Dielectric Nipple

[Stuart Brooks]

Ref: P2904.17.1 Connection of copper pipe to galvanized pipe (in general)
A connection between copper pipe and say a galvanized pipe nipple at at water heater must use a brass fitting or or approved dielectric fitting.

Now I have run into plumber's who say the nipple that came with the water heater is, according to the manufacturer of the water heater, an approved dielectric fitting. I search some and find a fitting that is galvanized steel with a PEX liner and sold as a dielectric fitting. Manufactured in China, no real manufacturer name, "Approved Vendor". This is what is shown in Grainger's catalog, "Dielectric Nipple, Pipe Size 3/4 In, Length 4 In, NPT Connection, Material of Construction Non-Alloy Steel And PEX, Schedule Schedule 40, WOG 150 At 200 Deg F PSI, Standards ASTM A 53 ANSI B1.20.1"

I can see why it would be classified as a dielectric fitting but I don't swallow the story. If the threaded joint is not perfectly sealed to prevent even a drop of water between the metals a galvanic reaction will occur.

How are we supposed to know if this nipple is used in an installation? There's nothing on the outside to identify it. How are we supposed to know if the threads were thoroughly sealed? I suppose if there is no visible corrosion or leak evidence we just go our merry way and assume all is well? It seems to me the best and cheapest solution would be to use a PVC, PEX, or other approved plastic material for a nipple or connection.

[Dana Bostick]

I'm not buying it! Having replumbed 1000's of houses, I do have some personal field knowledge about this.

They sell "dielectric" unions and they are crap IMHO. Every one that I've seen has rusted out in a year or less. I know because I had to go back and change out a crap load of them!:( Many times they are used for a "partial repipe" or a cheap out horizontal repipe.

The only successful transition is a good 6 inch brass nipple and the proper adapters. I've had the opportunity to see old jobs years later (while completing the rest of the repipe) and connections made this way were always fine. I was even able to salvage the nipples for re-use.

[Scott Patterson]

I do not call out the fittings on a water heater unless I see corrosion. If I see corrosion I do not call the fitting out as being improper I just say that it needs to be properly corrected by a qualified plumber.

[Markus Keller]

The plumper is BS'ing you on those nipples. The nipple you show in the pic, with the white insert, that come with the tank, are not intended as dielectric fittings. I don't remember the details because its been a few years. I had a dispute case involving a defective install a few years ago. I called State or AO Smith as part of my research, don't remember which for sure. Those nipples are designed to do something regarding heat transfer, heat insolation, something to do with heat/temperature at the top of the tank, not as a dielectric fitting though from my recollection.
The ones I have seen that came with the tanks all had markings on them. Anything that was specifically manufactured for, say dielectric purposes, would have a marking. The manufacturer would put something on there to identify their unique, better, more expensive product. They aren't going to want it to be anonymous.

[Stuart Brooks]

I do not call out the fittings on a water heater unless I see corrosion. If I see corrosion I do not call the fitting out as being improper I just say that it needs to be properly corrected by a qualified plumber.


This started out with leaking joints and corrosion. Even the CPVC adapter stubs were bad and leaking. The seller replaced the water heater. The buyer wanted me to reinspect several things within the limitations I established. He wanted to know if it was done correctly. This installation with copper fittings to galvanized nipples is not correct.

This was a 6 year old house that I didn't feel good about. A lot of things thrown in to finish a project, "Re-conditioned" by the seller. Bad shingles. Certainteed settled with contractors and perhaps some got back to the original owner but the shingles were left. A number of things implied there was no real code inspection. The seller was a flipper. No slack allowed for flippers. Work done was really poor.

Corrosion - I'm the same. No corrosion nothing to fix but I add a blurb about galvanic reaction and corrosion and leaks are possible in the future.

[Eric Barker]

Me three! If the dielectric is absent and the connection has been installed years without corrosion I ignore it and do not comment. If there's no indication of trouble after several years can you honestly say that the condition is not performing as intended?

[Ken Rowe]

I call out every one I see without the nipples. The water heater in my own house lasted 6 years with copper pipes connected directly to the water heater. It started leaking from the top of the tank at the connections. Upon "forensic" examination the steel tank had corroded at the copper pipe connections inside the tank and the anode rod was completely gone. There was no corrosion visible on the exterior.

The water heater manufacturers supply these with the water heaters for a reason.

[Stuart Brooks]

I call out every one I see without the nipples. The water heater in my own house lasted 6 years with copper pipes connected directly to the water heater. It started leaking from the top of the tank at the connections. Upon "forensic" examination the steel tank had corroded at the copper pipe connections inside the tank and the anode rod was completely gone. There was no corrosion visible on the exterior.

The water heater manufacturers supply these with the water heaters for a reason.

True, that will happen ken. It seems funny to me that the code specifies only galvanized steel pipe because I sure have seen a lot of tanks rusted at the top when male copper adapters were used to connect to the water lines. Doesn't make much difference if its plastic or copper pipe. The copper, in the presence of water will corrode the steel.

If I recollect my early science history correctly, I believe this phenomena first became a big time issue when ships went from wooden hulls to steel hulls. Copper plating had been used for some time on wood hulls to prevent fouling. Seems steel hulls fouled also so someone decided the thing to do was to copper plate the steel hulls. The hulls didn't collect barnacles and seaweed but the steel disintegrated in a short time resulting in sinking ships. The practice was stopped and the hunt for a new improved anti-fouling was on.