Bonding of water piping

Status
Not open for further replies.

iggy2

Senior Member
Location
NEw England
I searched, but did not see anything related to this situation....

Multi-family building, 3 stories. Copper H/C water mains running across the basement, with copper water pipe taps which go vertical and serve back-to-back stacks of dwelling unit plumbing fixtures. Copper pipe mains in the basement are being replaced with CPVC. Copper risers (H/C) will remain, and be tapped into the new CPVC mains.

Inspector (correctly) cited a created violation of 250.104, since bonding of the metal water piping system is now lost.

We were called in to tell the plumbing contractor to hire an electrical contractor (a plumber cannot bond, IMO) to basically run a bond from one of the choices in 250.104 to each set of copper risers, hot and cold.

Plumbing engineer was a little shocked, since he has done this project repeatedly at other sites, but this issue did not come up. When he asked me about it, I looked it up, then said "yes, of course you have to re-bond the metal piping...", just to be a smart guy. State agency funding the project was also surprised. I'm not sure what, if anything, will happen at the other sites.

Inspector says a #6 is required, assuming a 20 ampere branch circuit is the largest circuit in the unit(s) that would energize the piping. I haven't looked into that logic yet (I don't know if maybe there's a 50A for a range?). My initial read is that we have to size based on 250.66, which says per the building SE conductors. If there's say and 800A service, the bond may have to be 2/0 (!!!) per table 250.66, since this does not fall under 250.66 (A), (B) or (C).

Thoughts?
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
If the wiring to the individual units is metered and OCPD protected at a lower amperage, you might be able to argue that that amperage rather than the service amperage should be used to size the bond. Especially since the full ampacity service conductors are extremely unlikely to be the ones which energize the pipes.
The required size of the bond to the pipes on the underground side, which might end up substituting for a broken neutral, should not be relevant to the interior pipe bonding jumper.
Disclaimer: That is just logic, not formal code interpretation.



Tapatalk!
 
Last edited:

xformer

Senior Member
Location
Dallas, Tx
Occupation
Master Electrician
Inspector says a #6 is required, assuming a 20 ampere branch circuit is the largest circuit in the unit(s) that would energize the piping. I haven't looked into that logic yet (I don't know if maybe there's a 50A for a range?). My initial read is that we have to size based on 250.66, which says per the building SE conductors. If there's say and 800A service, the bond may have to be 2/0 (!!!) per table 250.66, since this does not fall under 250.66 (A), (B) or (C).

Thoughts?

IMHO... Look at 250.104(A)(2) :)
 

iggy2

Senior Member
Location
NEw England
Yes, that logic seems to be allowed per 250.104(A)(2), but ONLY if the plumbing of the dwellings units are isolated from one another - which is not the case here.

That's why I end up back at 250.66. But if the AHJ is OK with a #6, I guess I'll go with that. A 2/0 seems pretty silly.
 

xformer

Senior Member
Location
Dallas, Tx
Occupation
Master Electrician
Yes, that logic seems to be allowed per 250.104(A)(2), but ONLY if the plumbing of the dwellings units are isolated from one another - which is not the case here.

That's why I end up back at 250.66. But if the AHJ is OK with a #6, I guess I'll go with that. A 2/0 seems pretty silly.

Good Catch iggy2... I would agree with you and go back to 250.66 :)
 

roger

Moderator
Staff member
Location
Fl
Occupation
Retired Electrician
Had an inspector red tag this installation.

plasticbonding.JPG


He said it had to be 30 lb monofilament. :D

Roger
 

iggy2

Senior Member
Location
NEw England
I did the math, and the cross sectional area of a 2" copper pipe is about .5 square inches, which is about 676 kcmil. I guess 2/0 doesn't sound so silly anymore.

It doesn't seem like a 3/4"x3/4" copper bar, and a piece of 676 kcmil have the same area...but the numbers don't lie.
 

augie47

Moderator
Staff member
Location
Tennessee
Occupation
State Electrical Inspector (Retired)
It would seem to me that you have two choices.
If, as I believe you say, the water pipe to each unit is not isolated then a single electrode conductor based on 250.66 for the service size would be compliant.
If the piping sections to each unit are isolated the 250.104(A)(2) would allow you to connect to each isolated riser with a jumper base on 250.122 for the feeder to that unit.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
The apparent problem is that the OPs situation is in between those two. There are multiple isolated risers but each riser may serve four or more units.
So bonding each group of risers together and then running one long bond wire from each group to the selected bonding point seems to be what is required.

Tapatalk!
 

augie47

Moderator
Staff member
Location
Tennessee
Occupation
State Electrical Inspector (Retired)
The apparent problem is that the OPs situation is in between those two. There are multiple isolated risers but each riser may serve four or more units.
So bonding each group of risers together and then running one long bond wire from each group to the selected bonding point seems to be what is required.

Tapatalk!


Good point...
If that is the case, I would agree with your "solution" and, in my opinion, that bond would need to be sized by 250.66 since we don't have the "individual occupancy" option.
 
Status
Not open for further replies.
Top