." In the buildup you describe, "within the perimeter surface" would mean within the pavers or travertine, which is not an option. : - ) The 4" to 6" below is in reference to the subgrade, i.e. the surface supporting the perimeter surface of travertine or pavers, not the grade, the top of the perimeter surface. So that would be 4" to 6" below the top of the sand layer. As long as the sand and concrete together are 4" to 6" thick, just below the concrete would comply.

At least, that's my best effort to make sense of the wording used. It does seem odd, though, that if you had say 6" thick plain concrete as the perimeter surface that the copper bonding wire should either be within the concrete, or 4" - 6" below it, while the uppermost 4" below the concrete is excluded.

The 2023 TIA changes subgrade to "finished grade", so that is more in accordance with what you were expecting. It also specifies 4" to 6" only for unpaved surfaces; for conductive paved surfaces, it is within 6" below the finished grade (and now has to be a grid, not just a single conductor).

Cheers, Wayne
 
Last edited:
ICE said:
What does the grid look like?
Click to expand...
The TIA says:

(1) Copper grid is constructed of 8 AWG solid bare copper and arranged in accordance with 680.26(B)(1)(b)(3).
(2) Steel welded wire reinforcement is minimum ASTM 6x6‐W2.0 x W2.0 or minimum No. 3 rebar constructed in a 300
mm (12 in.) grid.

The reference in (1) is to the section on copper grid for conductive pool shells, which says 12" x 12" grid with a tolerance of 4".

Cheers, Wayne
 
This is typical, and the thicknesses can vary. Pool pavers are typically 2-3/8", the sand can be anything, the concrete can be 3-1/2" or even 4"-5", so there are always lots of questions when they call in their perimeter surface bond inspection. There are plenty of scenarios that can play out, including Travertine, which is usually 1-1/4" thick.

If there is rebar in the concrete, we may have to have a separate wire in the sand if the concrete is too thick and they can't make the maximum 6" depth.

Note: this is for reference only and NTS.

9C8D4F78-162F-4839-B191-8FD9F87E0CE3.PNG
 
Is there a fabricated copper grid commercially available? It seems to me that the splices would require silver soldering.
 
You must log in or register to reply here.
Back
Top