Thread: Water Heater BX Quiz

When I come across an installation such as pic 1 in an older, pre-80's house, I should:

A. call for a repair, an anti-short bushing and a clamp to the wall.
B. call for a repair, an anti-short bushing and a staple past the end of the BX.
C. suggest a wrap with tape, tape the BX to the NM.
D. ignore it, the AHJ approved this back then. I see this so often, it probably was allowed.
E. shove the end of the BX into the hole in the drywall and move on.
F. call for a repair, a junction box and a cable clamp.

The other pics are for reference or comments. The one behind the panel is in a mobile home.

Similar Threads:

• ABS Flue Pipe on LP Water Heater
• Tankless water heater comment
• Water Heater Location
• Water heater Question
• Reverse plumbing to water heater

The electrical feed to the water heater has been damaged. Have a licensed electrical contractor removed the existing and install a new feed and disconnect for safety. While he's at the WH, have him add the proper bonding cable across the hot & cold water pipes as required.

JK: I would choose "F".

Originally Posted by Darren Miller

The electrical feed to the water heater has been damaged. Have a licensed electrical contractor removed the existing and install a new feed and disconnect for safety. While he's at the WH, have him add the proper bonding cable across the hot & cold water pipes as required.

what code section requires you jump across the hot and cold water lines at the water heater ? The one picture has plastic piping in the structure and 2 of them you can not tell

Originally Posted by ken horak

what code section requires you jump across the hot and cold water lines at the water heater ? The one picture has plastic piping in the structure and 2 of them you can not tell

Don't know what you're talking about KH, I see copper, brass valve & steel nipples in the first picture which was the subject post for the cold inlets & hot outlets that should still be bonded . Look beyond the TPRV discharge in the first pic. Cannot tell on the fourth pic neither cold inlet or hot is shown, again TPRV discharge only shown usually mobile/manufactured homes have inlets/outlets on the side and TPRV discharge down through floor is the norm. I've rarely seen electric storage water heaters in mobile/manufactured homes - as the original panels are/were usually limited 60 or 100 amps on a SW and the WHs were usually gas with convertable propane/nat'l gas kits, and required to be bracket mounted. They are usually in a closet with door on the exterior, separated from the rest of the manufactured home, note no PAN in the fourth photo - a requirement here for a manuf'd home WH storage tank.

I too choose F, excepting the additional disconnect if there are two intermediate means in place (a main & a circuit disconnect at the panel) and either the panel is within view & control OR the breaker can be locked out for servicing. The one with the wall plate (not the first pic) may have a junction box in the wall. Bonding is required.

I note that one of the WH's (3rd pic with all the other stuff piled/leaning against the WH where it should remain clear ) is strapped too close to the top to be effective and does not have a flex connection does your area have a seismic zone requirement JK? Looks like a pointed screw on the removed panel in the fourth picture - further risk of hazard with unprotected NM.

Just curious -- in picture 2 (which looks like it's missing a discharge pipe), what are the white, Saturn-looking balls attached to the hot and cold?

Originally Posted by Jon Errickson

Just curious -- in picture 2 (which looks like it's missing a discharge pipe), what are the white, Saturn-looking balls attached to the hot and cold?

Water hammer devices. They do not have little green men inside.

Well then, who does the hammering??

Originally Posted by H.G. Watson, Sr.

the cold inlets & hot outlets that should still be bonded .

Like I asked in my original post - What code article ?

I don't believe it is required by any code article. But I would think that somewhere in 250.104 would be the closest to doing that. In my experience 99% of the time I saw the metal water pipes (hot and cold) bonded at the the hot water tank it wasn't necessary. Ken ... I'm sure you know the story to why this is becoming more commonly seen in homes so I won't go into that. In my experience any home that was plumbed with copper or metal pipes had the required bonding at the mixing valves for tubs and showers. I suppose you could add the additional bond at the hot water tank if it gave you little fuzzies or something.

So my answer is the code only requires it if you can prove isolation of cold and hot water pipes.

I think and I may be corrected that this bonding jumper between hot and cold pipes might be required in mobile homes where copper was used for the piping and plastic for the tub and shower valves and flexible or plastic supplies from shut off valves to fixtures .In this case proof of isolation might be possible but again I am not certain.

If it is likley that a enginerized metal piping systems can occure it needs bonding at removable fixtures, If connectin is questioned code requires bonding jumpers. (PRV, meters, H2O heaters dielectric fittings, ductwork, etc.)

250.104 Bonding of Piping Systems and Exposed Structural Steel.
(A) Metal Water Piping. The metal water piping system shall be bonded as required in (1), (2), (3), or (4) of this section. The bonding jumper(s) shall be installed in accordance with 250.64(A), (B), and (E). The points of attachment of the bonding jumper(s) shall be accessible.
(1) General. Metal water piping system(s) installed in or attached to a building or structure shall be bonded to the service equipment enclosure, the grounded conductor at the service, the grounding electrode conductor where of sufficient size, or to the one or more grounding electrodes used. The bonding jumper(s) shall be sized in accordance with Table 250.66 except as permitted in 250.104(A)(2) and (A)(3).

Those little plastic disks in the steel nipples pictured (heat traps/functioning as dielectrics) isolate, they must be jumpered. The CNC is clear on the subject (JK is in BC). Any non-metalic or die electric supply to a faucet also isolates. IIRC the Canadian codes also require a jumper to the washing machine.

Electric water heater, if an element fails and starts electrifying the water - ouch. We've already got open BX under a TPRV, gosh I hate seeing water over un protected electricity.

For the US, NEC already cited by others. Check the FPN and commentary for clarification.

Finally, remember the original post - regarding picture one and that it is a PRE 1980s home. If in the States that would mean floating grounds to metal water pipes were still allowed - and it didn't just have to be a cold water pipe.

It takes just one non-conductive supply to a faucet or a dieelectric, plastic, bronze stop cock, etc. and you've isolated. Its the other way around regarding bonding - prove the negative that there isn't any isolatation - with a storage type water heater and copper you won't be able to unless it is to the detriment (sacrifice) of the vessle when the anode is quickly consumed.

That particular home has all copper plumbing. Hot is bonded to cold at the tub faucet. Cold is bonded correctly at the main shutoff valve. No danger.
Plastic (polyB) in the mobile, except for the stub outs. It was strapped, missing a pan, no BX needed AFAIK.

Can the length of flexible conduit be terminated in the drywall?

Although 250.104 was mentioned, I realize SvH quoted 250.104(A), not what is on point which is 250.104(B):

2008 NEC:

250.104 Bonding of Piping Systems and Exposed Structural Steel.

(B) Other Metal Piping. Where installed in or attached to a building or structure, metal piping system(s), including gas piping, that is likely to become energized shall be bonded to the service equipment enclosure, the grounded conductor at the service, the grounding electrode conductor where of sufficient size, or to the one or more grounding electrodes used. The bonding jumper(s) shall be sized in accordance with 250.122, using the rating of the circuit that may energize the piping system(s). The equipment grounding conductor for the circuit that is likely to energize the piping shall be permitted to serve as the bonding means. The points of attachment of the bonding jumper(s) shall be accessible.

FPN: Bonding all piping and metal air ducts within the premises will provide additional safety.

Great that the bonding is at the tub faucet, but it is broken between the GEC bond and the cold at the WH dieelectric/heat trap as the hot is also isolated, thus the bonding is required. CEC expressley requires it (bonding at the electric WH) as well IIRC.

Properly terminated at a JB in the drywall with a cover, perhaps.

Thanks, HG, I think I get it. The pipe is bonded at one end but isolated at the other.
It needs suspenders in case the belt breaks.

Originally Posted by A.D. Miller

JK: I would choose "F".

AD and the guys who choose "F", a splice in a J box is an improvement over a continuous feed from the panel, right?
Or do you suggest pulling the continuous feeeder thru the clamps and the box and the cover? But the J box needs bonding.

They have stripped the NMSC and pushed it thru the bx loom. I would suggest a box new bx , and reterminate in the box properly. Put the box on the wall with the NMSC coming thru the back with a connector fasten the box to the wall and then run a new piece of bx to the heater with the proper connectors and terminations. This will comply and be safe.........

Of the choices, "F". Although, recommending repair should be enough - electricians prefer to determine their own methods of correcting a problem (if they agree that it is a problem, that is), in my experience.

This is what I see ...



These are heat traps with dielectric inserts but they do not electrically isolate. But this below will electrically isolate the connection to the metal tank of the copper pipe.... a dielectric union.



How can it be that any one dielectric connection to a fixture is isolation electrically for the entire metal water piping system? How does bonding at the hot water tank fix that as far as preventing difference of voltage potential between hot and cold pipes ? Diagram that for me please cause I'm not seeing it. If that was the case then it would be more than an FPN that suggests bonding for added safety.. the circuit likely to energize the hot water tank is its supply circuit and therefore bonding is met with the egc of that circuit. Same with any other electric powered water usage unit. How many bonds do you want? Other than that there are no other circuits that should be likely to energize metal water pipes as addressed in 250.104(B). Or are you saying that a bonding jumper at the hot water tank between hot and cold water pipes is required by 250.104(B)? Those steel nipples (heat traps) are not isolating the copper hot and cold piping in picture 1.

If there were dielectric unions of the type that insulates the connection from electrical continuity as pictured above then I would say due to the fact you could show a mechnical connection is preventing bonding of hot and cold it would be required to bond the pipes in accordance with 250.104(B)

Take away any electrical interface with a metal water pipe system and dielectric unions not used at the connection of the hot water tank then ' bonding' of hot and cold pipes would only be subject to complying with the fpn for added safety.


Now how I've been taught ... If the metal water piping system in a single family dwelling is a complete system with no nonmetallic sections then it is a 250.104(A)(1) installation otherwise it is a 250.104(B) installation and the egc of the circuit likely to energize the pipes mets the bonding requirement.

This has been upheld through several code proposals to require bonding.... period. Required bonding of hot and cold metal water pipes has been refused (IMO) by the ROP documents on several ocassions ... as stated it is either a 250.104(A) installation or its a 250.104(B) installation. If the hot water tank has dielectric unions that break electrical continuity then it is no longer a complete system as described in 250.104(B) and then if it is likely to be energized the egc in that circuit meets the bonding requirement for the pipes.

Originally Posted by Roger Frazee

This is what I see ...



These are heat traps with dielectric inserts but they do not electrically isolate. But this below will electrically isolate the connection to the metal tank of the copper pipe.... a dielectric union.



How can it be that any one dielectric connection to a fixture is isolation electrically for the entire metal water piping system? How does bonding at the hot water tank fix that as far as preventing difference of voltage potential between hot and cold pipes ? Diagram that for me please cause I'm not seeing it. If that was the case then it would be more than an FPN that suggests bonding for added safety.. the circuit likely to energize the hot water tank is its supply circuit and therefore bonding is met with the egc of that circuit. Same with any other electric powered water usage unit. How many bonds do you want? Other than that there are no other circuits that should be likely to energize metal water pipes as addressed in 250.104(B). Or are you saying that a bonding jumper at the hot water tank between hot and cold water pipes is required by 250.104(B)? Those steel nipples (heat traps) are not isolating the copper hot and cold piping in picture 1.

If there were dielectric unions of the type that insulates the connection from electrical continuity as pictured above then I would say due to the fact you could show a mechnical connection is preventing bonding of hot and cold it would be required to bond the pipes in accordance with 250.104(B)

Take away any electrical interface with a metal water pipe system and dielectric unions not used at the connection of the hot water tank then ' bonding' of hot and cold pipes would only be subject to complying with the fpn for added safety.


Now how I've been taught ... If the metal water piping system in a single family dwelling is a complete system with no nonmetallic sections then it is a 250.104(A)(1) installation otherwise it is a 250.104(B) installation and the egc of the circuit likely to energize the pipes mets the bonding requirement.

This has been upheld through several code proposals to require bonding.... period. Required bonding of hot and cold metal water pipes has been refused (IMO) by the ROP documents on several ocassions ... as stated it is either a 250.104(A) installation or its a 250.104(B) installation. If the hot water tank has dielectric unions that break electrical continuity then it is no longer a complete system as described in 250.104(B) and then if it is likely to be energized the egc in that circuit meets the bonding requirement for the pipes.

The plastic inserts by design impede/break continuity/free passage/exchange of electrons between the galvanized nipple and the copper electrons so that they do NOT pass directly between the two metals, and this is of course its intended purpose so as to arrest GALVANIC corrosion.

I believe you have missed the mark as to just what a dielectric insert or union by design accomplishes as far as inhibiting a transfer of electrons not only one direction as in DC voltage but both directions transfer in AC voltage.

The metal system has to be electrically continuous/bonded. The hot water piping system begins at the water heater, the outlet which in this case is isolated (both inlet and outlet of the WH) with those nipples (otherwise we'd have corrosion between the Steel and the copper - thus inhibiting a good bond anyway and further causing quick decomposition/destruction/failure of the WH tank sacrifical to that copper plumbing pipe. You need a complete bonding circuit path for the potable hot water system. Although storage tanks of such steel vessles are often "glass" lined (SS - non conductive; or copper tank excepted), rarely, if ever, is the coating perfect, either.

However, the CEC expressly requires the bonding here IIRC. IIRC so would NY, MA, IL and NJ.

Remember, in the instant case the storage type WH is an ELECTRIC one(element/s) in BC Canada. The instant case and accompanying examples are also regarding PRE-1980s Residential Electrical Systems according to the original post.

I agree with the others except in your area do you require Eath Quake straps??



Rolland Pruner

Livermore, Ca

I agree with the others except in your area do you require Earth Quake straps??



Rolland Pruner

Livermore, Ca

I apologize if I wasn't clear on the purpose of the dielectric union. I did not mean to imply it had any electrical use as far as bonding or not bonding the pipes electrically.. Its purpose is as you stated. Plumbers install them not electricians....

Speaking in the USA

The continuity for the required bonding in a 250.104(A) installation is compromised with a dielectric union assuming that you are not going to consider metal mixing valves for tubs and showers upstream as bonding the hot and cold pipes as you were stating. In this case the dielectric union would be like a short piece of pvc and the piping would fall under 250.104(B). The egc in the supply circuit would satifsy the bond to the pipes for that electric hot water tank.

I have never heard of nor ever been told or ever attended a class that told me that metal hot and cold water pipes were not bonded thru the mixing valve assemblies in the downstream metal plumbing. I have been directed to consider the fpn. To say that if a non metallic type connection is made from a shutoff to sink fixture dictates bonding of hot and cold pipes is hard for me to see. Thinking of it in that way would mean I would need to install a bonding jumper from the metal fixture to the metal pipe, ... bonding somewhere between the pipes downstream would not bond the fixture. In fact to bond the metal water pipe system under that mind set you would need multiple bonding jumpers.

Anyway I believe you are saying that it is required to bond the hot and cold metal water pipes in a single family dwelling in the USA or maybe I'm mistaken. If so I always welcome correcting my misunderstandings. I would welcome any one to expalin where I am in error so that I can not continue to believe what I have been instructed for years now. I do know this it isn't required by our local codes last I looked and I can probably count on one finger how many times I've seen this done on single family dwellings with metal water pipes in my area.

Having been retired for awhile now this is going to cause me to go over to a new constructuion home sub-division and see if this bond at the hot water tank is being done. That is if I can find a new construction home sub-divison in these times


I have no idea the requirements of Canadian code. I failed to check the location.... my mistake. I do not believe it is a code requirement to bond the hot and cold pipes in the USA.

John Kogel

I pick F.

As for the other comments, I do want to see a grounding bonding strap
across the cold to hot water pipe if copper tubing is used.

I perfer to see greenfield or bx not NM-B. But I don't make a big deal
out NM-B, it just look like a nicer looking job.

Lets refer to the below diagram. This is how I am assuming what Canadian code is mandating. This is for my understanding I'm not wanting to cause argument. That bonding is not met between hot and cold water pipes via upstream metal mixing valves or if any one fixture has a nonmetallic connection... The heat traps are acting like a dielectric ( picture 1 in this thread ) and the tank is not allowing bonding (shown by the breaks in the gold lines). What we would have is a hot water pipe effectively isolated considering that understanding... so the bond at the hot water tank between hot and cold is required by CEC discounting any like language (USA) for the egc to meet the bonding requirement.

Am I even close......

Roger nice schematic

It show the grounding conductor connected to the water tank, yes that
right.

I show a bonding jumping across hot & cold water pipes with the wire
leading off back to the panel grounding electrode, that how I read it.

Some one will correct me, but I was taught that you need only to just
connect a bonding jumper, (a wire #6 solid or stranded) and bond it
to both hot and cold water pipe. That it, nothing more.

The reason I was told, is to make sure the hot water pipe is grounded,
with the cold water pipe.

Originally Posted by Roger Frazee

I apologize if I wasn't clear on the purpose of the dielectric union. I did not mean to imply it had any electrical use as far as bonding or not bonding the pipes electrically.. Its purpose is as you stated. Plumbers install them not electricians....

Speaking in the USA

The continuity for the required bonding in a 250.104(A) installation is compromised with a dielectric union assuming that you are not going to consider metal mixing valves for tubs and showers upstream as bonding the hot and cold pipes as you were stating. In this case the dielectric union would be like a short piece of pvc and the piping would fall under 250.104(B). The egc in the supply circuit would satifsy the bond to the pipes for that electric hot water tank.

I have never heard of nor ever been told or ever attended a class that told me that metal hot and cold water pipes were not bonded thru the mixing valve assemblies in the downstream metal plumbing. I have been directed to consider the fpn. To say that if a non metallic type connection is made from a shutoff to sink fixture dictates bonding of hot and cold pipes is hard for me to see. Thinking of it in that way would mean I would need to install a bonding jumper from the metal fixture to the metal pipe, ... bonding somewhere between the pipes downstream would not bond the fixture. In fact to bond the metal water pipe system under that mind set you would need multiple bonding jumpers.

Anyway I believe you are saying that it is required to bond the hot and cold metal water pipes in a single family dwelling in the USA or maybe I'm mistaken. If so I always welcome correcting my misunderstandings. I would welcome any one to expalin where I am in error so that I can not continue to believe what I have been instructed for years now. I do know this it isn't required by our local codes last I looked and I can probably count on one finger how many times I've seen this done on single family dwellings with metal water pipes in my area.

Having been retired for awhile now this is going to cause me to go over to a new constructuion home sub-division and see if this bond at the hot water tank is being done. That is if I can find a new construction home sub-divison in these times


I have no idea the requirements of Canadian code. I failed to check the location.... my mistake. I do not believe it is a code requirement to bond the hot and cold pipes in the USA.

1. I have repeated the location multiple times in resonses to your questions/posts.

2. I have pointed out multiple times the VINTAGE of the homes (Pre-1980s electrical). Relatively recently have four wire circuits been required for 240v appliances.

3. I have listed several states which expressly require or at least did. You'll note most have UPC based but highly rewritten plumbing codes as well.

4. Although that section of the NEC has become progressively WEAKER in this regard, even between 05 and 08, the bonding requirements are still there. You keep ASSUMING that EVERY home with metalic/conductive potable plumbing has a complete electrical path for the purpose of bonding at least in ONE Mixing valve - one that allows a continuous sufficiently non-resistant connection between hot and cold and that this path is ALWAYS uninterrupted, least resistant, etc. AND that any circuit that MIGHT energize the metalic plumbing or the water would also make circuit ecg contact, AND that the circuit ecg would be SUFFICIENT for the ENERGIZATION. You assume every circuit in every home of a vintage EXCEEDING 30 years would even have a CONTINUOUS ecg! However, the subject was bonding not grounding and we are discussing ELECTRIC Storage type Water Heaters.

I want you to THINK about that for a few moments, really THINK about it. Then I want you to remember what was allowed in the 70s and even the early 80s when modifying OLDER systems. Think outside of the city water/sewer box too. Expand your thinking to the mechanical, plumbing codes and PRE-1980s Electrical, not just limited by your personal interpretation of the instant unammended NEC. Think about the "improvements" in plumbing supplies - ceramic, plastics, anti-scald devices (not all the same), pot metal, the days of 100% brass (and the lead) are no longer the norm, as are un-insulated soft copper supplies. More and more there are stop cocks for supplies as well (esp. in older homes where somebody has had to have a swap-out or repair) few inches of red brass and you've created a dielectric - that cock may have nylon, plastic, etc..

Finally, and no personal slight is intended here, but Kansas was not and is not "known" for its stellar performance, standards of code compliance, application, quality, etc. begining with stability and suitability of the site itself for building in the first place, quite the opposite.

P.S. your later post, not quite. I think you might be confusing the GEC with an ECG and confusing bonding with grounding??

The bond is cold pipe supply side of the shut off valve to the WH to the hot outlet pipe (just like you would bond/jumper over the cold if there were valves and a water meter in the home after your GEC bond to the water pipe).

The HOT plumbing pipes must be bonded to the cold - the dielectric nipples (which may or may not also contain heat traps) don't "act like a dielectric" they ARE dielectric - as long as the insulator (the plasticy looking isolator/insulator/inhibitor) hasn't failed, which when the average DIY solders on the coupling connection so close, often do.

From post #20

Originally Posted by H.G. Watson, Sr.

The plastic inserts by design impede/break continuity/free passage/exchange of electrons between the galvanized nipple and the copper electrons so that they do NOT pass directly between the two metals, and this is of course its intended purpose so as to arrest GALVANIC corrosion.

I believe you have missed the mark as to just what a dielectric insert or union by design accomplishes as far as inhibiting a transfer of electrons not only one direction as in DC voltage but both directions transfer in AC voltage.

H.G.
Are you saying there is not electrical continuity between the hot and cold at the water heater?

What prevents the electrons from jumping from the metal threads of the tank, onto the metal threads of the nipple, crawling up the side of the nipple, and finally jumping onto the copper threads of the pipe fitting?

The plastic insert inside the nipple only prevents the electrolyte (water), which is needed for galvanic action, from contacting the dissimilar metals in close proximity.

Originally Posted by Vern Heiler

From post #20

H.G.
Are you saying there is not electrical continuity between the hot and cold at the water heater?

What prevents the electrons from jumping from the metal threads of the tank, onto the metal threads of the nipple, crawling up the side of the nipple, and finally jumping onto the copper threads of the pipe fitting?

The plastic insert inside the nipple only prevents the electrolyte (water), which is needed for galvanic action, from contacting the dissimilar metals in close proximity.

Next you'll tell me you've grabbed your meter and found continuity with water in the pipes.

Glass is an insulator. That "tank" if steel (not copper, not stainless steel) would be glass lined (glazed).

The dielectric nipple interupts the connection. It (the metal piping system both cold potable and hot potable water) must be bonded, this is why we must jumper (from the cold conductive metal supply pipe upstream of the WH shut off valve to the output side hot water line), just as we do over shut off valves and water meters, amongst others. This is regarding the PICTURED and DESCRIBED installation as presented in the Original Post of this topic thread (electric WH, pre 1980s electrical system, BC Canada, copper plumbing).

Mark Ode and the other enginers at UL have published on this very subject explaining the WHY (imagine - advances in plumbing materials!) and HOW to interpret this very code section and evaluate the situation. Mark's article in Electrical Contractor February 2003 should be an easy enough read for you, VH. We want the least resistive path back to the panel so the circut breaker will open not via a hot fixture and through you.

Originally Posted by Rolland Pruner

I agree with the others except in your area do you require Earth Quake straps??



Rolland Pruner

Livermore, Ca

How is bis in your area Rolland?

best

ron

Originally Posted by Vern Heiler

From post #20

H.G.
Are you saying there is not electrical continuity between the hot and cold at the water heater?

What prevents the electrons from jumping from the metal threads of the tank, onto the metal threads of the nipple, crawling up the side of the nipple, and finally jumping onto the copper threads of the pipe fitting?

The plastic insert inside the nipple only prevents the electrolyte (water), which is needed for galvanic action, from contacting the dissimilar metals in close proximity.

I have to admit that I have always assumed that both hot and cold are soldered into screw connectors, which are screwed as tight as they can be into the top of a steel tank. The teflon tape is the only thing in the way of the bond to the steel, AFAIK. Somebody with a DMM should measure the resistance through a teflon-taped pipe joint. If it's higher than zero ohms, they're correct, we need a bonding strap.

HG, if Canadian plumbers are supposed to install a bonding jumper on any electric or gas WtrHtr that has copper plumbing, I guess it's no wonder they've all switched to Pex.
I do know that when they reinstall a tank in an old copper-plumbed house, they do not do any of the things you guys say is law up here. I'll upload a few more tanks for your scrutiny. No bonding strap to be seen anywhere.
Earthquakes only affect some of the new houses here, must be the OSB.
The old ones are built out of old growth douglas fir, and no earthquake straps are needed. Like a '56 Ford with no seatbelts.

Next you'll tell me you've grabbed your meter and found continuity with water in the pipes.

Why would I need water? Its not a good conductor anyway!

Glass is an insulator. That "tank" if steel (not copper, not stainless steel) would be glass lined (glazed).

Are you saying the female thread fittings in the top of the tank are not welded to the metal tank?

Your link to article by Mark Ode did not work. I would like to read the article if you can direct me to it.

No I'm not confusing the egc with gec or bonding with grounding I'm referring to the exact code language of 250.104(B) that the egc satisfies the bonding of the pipes to the service grounded conductor. Anyway you are correct it is possible that an egc is not present in vintage home wiring in picture 1 and it sounds like it doesn't matter in Canada anyway.

I also was keeping my head in the NEC not the local codes for every jurisdiction in the USA .. I do not have access to that information. And keeping my head in the NEC I do not see where it is "required" to bond the hot water pipe to the cold water pipe at the water heater or anywhere else.

The bond across the water meter is for the continuity of the GEC when the metal pipe is being used as the grounding electrode. Thinking along those lines of bonding then I would need to bond across all those modern materials you mentioned if they have metal pipes supplying them. If I bond at the hot water tank how does that bond across those devices and fixtures like your water meter example? How do you bond the metal shower pipe to the shower head if the mixing valve is not bonding the hot and cold to it ? And I am not assuming that mixing valves are all metal, it just that I haven't seen one in a single family home ever....RV..yes... mobile home.. yes.

However thinking as you asked me too about new materials in plumbing I can only consider that there will be at least one plumbing device that is non-metallic disrupting the bond bewteen hot and cold water pipes. But I'm having a hard time seeing how one non-metallic plumbing connection would isolate the metal pipes..and require a bonding jumper...it might isolate a fixture. But fixtures have been supplied for many years with non-metallic connections. As for mixing valves I have never seen one that was other than metal but again that is my little window of experience. But we Kansans are just not up with big boys I guess when it comes to bonding metal water pipes.

EDIT: Now for what it is worth I was on the phone earlier with a fella at a commercial building about 5 years old and I asked him to check the utility room to see if the bonding was there....guess what?... he informed me that it indeed had a bonding jumper across the water pipes above the hot water tank. So let it be known that Kansas has got a least one building with a hot to cold bonding jumper...

But lets look at the NEC 250.104(A) COMMENTARY from the nfpa hand book. It says " when it cannot be reasonably concluded that the hot and cold water pipes are reliably bonded through mechanical connections an electrical bonding jumper is required to ensure that the bonding connection is made". Some judgment is necessary for each installation.

Now assuming this is the intent of the cmp then the requirement is only if you conclude that the hot and cold are not bonded by mechanical means. Your saying the only conclusion considering modern day materials used in the metal water piping system is that the bonding doesn't exist no other conclusion is possible.

Now I agree with you the code is weak about what entails "reasonable conclusion" but IMO it sure as heck doesn't agree with your interpretation does it? I see nothing in the commentary or the code language that says it is required to bond hot and cold regardless of any conclusions to the contrary.

If I stick a multimeter across the metal pipes at the water heater and I get a good solid tone what conclusion do you get? If I stick a multimeter across a mixing valve and all three connected pipes at the shower and I get a tone what conclusion to you get? Not saying this would be the test to reach any conclusion but just exactly what is the test or judgment to reach a conclusion whether bonded or not?

Now assuming I don't have access or knowledge of the mechanical connections upstream of the hot water tank but I have continuity through the hot water tank to both pipes hmmmm... my point being if a conclusion can be reached as suggested in the commentary that does not require bonding of hot and cold then there can be no "required" bonding in the NEC code language.

If a local jurisdiction amends the NEC then I concede that bonding is required.

Now looking at 250.104(B) which is talking about metal water pipes that are likely to be energized and assuming picture 1 is a branch circuit without egc. Then a bond must be installed between hot and cold but again you must conclude that no bonding already exists between hot an cold...but this is interesting because your telling me that this vintage house probably doesnt have an egc in the supply to that hot water storage tank. Your also telling me that I cannot consider that the tank bonds the hot and cold for reasons mentioned ...how is it then that if an egc is present it is allowed to bond the metal piping by 250.104(B) according to the NEC?

If a local jurisdiction amends the NEC then I concede that bonding is required.

But I think it was clear when asked earlier to cite in the code the language that requires that hot and cold pipes be bonded.

I do not know if Ken was speaking to the CEC or NEC I thought it was the NEC ... again I made another assumption. I need to be more careful in my reading into things..

Now Kansas may not be on your scale of one to 10 but it's your scale and no one elses no slight intended...

After thinking (as requested) and discussing with others in different forums it appears I may be mistaken about 250.104(B). Meaning 250.104(A) is addressing metal water pipes and 250.104(B) is talking about other than metal water piping systems .... what say ye? The metal water pipe system is not subject to an egc bonding allowance .. a mechanical connection is required if not mechanical then an installed jumper sized correctly between hot and cold pipes regardless of any electrical interface like a electric hot water tank or so the consensus is leaning.

One last thing I should mention this discussion was about whether or not the NEC "required bonding of hot and cold" I still say it does not.

I am told quite emphatically that I am backwoods and should THINK about all the new space age products out there that any reasonable mind would concur that bonding of hot and cold is now required even though I do not see that in the NEC language. And that is what this discussion was intended in the beginning.. I still don't see the requirement language. If the commentary is the intention of the cmp then the question is about the "some judgement being used to determine bonding". If the idea is because of the advent of the new improved non-metallic products that you can reach no other conclusion than a installing a bonding jumper, then I wonder why the aurthors of the NEC have not just said a bonding jumper must be installed between hot and cold pipes? Why leave it up to ' judgment ' if no other conclusion can be made ?? Again this is assuming the commentary is correct however like I said I have been discussing with many others since this thread began and that is the concensus that the commentary is the intent of the 250.104(A) language.

No for what it is worth at this point in 2005 I installed a bonding jumper above my hot water tank from hot to cold pipes even though I concuded that I had multiple bonding points and only non-metallic connections to sink fixtures...that jumper didn't bond the metallic sink fixtures to anything. I installed it after being instructed in continued education workshops that it is a good idea in the event of a new owner modifing the plumbing system or modifications by me.

And no I didn't run downstairs in the last day in panic and install that bonding jumper. It was one of those decisions you make to improve safety above the minimums set forth by the NEC.

That said .. I was responding to Kens question...does the code (I thought the NEC) require this bonding. The answer is yes if bonding doesn't exist ...which means that you can also not bond if you have a mechanical connection that bonds the hot and cold somewhere. The bonding jumper must be accessible if installed .. so the hot water tank is a likely place to do this and makes the most sense IMO.

So I still say the language in the NEC on this subject does not say install a bonding jumper between hot and cold and no other choice. That was the intent of my posts before all the usual distraction started.

One last addition here and that is a quote from John

That particular home (picture 1) has all copper plumbing. Hot is bonded to cold at the tub faucet. Cold is bonded correctly at the main shutoff valve. No danger.

CEC requires bonding between hot and cold or so I'm being told. That house in the USA would likely never see a bonding jumper intentionally installed across the hot and cold at the hot water tank or anywhere else at the time it was built, it might be installed at a later date by someone knowledgable due to modifications causing loss of a mechanical bonding. I wonder though how often that is getting done?? I do hope the knowledge becomes wide spread and the NEC will mandate bonding jumpers in new construction so that all jurisdictions do not have to rely on local amendments to the NEC to require the 'intentional bonding' of hot and cold metal water pipes.

As far as a plumber coming along in existing construction and adding one non-metallic device breaking the bond between hot and cold metal pipes at that one point and then going to the hot water tank and installing a bonding jumper ... I'm not sure ... but I only see that possible if the plumbing codes require him to do so ... I don't know if that is the case. I would suspect this is an ongoing change and there are some plumbers that do and some that don't.

RF,

You are changing what I posted, and make conclusions missing the mark which are not logical and ascribing them to me.

What you proposed is that the bonding of orphaned potable hot water metalic piping system at ITS origin which is from the electric water heater is only required if the NEC specificially states it is orphaned, or must be at that specific location (origin, accessible, and exposed), and/or that one can presume an electrically continuous system with the potable cold water metalic piping system, and/or that if present the egc for the water heater circuit somehow is providing a bond to the origin of the metalic hot water distribution pipe, and/or the onus is to prove a negative - it is the opposite you must prove the positive - that the metalic pipe supplying the hot water is sufficiently bonded. What I stated was that if one cannot CONCLUDE an electrically continuous and least resistant path back to the panel upon untested assumptions (yours), and therefore to ground one MUST bond. (A) does not preclude or exclude requirements of (B) and vice versa. The code does not put the onus on proving a negative - but instead requires the conlcusion/proof of a positive (that the metalic piping systemS themselves are continuous and sufficiently bonded) without an orphaned section. Limitations as to the location of the origin of a metalic piping system and required bonding locations were enhanced on/about early 90s edition(s) of the unammended NEC. Through the tank, two dielectrics and a shut off valve is not bonded. The bonding (jumper) is required in the instant example for the ORIGIN of the potable hot water system "circuit".

Re-read the entire section, also please review and clarify your use of egc in the opening of your second to last post. So many errors of logic and application that I gave up after timing out so many times trying to reply, since the post contridicts itself so often, and inaccurately represents what has been said or what the NEC says on the subject repeatedly.

Originally Posted by Vern Heiler

Why would I need water? Its not a good conductor anyway!




Are you saying the female thread fittings in the top of the tank are not welded to the metal tank?

Your link to article by Mark Ode did not work. I would like to read the article if you can direct me to it.

Are you saying they are?

I didn't link to it, I referred to it, regarding modern plumbing materials and changes over the decades to be remembered.

As far as WHAT A dielectric IS that's 101. A dielectric nipple is just that. A dielectric union is just that. A threaded connection is just that. Conductive metal is just that. Dissimilar metals - just that. Teflon, is just that. Insulators are just that. Relying on the water in the pipes and an insulated (glass lined) tank for continuity or as a least resistive path to the panel is not valid, especially for an electric water storage heater.

You can't find the article yourself? Hmmmm. Threaded bonds presumed with dielectrics present - hmmm. Since I was referring to the printed article didn't think the ONUS was upon me to provide it to you, guess you're saying it is up to me to do so. I'll see what I can do to make it easy for you, I don't pack a scanner anymore. Give me a few minutes.

Seems ECmag has removed some of the older archives now only back to 05.

Lets see if this works I'll attach pdf file hosted elsewhere now let us see if this works:

JK, Perhaps I'm the only blind one here, not wanting to make assumptions, I can't tell if; this is BX cable w/armor stripped, AC cable, or NM cable sheathed in 3/8 flex metal conduit.

I don't want to assume any AHJ had approved #1 pic illustrated.
Making the assumption this is NM cable, it would usually need to be sleeved back into the sheet rock with a bushing or fitting end. Short sections of NM cable do not always require securing when the AHJ determines that they are not likely to suffer physical damage in normal use.

I know of a few AHJ's that allow the NM w/out a sleeve of protection if the cable exits behind & above the WH, and routed without contacting pipes. I personally require a protective sleeve and securing within a reasonable distance to the WH termination (how many times have we seen FMC or MC/AC pulled out from their connector)!

Note that the requirement for bushing/fitting at the open end of a sleeved NM raceway is a more recent one. AHJ's do not have authority to require an installation to be brought up to current code if the installation conformed to code in place at that time. One exception would be the AHJ determining a situation poses 'immanent danger' or deterioration.

It has been noted in this thread that the HI should only suggest a licensed electrician look at the installation. I agree with this opinion, however, it is always interesting to read all the opinions and the various directions a simple post may lead us. Remember wire nut taping?

Bob Smit, County EI

EDIT: This article was in the information packet for the continuing education class which prompted me to bond my hot cold pipes back in 2005. It is also a very good read and it does support Mr. Watsons information about new products and materials.

Bonding Metal Piping Systems | IAEI Magazine Online


BTW

Mr. Odes article .... http://www.homebuyeradvocates.com/PD...pe_bonding.pdf

Not sure if it is the right one the link posted earlier didn't work for me either. I read and have on file nearly all of Mr. Odes commentary articles. He is a very good authoritive source IMO.

Sorry I see Mr. Watson has corrected the link.

Hg

I really have spent all the time on this I care to...I understand where your 'coming from' and I never intended to disagree with anything you said about the difficulty in determining mechanical bonding. I was simply saying that the NEC does not emphatically require the bonding jumper.
I apologize if it appears I am changing what you say that is not my intention. You have a broad base of experience I do not. I worry that I am not communicating properly what my understandings are as to this subject. So I am trying to be more precise in what I say. If I was confusing on the egc vs bonding I can only say I will try harder to be more clear. I want to let this go for now I think enough has been said after the fact.

I welcome your responses to my posts and I do pay attention to what you say. I will admit when you are right and I am wrong. I believe I have shown the ability to do so in this thread and others. What was really just a simple response about the NEC code language has turned into another drawn out affair that I doubt anyone is going to get any use out of reading ...unless you have all day to read it.....

Your points of knowledge in this subject have been well taken by me and I thank you for your input and time but I need a rest from this....it is the weekend you know.

Originally Posted by bob smit

JK, Perhaps I'm the only blind one here, not wanting to make assumptions, I can't tell if; this is BX cable w/armor stripped, AC cable, or NM cable sheathed in 3/8 flex metal conduit.

I don't want to assume any AHJ had approved #1 pic illustrated.
Making the assumption this is NM cable, it would usually need to be sleeved back into the sheet rock with a bushing or fitting end. Short sections of NM cable do not always require securing when the AHJ determines that they are not likely to suffer physical damage in normal use.

I know of a few AHJ's that allow the NM w/out a sleeve of protection if the cable exits behind & above the WH, and routed without contacting pipes. I personally require a protective sleeve and securing within a reasonable distance to the WH termination (how many times have we seen FMC or MC/AC pulled out from their connector)!

Note that the requirement for bushing/fitting at the open end of a sleeved NM raceway is a more recent one. AHJ's do not have authority to require an installation to be brought up to current code if the installation conformed to code in place at that time. One exception would be the AHJ determining a situation poses 'immanent danger' or deterioration.

It has been noted in this thread that the HI should only suggest a licensed electrician look at the installation. I agree with this opinion, however, it is always interesting to read all the opinions and the various directions a simple post may lead us. Remember wire nut taping?

Bob Smit, County EI

Thanks, Bob. It's NM with a bit of damage to the insulation, old house has had several water heater installations.
I call for repairs, but would not call for a continuous feed to be cut so that the flexible conduit can be terminated in a J box. Apparently there are many here that are saying they would.
It seems I missed several "deficiencies" this week.

The first one here was leaking, so I called for a new tank. How they put it in will be the plumber's call.
The second and third one looked ok to me.
The Pushmatic needs cleaning up, though.

Picture # 3 - that is an incorrect method of installing NM in a flexible "sleeve". NM cable must be terminated in an approved fitting, stuffing it down some Flex and using black tape is not an approved method.
I would also look into the raceway fill for that installation. If that is 3/8 flexible Metal Conduit and someone stuffed #10 conductors into it - that is a violation in itself.
NEC article 348 covers this. Table 348.22 gives you the maximum number of conductors for 3/8 FMC.

You also have the issue of dissimilar metals - The Metallic flex is in direct contact with the copper pipe.

As far as cutting a continuous feed to install a junction box and such - One would not need to cut the feed. One would simply disconnect the cable from the water heater then one could install said box.

Needing to "sleeve" NM is a border line call at best anyway, but that is a discussion for another time.

Gentlemen;

I am aware that every state (and within most states, every town) has different requirements for each 'sub-code'.
Here is the official requirement for NJ on who is responsible for checking the bonding jumper.

Also, the PVC/ABS glue connection is interesting reading too.

Originally Posted by ken horak

Picture # 3 - that is an incorrect method of installing NM in a flexible "sleeve". NM cable must be terminated in an approved fitting, stuffing it down some Flex and using black tape is not an approved method.
I would also look into the raceway fill for that installation. If that is 3/8 flexible Metal Conduit and someone stuffed #10 conductors into it - that is a violation in itself.
NEC article 348 covers this. Table 348.22 gives you the maximum number of conductors for 3/8 FMC.

Thanks, Ken. That is a point we had not considered.

You also have the issue of dissimilar metals - The Metallic flex is in direct contact with the copper pipe.

It is actually hanging in space on that loose zap strap. Very creative.

As far as cutting a continuous feed to install a junction box and such - One would not need to cut the feed. One would simply disconnect the cable from the water heater then one could install said box.

How would you provide bonding for the J box without cutting the feed? Using the flex for that?

Needing to "sleeve" NM is a border line call at best anyway, but that is a discussion for another time.

Now is an OK time.

In respose to my statement: Are you saying the female thread fittings in the top of the tank are not welded to the metal tank?

Originally Posted by H.G. Watson, Sr.

Are you saying they are?

H.G. I gota say; when your right your right!








But when your wrong, you are soooo wrong!

As shown continuity does exist between the heat traps .. after all they are threaded into steel and that steel tank connects the hot and cold pipes and the copper pipes are threaded on to the heat traps so bonding is present.

I do not want to speak for HG but I interpret what he is pointing out is that corrosion at the tank or modifications using the addition of valves or other plumbing devices that do not bond thru to both hot and cold down stream of the tank due to being made of modern non conductive materials or parts .. then bonding will likely be suspect or lost eventually.

By installing an intentional bonding jumper sized correctly you will never lose that bond regardless of future events (unless removed) and it will be sufficient for lasting bonding and will carry any fault current imposed on the pipes because its amperage carrying capacity will never weaken ... new construction or old.

My whole point was just simply trying to keep things simple as to whether or not the NEC "required" that an intentional bonding jumper between hot and cold be installed ....no choice.

I say the NEC does give you a choice provided you can prove bonding throughout the water pipe system at the time of new construction or existing. And because a choice is given the the NEC does not require it.

I suppose bonding would be called for if you could prove bonding has been lost. For instance if your ohm meter should show a big resistance at the water tank or open.

Now in my limited experience I have never shown a loss of continuity at the tank but I have found 'open' between pipes directly above and below a dieelectic union like I posted earlier. Never between hot and cold metal pipes.

Knowing that the NEC is hovering on minimum standard and knowing the things HG pointed out I would agree that bonding is prudent regardless.

I was simply saying the language in the NEC does not say it is absolutely a requirement because you are given a means to determine whether to bond or not bond the hot and cold.

Now reading the local amendment that Darren posted it appears that the bonding jumper is only required on gas water heater replacements and is the responsibility of the plumber ...I also believe it is requiring a jumper if the heater is electric egc present or not, obviously due to the fact the egc would not satisfy 250.122 for awg.

Last post ....

Roger, I am sure H.G. appreciates you taking a swing at it for him, but that is not at all what he was saying.

(High-lighting is mine.)
H.G.

Next you'll tell me you've grabbed your meter and found continuity with water in the pipes.

Well I hadn't, but then I did! (without water!)

Glass is an insulator. That "tank" if steel (not copper, not stainless steel) would be glass lined (glazed).

The dielectric nipple interupts the connection. (No it doesn't!)It (the metal piping system both cold potable and hot potable water) must be bonded, this is why we must jumper (from the cold conductive metal supply pipe upstream of the WH shut off valve to the output side hot water line), just as we do over shut off valves and water meters, amongst others. This is regarding the PICTURED and DESCRIBED installation as presented in the Original Post of this topic thread (electric WH, pre 1980s electrical system, BC Canada, copper plumbing).

Mark Ode and the other enginers at UL have published on this very subject explaining the WHY (imagine - advances in plumbing materials!) and HOW to interpret this very code section and evaluate the situation. Mark's article in Electrical Contractor February 2003 should be an easy enough read for you, VH. (So how could I let this go?)We want the least resistive path back to the panel so the circut breaker will open not via a hot fixture and through you. (By the way, I did read the article. If you read it you will see that Mark Ode is talking about dielectric unions.)

John -
Bonding the box would be simple seeing how it is an NM cable. Insert a ground screw or ground clip with the grounding conductor from the NM cable around it.
I'm willing to bet that that NM Cable has had the outer sheathing removed in order to get the conductors into that flex.

Hanging loose in that tie wrap ? Then it is not supported. (another issue to correct )

Vern

Believe me I understand what your saying and the very points you highlighted baffled me also and I agree about the Mark Ode article and dielectric unions..but heat traps are not dielectric unions...

I am trying to be civil and proceed with a professional approach. I know a BS'er when I see one. I was trying to say politely what you are pointing out in my last post when I said I have never seen heat traps (as you show) break the bond between the steel tank and the pipes.


HG's explanation for those heat traps breaking the bond is off base entirely and you IMO have proved it also what glassed lined has to do with anything I'll be interested to discuss.

Originally Posted by Roger Frazee

Vern

Believe me I understand what your saying and the very points you highlighted baffled me also and I agree about the Mark Ode article and dielectric unions..but heat traps are not dielectric unions...

I am trying to be civil and proceed with a professional approach. I know a BS'er when I see one.I'll be interested to discuss.

Roger, do you know a BX'er when you see one?

I have another thread under Plumbing which you may like to contribute to. I think it is a case of poor bonding between the builder and the buyer.
We are having a BX session here.

Oh yeah, here's the results of the teflon test. The pipe threads cut the tape and there is less than one ohm of resistance.

Hi John

Jeeez I feel like I owe you an apology. I thought this was going to be simple....

I'll check out the BX'er thread.

Originally Posted by Roger Frazee

Hi John

Jeeez I feel like I owe you an apology. I thought this was going to be simple....

I'll check out the BX'er thread.

Nope, you owe me nothing so far.
Only read what I want to read, 2 or 3 sentences max. .

John K.

Thanks for the pictures.

Picture number four, "the Pushmatic needs cleaning up."

I have an on going job, with a Pushmatic that looks a lot like the
one shown in picture number four.

Now the E.I. is involve, saw there was now service disconnect either
inside or outside, and wants one installed. I said no way.

But not want to get into p. contest, with him, may I look to you for some
help.

The house is older than god, and the service panel is old, I cannot go
back thru old edition of the NEC for what was require, in what years
when its comes to panels. The whole is running thru this one electrical
panel, without any Service Disconnect.

And yes it bolt down onto a hot buss.


Any help here. This panel is Connecticut Electric.

Originally Posted by Robert S. Mattison

John K.

I have an on going job, with a Pushmatic that looks a lot like the
one shown in picture number four.

Now the E.I. is involve, saw there was now service disconnect either
inside or outside, and wants one installed. I said no way.

The house is older than god, and the service panel is old, I cannot go
back thru old edition of the NEC for what was require, in what years
when its comes to panels. The whole is running thru this one electrical
panel, without any Service Disconnect.

And yes it bolt down onto a hot buss.

Any help here. This panel is Connecticut Electric.

IMHO, in my humble opinion, if you install a service disconnect ahead of that antique panel, you will be raising the value of your home and maybe preventing a fire as well.
The Pushmatic from my inspection was not the service equipment, it is a "remote distribution panel" only. If the electrician has any doubts about that panel, he may upgrade to new equipment, a wise choice.
Why are you saying "No way" to the EI, if you know that the rules were changed to make you safer in your old home. The new breakers are far superior to the old stuff, even the Pushmatics, which I hear are good, but what about rust and tarnish?

Thanks for your reply. Very helpful. Robert

Originally Posted by Roger Frazee

Vern

Believe me I understand what your saying and the very points you highlighted baffled me also and I agree about the Mark Ode article and dielectric unions..but heat traps are not dielectric unions...

I am trying to be civil and proceed with a professional approach. I know a BS'er when I see one. I was trying to say politely what you are pointing out in my last post when I said I have never seen heat traps (as you show) break the bond between the steel tank and the pipes.


HG's explanation for those heat traps breaking the bond is off base entirely and you IMO have proved it also what glassed lined has to do with anything I'll be interested to discuss.

No it is you and VH who are OFF BASE. How did I know someone was going to pull out a multimeter throw a few MILI-amps and play the continuity game and think that in anyway had anything to do with the requirements of bonding here. <yawn>

CMP5 has repeatedly reiterated on the subject of split threads.

THOSE ARE DIELECTRIC NIPPLES. Those threaded nipples to the glass lined tank (EQUIPMENT) DO NOT CONSTITUTE A "BONDING" of the metalic piping systems to each other, and EVEN IF you could PROVE IT WAS A SUFFICIENT BONDING (which btw CMP-5 has been expresly stating IT IS NOT since at least 1987, threaded into the bungs - directly? no tape or teflon containing dope?) THE COLD water metalic piping system AND THE HOT water metalic piping system HAVE TO HAVE A JUMPER THERE - for WHEN/IF that Electric Water Heater is MECHANICALLY and electrically DISCONNECTED for servicing, repair, or REPLACEMENT!

THERE ARE TWO, YES 2!!!! WATER PIPING SYSTEMS. COLD POTABLE WATER PIPING SYSTEM and HOT POTABLE WATER PIPING SYSTEM.

The Electric Storage Type Water Heater is NOT A PIPE and is NOT PART OF EITHER WATER PIPING SYSTEM.

Can you AGREE that there is a HIGH LIKELYHOOD that the Water Heater will require being electrically and mechanically DISCONNECTED at some point in the future (and has already in the past) for SERVICING, MAINTENANCE, REPAIR, or REPLACEMENT?!?!?!??!?!?!??!?!??!

Do you UNDERSTAND that your continuity test with your Multimeter is MEANINGLESS as regards to a test which proves ability to safely and reliably offer a least resistant path for SERVICE current?!?

Do you UNDERSTAND the PURPOSE is to maintain an EQUAL POTENTIAL for BOTH THE COLD WATER PIPING SYSTEM AND THE HOT WATER PIPING SYSTEM, EVEN WHEN THE WATER HEATER ITSELF IS PHYSICALLY MECHANICALLY and ELECTRICALLY (in the process of being or has been completely) DISCONNECTED?

When the NEC was simplier/shorter the requirement was combined and spelled out in ONE place. The CMP has broken up 250 and spoken on this subject so many times, it is as I stated. You have to CONFIRM there is a RELIABLE BOND ELSEWHERE (THE HOT WATER METALIC PIPING SYSTEM), OR ELSE IT MUST BE BONDED HERE as I have said BEFORE, the Onus is to Confirm elsewhere BUT if the Water Service entrance IS the grounding electrode (and hopefully there is a supplemental rod or plate also), there MUST be a bond before the cold inlet shut off valve from the cold water piping system to the hot water output pipe (origin of the HOT POTABLE WATER METALIC PIPING SYSTEM).

Please See 250.68.

2005 NEC:

250.68 Grounding Electrode Conductor and Bonding Jumper Connection to Grounding Electrodes.

(A) Accessibility. The connection of a grounding electrode conductor or bonding jumper to a grounding electrode shall be accessible.

Exception No. 1: An encased or buried connection to a concrete-encased, driven, or buried grounding electrode shall not be required to be accessible.

Exception No. 2: An exothermic or irreversible compression connection to fire-proofed structural metal shall not be required to be accessible.

(B) Effective Grounding Path. The connection of a grounding electrode conductor or bonding jumper to a grounding electrode shall be made in a manner that will ensure a permanent and effective grounding path. Where necessary to ensure the grounding path for a metal piping system used as a grounding electrode, effective bonding shall be provided around insulated joints and around any equipment likely to be disconnected for repairs or replacement. Bonding conductors shall be of sufficient length to permit removal of such equipment while retaining the integrity of the bond.

2008 NEC:

250.68 Grounding Electrode Conductor and Bonding Jumper Connection to Grounding Electrodes. The connection of a grounding electrode conductor at the service, at each building or structure where supplied by a feeder(s) or branch circuit(s), or at a separately derived system and associated bonding jumper(s) shall be made as specified 250.68(A) and (B).

(A) Accessibility. All mechanical elements used to terminate a grounding electrode conductor or bonding jumper to a grounding electrode shall be accessible.

Exception No. 1: An encased or buried connection to a concrete-encased, driven, or buried grounding electrode shall not be required to be accessible

Exception No. 2: Exothermic or irreversible compression connections used at terminations, together with the mechanical means used to attach such terminations to fire-proofed structural metal whether or not the mechanical means is reversible, shall not be required to be accessible.

(B) Effective Grounding Path. The connection of a grounding electrode conductor or bonding jumper to a grounding electrode shall be made in a manner that will ensure an effective grounding path. Where necessary to ensure the grounding path for a metal piping system used as a grounding electrode, bonding shall be provided around insulated joints and around any equipment likely to be disconnected for repairs or replacement. Bondiing jumpers shall be of sufficient length to permit removal of such equipment while retaining the integrity of the grounding path.

Now, getting BACK to the SUBJECT AT HAND, i.e. PRE 1980 ELECTRICAL SYSTEM,

1975 NEC:

250-80 Bonding of Piping Systems. All interior metal water and gas piping that may become energized shall be bonded together and made electrically continuous. An equipment bonding jumper sized in accordance with Table 250-95 shall be connected between the bonded piping system(s) and the grounding electrode conductor at the service disconnecting means.
Bonding to sewer piping and metal air ducts within the premises will provide additional safety.

CMP-5 circa 1987:

"The comment noted that grooved-joint piping systems have questionable continuity due to the rubber gaskets between joints, but that other systems such as those with sweated joints had unquestionable continuity. Even the grooved-joint systems can be made continuous with bonding jumpers "

The reason for the rule "is the prevalence of other trades disrupting continuity of interior water piping without appropriate supervision."


Whether or not the cold water service is being used as the grounding electrode, the HOT potable metalic piping system must be BONDED to the grounding electrode SYSTEM (as must the cold potable metalic piping system). All metal piping systems must be at the same potential. With the metalic water piping systemS this is very important should either water metal piping system become ENERGIZED.

H.G.
Without going into the fact that you can't defend your assertion, that two galvanized nipples screwed into a metal tank do not have electrical continuity, I must point out yet another mistake you have made.

You forgot to high light this part:

250.68 Grounding Electrode Conductor and Bonding Jumper Connection to Grounding Electrodes.

(A) Accessibility. The connection of a grounding electrode conductor or bonding jumper to a grounding electrode shall be accessible.

Exception No. 1: An encased or buried connection to a concrete-encased, driven, or buried grounding electrode shall not be required to be accessible.

Exception No. 2: An exothermic or irreversible compression connection to fire-proofed structural metal shall not be required to be accessible.

(B) Effective Grounding Path. The connection of a grounding electrode conductor or bonding jumper to a grounding electrode shall be made in a manner that will ensure a permanent and effective grounding path. Where necessary to ensure the grounding path for a metal piping system used as a grounding electrode, effective bonding shall be provided around insulated joints and around any equipment likely to be disconnected for repairs or replacement. Bonding conductors shall be of sufficient length to permit removal of such equipment while retaining the integrity of the bond.

Originally Posted by Vern Heiler

H.G.
Without going into the fact that you can't defend your assertion, that two galvanized nipples screwed into a metal tank do not have electrical continuity, I must point out yet another mistake you have made.

You forgot to high light this part:

250.68 Grounding Electrode Conductor and Bonding Jumper Connection to Grounding Electrodes.

(A) Accessibility. The connection of a grounding electrode conductor or bonding jumper to a grounding electrode shall be accessible.

Exception No. 1: An encased or buried connection to a concrete-encased, driven, or buried grounding electrode shall not be required to be accessible.

Exception No. 2: An exothermic or irreversible compression connection to fire-proofed structural metal shall not be required to be accessible.

(B) Effective Grounding Path. The connection of a grounding electrode conductor or bonding jumper to a grounding electrode shall be made in a manner that will ensure a permanent and effective grounding path. Where necessary to ensure the grounding path for a metal piping system used as a grounding electrode, effective bonding shall be provided around insulated joints and around any equipment likely to be disconnected for repairs or replacement. Bonding conductors shall be of sufficient length to permit removal of such equipment while retaining the integrity of the bond.

Vern you beat me to it ... that section of code is talking about equipment that interrupts continuity of the cold water pipe from being an electrode.... example being a water meter or a water filter supply side of the connection point of the water pipe bond from the service. It isn't talking about storage water heaters.

NEC 2005

1. Building Supplied by a Service. The metal water piping system of a building or structure must be bonded to one of the following: service equipment enclosure; grounded neutral service conductor; grounding electrode conductor where the grounding electrode conductor is sized in accordance with Table 250.66; or one of the electrodes of the grounding electrode system. The metal water pipe bonding jumper must be sized in accordance with Table 250.66, based on the largest ungrounded service conductor. Where hot and cold water pipes are electrically connected, only one bonding jumper is required, either to the cold or hot water pipe. Otherwise, a single bonding jumper, sized in accordance with 250.104(A)(1), must be used to bond the hot and cold water piping together.

There is nothing there about hot water steel tanks not being an electrically conductive path between pipes. There is nothing afaik in the NEC that requires this bonding across the hot water tank. The only time I have talked to an inspector who told me he would red tag was when hot water tank was connected to the metal piping system by non-conductive means. So somebody considers it a good bond for the pipes.

THERE ARE TWO, YES 2!!!! WATER PIPING SYSTEMS. COLD POTABLE WATER PIPING SYSTEM and HOT POTABLE WATER PIPING SYSTEM.

HG
As I read the above section posted it seems clear to me that the NEC does not consider the hot and cold piping system two systems as you lead us to believe. That is your interpretation not the NEC.

As for cmp 5 ... who's pulling who's leg they have addressed and not required this about the hot water tank requiring bonding across the hot and cold if it is replacement equipment...or any other reason Have you ever been to a sectional meeting of the CMP 5 panel ??? This question of that hot cold bonding is not the first time they have heard it ... and speaking about the NEC and not local codes they did not require bonding around the hot water tank as you lead us to believe.

Cmp 5 sees one water pipe system and does not see the hot and cold pipes as two different systems

.

This above is out of soares grounding and bonding commentary.

Now what good is it to bond across the hot water tank if at some time a section of cold water pipe bursts between the hot water tank and the water pipe bond at the piping system entry to the home if the plumber makes repairs with plastic pipe or the homeowner makes repairs with non conductive material? Would a 2/0 solid copper bonding jumper across the hot water tank do any good? The NEC cannot foresee what ifs on down the road. You can only deal with present.

You can get an answer to questions as to where a bonding jumper is required across a water heater as long as it is formed to where a yes or no answer can be given by writing to NFPA P. O. Box 9101 Quincy Massachusetts 091s00-7471

This is posted on the IBEW bulletin board at the local headquarters. So there you go have at it.

Now I've told you my position on bonding the hot and cold water pipes with an intentional jumper.. I support it. Whether it is just because it makes me feel better or cuddly is irrelevant it is recommended by many in the trade who are authorities on the subject but it is not a requirement by the NEC as you keep trying to make us believe.

I am really tired of your lack of civilty and need to demean and humilate using inacurrate pastes of code, name calling and a general pattern of disrespect to any one you can't beat into agreeing with you.... shrinks have a term for your illness "Approval Syndrome".

Your are back on my ignore list ... it makes me all FLUFFY inside..when I see .." post hidden member is on your ignore list " .....

RF,
It is you who started with the "BS", and in that vain have displayed that you are indeed without a clue. VH's adventures with his multimeter are meaningless I too can display continuity with a lone remaining strand of an otherwise intermediately broken stranded cable properly terminated with miliamp loading - without a perfect ground reference this measurement is likewise meaningless.

Again we were discussing a PRE 1980s electrical system - compromised installation as DISPLAYED IN THE FIRST PHOTO.

I am sure the specifics and nuances of grounding, earthing and bonding are quite beyond both you and VH.

This discussion is more frustrating than one with a KBR reject, its over your head, you are NOT correct even in your out of context selective quoting.

Step ONE under present NEC to address the original photo and situation as described by JK in the original post, would be to confirm the hot & cold metalic piping systemS' pipes are properly and sufficiently bonded INDEPENDENT of the Water Heater itself elsewhere. If this cannot be CONFIRMED, then a sufficiently sized BONDING JUMPER would be installed as I described. THEN and ONLY THEN could the next step (PER THE NEC) in addressing the other defects, maintenance, repairs and replacements take place.

H.G.

Hit the shower. You have lost. That was game, set, match!

Originally Posted by Roger Frazee

I was simply saying the language in the NEC does not say it is absolutely a requirement because you are given a means to determine whether to bond or not bond the hot and cold.

Bonding of metal water piping systems is not elective, it is required.

VH, regarding your shower and lost comment, no and no. If the shower is in a place either you or RF (or similar KBR-like flunkie) has interacted with, I'll pass: wouldn't want to risk the potential your incomprehension exposes the user to risk.

Jeez guys ...

1) *IF* there is electrical continuity from through the cold water piping to the water heater, through the water heater, through to the hot water piping, then the bonding of either side is done by bonding one side.

2) *IF* that path is broken or incomplete in any way for any reason, then BOTH the cold and the hot piping systems need to be bonded, i.e., a jumper installed.

3) THERE ARE jurisdictions which presume that the electrical continuity COULD BE lost and thus require bonding the two piping systems together without regard to any help from the water heater itself.

4) THERE ARE jurisdictions which presume that a metal fitting screwed into a metal tank which has a metal fitting screwed into it has electrical continuity through it whether it does not not (i.e., just the reverse of 3), these jurisdiction believe that this connection will never fail.

Okay, that is about the ELECTRICAL CONTINUITY of a FRIGGIN' METAL NIPPLE, and those metal nipples DO HAVE ELECTRICAL CONTINUITY through them.

The inner sleeve is to deal with electrolysis caused by the flow of water between the copper piping the cold side, the steel fitting, the steel tank, the other steel fitting, and the other copper pipe. That is separate from the electrical continuity of the steel itself.

It is supposed to basically as like a brass or bronze fitting does: breaks up the mis-match of dissimilar metal, in the case of brass and bronze those metal are compatible with both steel and copper, eliminating the dissimilar metals problem by connecting with similar metals to each, whereas the lined nipples work by isolating one metal from the other. There is science behind the lined nipples, but I prefer using brass or bronze (preferably brass) fittings, next would be a clean break dielectric union with the plastic isolation piece (but give me the brass fitting any day).

Welcome to the fray Jerry. (I could feel you watching )

As long as you are here, what is your interpretation of the code reference:
(B) Effective Grounding Path. The connection of a grounding electrode conductor or bonding jumper to a grounding electrode shall be made in a manner that will ensure a permanent and effective grounding path. Where necessary to ensure the grounding path for a metal piping system used as a grounding electrode, effective bonding shall be provided around insulated joints and around any equipment likely to be disconnected for repairs or replacement. Bonding conductors shall be of sufficient length to permit removal of such equipment while retaining the integrity of the bond.

Is this not regarding the use of the water pipe as the SEC grounded electrode. And would this make the whole question of a ground jumper at the water heater, only apply to the very small percentage of homes that use the water pipe as the ground electrode?

1) *IF* there is electrical continuity from through the cold water piping to the water heater, through the water heater, through to the hot water piping, then the bonding of either side is done by bonding one side.
Understood from the begining.
2) *IF* that path is broken or incomplete in any way for any reason, then BOTH the cold and the hot piping systems need to be bonded, i.e., a jumper installed.

3) THERE ARE jurisdictions which presume that the electrical continuity COULD BE lost and thus require bonding the two piping systems together without regard to any help from the water heater itself.

4) THERE ARE jurisdictions which presume that a metal fitting screwed into a metal tank which has a metal fitting screwed into it has electrical continuity through it whether it does not not (i.e., just the reverse of 3), these jurisdiction believe that this connection will never fail.

Okay, that is about the ELECTRICAL CONTINUITY of a FRIGGIN' METAL NIPPLE, and those metal nipples DO HAVE ELECTRICAL CONTINUITY through them. (This was my original challenge to H.G.) I actually deleted first post as it seemed too obvious.

The inner sleeve is to deal with electrolysis caused by the flow of water between the copper piping the cold side, the steel fitting, the steel tank, the other steel fitting, and the other copper pipe. That is separate from the electrical continuity of the steel itself.

It is supposed to basically as like a brass or bronze fitting does: breaks up the mis-match of dissimilar metal, in the case of brass and bronze those metal are compatible with both steel and copper, eliminating the dissimilar metals problem by connecting with similar metals to each, whereas the lined nipples work by isolating one metal from the other. There is science behind the lined nipples, but I prefer using brass or bronze (preferably brass) fittings, next would be a clean break dielectric union with the plastic isolation piece (but give me the brass fitting any day).
__________________
Jerry Peck, Construction / Litigation Consultant
Construction Litigation Consultants, LLC ( www.ConstructionLitigationConsultants.com )
www.AskCodeMan.com

Originally Posted by ken horak

John -
Bonding the box would be simple seeing how it is an NM cable. Insert a ground screw or ground clip with the grounding conductor from the NM cable around it.
I'm willing to bet that that NM Cable has had the outer sheathing removed in order to get the conductors into that flex.

Hanging loose in that tie wrap ? Then it is not supported. (another issue to correct )

Thanks, Ken. I don't see a better way to support the cable in that installation, but I'll accept it could be better.
Re: The termination of the flexible conduit, clamped to a stud is the standard I see here. Whether the end is taped or protected with a bushing can't be seen once the drywall is on.

I saw no need to improve on the tape in that case, unfinished basement, the wires are protected from damage, no?

I've attached a pic of a 2010 water heater installation, so this is the standard my local authorities are accepting. I suspect the Flex conduit is neither taped nor bushed.

Originally Posted by Vern Heiler

Welcome to the fray Jerry. (I could feel you watching )

I've been eating popcorn and enjoying the show, but then I ran out of popcorn, so ...

As long as you are here, what is your interpretation of the code reference:
(B) Effective Grounding Path. The connection of a grounding electrode conductor or bonding jumper to a grounding electrode shall be made in a manner that will ensure a permanent and effective grounding path. Where necessary to ensure the grounding path for a metal piping system used as a grounding electrode, effective bonding shall be provided around insulated joints and around any equipment likely to be disconnected for repairs or replacement. Bonding conductors shall be of sufficient length to permit removal of such equipment while retaining the integrity of the bond.

Is this not regarding the use of the water pipe as the SEC grounded electrode. And would this make the whole question of a ground jumper at the water heater, only apply to the very small percentage of homes that use the water pipe as the ground electrode?

First, if the water heater is in the path of the grounding electrode conductor ... someone does not know they are required to connect the grounding electrode conductor within the first 5 feet of cold water pipe after it enters the structure (if not connected to it outside the structure) and they have just screwed up royally anyway.

The solution for that is to relocate the connection of the grounding electrode conductor to in front of the water heater on the cold water line ... the hot water line should never come into the discussion.

The reason for jumping a water meter in the cold water line which is being used as a grounding electrode is for when the water heater is removed - one would not want to remove the water meter while it was being used as part of the grounding electrode path.

However, back to the water heater, which should not be discussed in the same breath as "grounding electrode conductor" ...

... the water heater and the metal water piping which is connected to it are part of the interior metal water piping system, and that system is required to be bonded. NOTHING SAYS THAT BONDING HAS TO BE DONE AT THE WATER HEATER ... yep, nothing. The water heater is, however, an excellent place to bond the water piping as the bonding connections can easily be made, and seen (i.e., inspected) on the exposed piping to the water heater. The typical house does not have many places where the piping is exposed like that, so the water heater is a CONVENIENT PLACE to make those bonding connections.

Now, the NEC does not say to "only bond the interior cold water piping system', it simply states to bond the 'interior water piping system', which includes the hot water piping system too, thus the need to bond the metal hot water piping also.

I understand and agree with all you said, but what does this mean?

Where necessary to ensure the grounding path for a metal piping system used as a grounding electrode