Pool bonding question

[sidewinder770]

Hi all. I have been to this forum several times and found most of what I needed to know by reading other threads but now I need a little help and can't seem to find the answer elsewhere. I don't do much work on pools at all so when it's a little out of the norm I figure it's best to ask for help from the pool experts.

Here's what I got- I'm helping a friend install a subpanel for a pool equipment relocation project. That's pretty straight forward and were actually using GFi breakers on all circuits even ones going to some fixed horse arena lights because he wants to be over safe.

The problem we're running into is the pool bonding. The pool light is in a red bronze conduit to a deck box (already existing) and will be moved a little but no extending of the conduit/pool light will be needed. The deck box will end up being about 15 feet away from the other equipment (sub panel, pool pump, blower motor, etc). He has dug around all the existing plumbing and pool light pipe and is ready to relocte them when I found out he has no bonding wire from the pool grid (inground pool about 20 years old). I thought that was kind of odd so I'm not sure what to do about the light and the bonding now as I was planning to just extend the bond wire to the equipment area but now that I see there isn't one what do I do? We will be extending the pool light from the deck box to the new panel using PVC which from what I know is acceptable but how do I bond the equipment without a bonding wire from the pool? we didn't want to have to extend the bronze pipe to the equipment area because of cost and also because he would have to buy a new pool light since the old cord would be to short. Can I attach a #8 bond wire to the bronze pipe for the pool light (at the deck box) and bring that over to the equipment area? I'm thinking it might be a good idea to put a ground rod in by the new equipment as well to bond to just to be safe but I want to make sure all this is up to code. Just kind of weird that there is not a bond wire from the pool but maybe back then they just used the bronze pipe as the bond?

any help would b great and thanks!

 

[barbeer]

I would at minimum take a bond wire from brass? to equip. area, ground rod is not necessary. You are bonding all these together to remove potential differences not to "ground" to earth.

 

[Dennis Alwon]

First off I am not sure you need to bond the pool at this point. I am not sure when equipotential bonding showed up in the code but if this is an existing install I am not sure whether you can adequately bond the pool without tearing up the whole area. Art. 680.26 is what you should read.

Now, I think you are talking about the pool light bonding.... In that case a number 8 copper conductor from the light shell to the deck box is what you need. You then just wire the egc to the deck box.

 

[sidewinder770]

Thanks for the reply. you say "at a minimum"- what else could be done along with the #8 bonding wire from the brass pipe located at the deck box to the eqipment area? I wasn't sure about the ground rod but not having a bonding wire kind of through me for a loop- must have allowed the use of the bronze pipe as a bond back then or something.

 

[dnem]

The problem we're running into is the pool bonding.

Pool bonding is a problem for 2 main reasons
1] the NEC still isn't clear about many issues
2] it's common place to confuse which conductor is at question

The jobs of bonding [and grounding] wires are separated into 3 main categories
1] electrode
2] equipotential
3] equipment

1] electrode, 250.32 says, "Building(s) or structure(s) supplied by feeder(s) or branch circuit(s) shall have a grounding electrode ..... installed .....". . Is the pool a "structure" ? . Yes. . Is it supplied by more than "a single branch circuit" or a shared neutral "multiwire branch circuit" ? . Usually. . Should this rule apply to a pedestal mounted pool feeder panel ? . My answer is yes. . But don't confuse a electrode requirement with a equipotential requirement.

Can I attach a #8 bond wire to the bronze pipe for the pool light (at the deck box) and bring that over to the equipment area? I'm thinking it might be a good idea to put a ground rod in by the new equipment as well to bond to just to be safe but I want to make sure all this is up to code. Just kind of weird that there is not a bond wire from the pool but maybe back then they just used the bronze pipe as the bond?

You're mixing electrode, equipment, and equipotential into one big confusing ball. . A ground rod is an electrode. . It's conductor goes to the panel enclosure. [250.32(B) connected to the ..... disconnecting means]

2] equipotential, When you say "bond wire from the pool" you're probably talking about equipotential, which bonds to the motor housing. . This would also bond to the lug on the outside of the wet niche fixture and also to the pool metal and also to the pool deck/sidewalk structure. [680.26(B)(1)-(7)]

3] equipment, Your equipment ground [grounding conductor] is required to be #12 or larger insulated copper run with the other circuit/feeder conductors. [250.134(B), 300.3(B)]. . The equipment ground for a nonmetalic conduit [or if metal for only part of the way from "forming shell to a junction box or other enclosure", 680.23(B)(2)] is required to be #8 and insulated [680.23(B)(2)(b)]. . The equipment ground for the light has limits on joints and splices [680.23(F)(2)]. . The bond lug on the inside of the wet niche fixture is for equipment ground and must be "encapsulated in, a suitable potting compound". [680.23(B)(4)]

Now
If you get all of your categories straight, do you still have a question or is it cleared up ?

 

[augie47]

3] equipment, Your equipment ground [grounding conductor] is required to be #12 or larger insulated copper run with the other circuit/feeder conductors. [250.134(B), 300.3(B)]. . The equipment ground for a nonmetalic conduit [or if metal for only part of the way from "forming shell to a junction box or other enclosure", 680.23(B)(2)] is required to be #8 and insulated [680.23(B)(2)(b)]. . The equipment ground for the light has limits on joints and splices [680.23(F)(2)]. . The bond lug on the inside of the wet niche fixture is for equipment ground and must be "encapsulated in, a suitable potting compound". [680.23(B)(4)]

quote]
David, Am I not correct that if his conduit from the shell to the pool j box is brass, the #8 from inside the niche is not needed? (680.23{B}(2)(a) vs (2)(b).

I would think he would still need to add the external equii-potential bonding wire to the grid.

 

[sidewinder770]

Thanks for the reply and that does make sense. I think we are on the same page and probably got the "common place to confuse which conductor is in question" problem

1- Electrode- Yes I was planning to hook the ground rod to the subpanel and not to the motor or any other equipment- just a ground location for the subpanel.

2- equipotential- This is where I am not sure as I don't have a wire (bare or insulated) coming from the pool light niche or anywhere else to the deck box (either inside the bronze pipe or outside). Is the bronze pipe from the light to the deck box enough for this part of the bonding equation? Also- I know that the equipment (motor, etc) needs to be bonded (by way of the external bonding lug) to the pool structure but since there is no bonding wire coming from the pool will attaching a clamp to the bronze pipe at the deck box and runnng a #8 solid uninsulated wire to the equipment (not in a conduit) going to satisfy this requirement?

3- equipment- for this there will be a #12 insulated circuit run from the panel (GFI breaker) to a light switch (with seperate ground wire for the switch) to the deck box in PVC, All wires #12 insulated and not attaching the ground wire to the deck box).

does all that sound right or is there something else I should be doing. Not having any bonding wire from the pool is what has me guessing.

 

[sidewinder770]

David, Am I not correct that if his conduit from the shell to the pool j box is brass, the #8 from inside the niche is not needed? (680.23{B}(2)(a) vs (2)(b).

I would think he would still need to add the external equii-potential bonding wire to the grid.

That's what I'm thinking as well- if it was pvc from the niche it would need a #8 inside but being bronze it doesn't. The problem lies in how do I connect the equipment to the grid if there is currently no bonding wire coming from the pool- will the bronze pipe work for this?

 

[benaround]

3- equipment- for this there will be a #12 insulated circuit run from the panel (GFI breaker) to a light switch (with seperate ground wire for the switch) to the deck box in PVC, All wires #12 insulated and not attaching the ground wire to the deck box).

does all that sound right or is there something else I should be doing. Not having any bonding wire from the pool is what has me guessing.

Why would you not connect the EGC to the deck box?

 

[M. D.]

I think you are required to do so .....

http://ecmweb.com/mag/electric_grounding_vs_bonding_11/

 

[dnem]

3] equipment, Your equipment ground [grounding conductor] is required to be #12 or larger insulated copper run with the other circuit/feeder conductors. [250.134(B), 300.3(B)]. . The equipment ground for a nonmetalic conduit [or if metal for only part of the way from "forming shell to a junction box or other enclosure", 680.23(B)(2)] is required to be #8 and insulated [680.23(B)(2)(b)]. . The equipment ground for the light has limits on joints and splices [680.23(F)(2)]. . The bond lug on the inside of the wet niche fixture is for equipment ground and must be "encapsulated in, a suitable potting compound". [680.23(B)(4)]

David, Am I not correct that if his conduit from the shell to the pool j box is brass, the #8 from inside the niche is not needed? (680.23{B}(2)(a) vs (2)(b).

Yes, that's right. . The #8 [or larger] is not required to extend to the equipment. . Light niche to jbox is acceptable [680.23(B)(2)(b)].

The OP wanted to know if that wire could continue from the jbox to the equipment [motor etc] and the answer to that is, yes.

"We will be extending the pool light from the deck box to the new panel using PVC which from what I know is acceptable but how do I bond the equipment without a bonding wire from the pool? we didn't want to have to extend the bronze pipe to the equipment area because of cost and also because he would have to buy a new pool light since the old cord would be to short. Can I attach a #8 bond wire to the bronze pipe for the pool light (at the deck box) and bring that over to the equipment area?"

I would think he would still need to add the external equii-potential bonding wire to the grid.

Absolutely !
An equipment ground is NOT an equipotential grid. . You can read about the grid and everything that it needs to bond to in 680.26(B)(1)-(7).

 

[dnem]

Thanks for the reply and that does make sense. I think we are on the same page and probably got the "common place to confuse which conductor is in question" problem

1- Electrode- Yes I was planning to hook the ground rod to the subpanel and not to the motor or any other equipment- just a ground location for the subpanel.

2- equipotential- This is where I am not sure as I don't have a wire (bare or insulated) coming from the pool light niche or anywhere else to the deck box (either inside the bronze pipe or outside). Is the bronze pipe from the light to the deck box enough for this part of the bonding equation? Also- I know that the equipment (motor, etc) needs to be bonded (by way of the external bonding lug) to the pool structure but since there is no bonding wire coming from the pool will attaching a clamp to the bronze pipe at the deck box and runnng a #8 solid uninsulated wire to the equipment (not in a conduit) going to satisfy this requirement?

3- equipment- for this there will be a #12 insulated circuit run from the panel (GFI breaker) to a light switch (with seperate ground wire for the switch) to the deck box in PVC, All wires #12 insulated and not attaching the ground wire to the deck box).

does all that sound right or is there something else I should be doing. Not having any bonding wire from the pool is what has me guessing.

From what you've posted so far, I don't think you have any grid.
If you walk away from this pool without any equipotential, I believe you will take some liability with you even tho it was an existing pool. . Even if you're found not liable in court, you'll have to pay a lawyer to defend your scope of work.

At the very least, you should ohm meter from the motor housing to anything metal in or around the pool [ladder railings, cover enclosure]. . Use a piece of wire attached to the motor lug and test wire to railing [measure wire first to subtract it's ohmic value]. . Measure railing to railing, etc. . If you get anything over a couple of ohms, there's something wrong ! . Tell the customer they have a serious problem and they need to expand your scope of work to fix it. . If they don't want to listen, will need to give them a letter documenting your attempt to inform them of the seriousness of the danger. . But don't hand them the letter, even if you're seeing them in person. . Send the letter registered mail. . In the future that letter could be as good as gold.

There's a couple of things that's adding to the confusion of the whole pool bonding situation and I see it here also.

A] Problem: #8 wire
The equipotential bonding wires for a pool are speced at min #8 [680.26(B)]. . The equipment grounding wires for a pool are usually speced at min #12 [680.7(B), 680.21(A)(1)]. . When electricians look at 680.23, they see a #8 [680.23(B)(2)(b)], but this is not an equipotential bonding wire, it's an equipment grounding wire. . It does nothing to satisfy the equipotential requirements of 680.26.
Answer:
Don't look at min wire size to figure out what's what. . 680.26 is equipotential. . The stuff in 680 before .26 is all equipment grounding. . Also notice that 680.23(B)(2)(b) is an insulated conductor, a dead give away that it's equipment grounding.

B] Problem: Niche shell redundancy
The wet niche shell has 3 different grounding conductor connections involved from 2 different grounding categories. . Starting from the outside, [#1] the outside shell lug always gets connected to an equipotential bonding wire. . [#2] The inside shell lug gets connected to an equipment grounding wire but is only required if the conduit is not metal [don't forget the potting]. . And, to add to the confusion, this is the one time that the equipment grounding is speced at #8. . [#3] The wiring to the lamp/bulb holder itself is a multiwire cord that contains yet another equipment grounding wire, this one a #12.
Answer:
Slow down when you're dealing with a 120v wet niche light. . The shock and drowning hazard of a screwed up light install is far far greater than that of the motor. . Make sure you've considered all 3 grounds and accounted for each.

3] The NEC itself adds confusion. . Electrode and equipotential are supposed to be kept separate so that the requirements for each are clear. . But follow this trail. . A ground rod is clearly an electrode covered by 250.52+53. . A service ground mat can also be used as an electrode and installed according to 250.52+53. . The electrode bonds the building and the equipment to the earth and connects to the service enclosure [or 2nd building feeder main disconnect]. . An equipotential grid bonds the area to anything metal within the area and connects to the equipment housing itself [pool motor etc]. . It's primary purpose is to protect people. . But when you jump over to 682.33(A) you find a grid connected to the service enclosure and they call it a equipotential grid.
Answer:
Since the 682 grid not only bonds to and grounds the service but also provides equipotential for the person standing in front of the grid, it should be considered both an electrode and an equipotential. . Consider 682 to be a unique situation and don't let it confuse you when you're in an installation under 680. . 680 has separate requirements for equipment grounding and equipotential bonding plus you can't ignore the electrode requirements back in 250 for any subpanel involved.

 

[dnem]

At the very least, you should ohm meter from the motor housing to anything metal in or around the pool [ladder railings, cover enclosure]. . Use a piece of wire attached to the motor lug and test wire to railing [measure wire first to subtract it's ohmic value]. . Measure railing to railing, etc. . If you get anything over a couple of ohms, there's something wrong ! . Tell the customer they have a serious problem and they need to expand your scope of work to fix it. . If they don't want to listen, will need to give them a letter documenting your attempt to inform them of the seriousness of the danger. . But don't hand them the letter, even if you're seeing them in person. . Send the letter registered mail. . In the future that letter could be as good as gold.

If the ohmic value is high, the danger is high. . But even if the value is low, the absence of an equipotential wire at the motor housing lug is a concern. . Without a conductor, resistance could vary depending on moisture content of the soil. . Someone could get shocked during the dry season if the ground is solely used as part of the path. . Even if the measurement value is low, I would still write a letter to the owner documenting the missing bond and clearly state that the pool and any concrete was existing. . Send it certified.

 

[dnem]

A] Problem: #8 wire

B] Problem: Niche shell redundancy

3] The NEC itself adds confusion

Plus I add confusion for labeling A, B, + 3
Should be A, B, + C but I'm not being allowed to edit.