New Hot Water Tank

[RAKocher]

This is the hot water connection on a 4 month old electric hot water tank. This is not due to a leak. The cold water connection looks as pristine as it did on day one. The manufacturer, without hesitation and fully under warranty, is sending a new tank nibble along with a plumber to install it. The tank nipples are dielectric. I didn't give it a thought at the time, but the tank that this one replaced failed in the same manner. This is an emergency electric standby tank plumbed in parallel to an aprox 12 yr old heat exchanger tank heated by a propane boiler. The main tank's hot & cold water connections are perfect.
Should I bond the home's hot water pipe to the cold water pipe?

 

[tom baker]

Why are you heating hot water? Sorry...
The hot and cold may already be bonded at a shower or clothes washer mixing valve.
My guess is the hot water accelerates any corrosion. But interesting no corrosion on the propane fired tank

 

[infinity]

1

 

[RAKocher]

Why are you heating hot water? Sorry...

More than one plumber has asked me the same thing. Haha

The hot and cold may already be bonded at a shower or clothes washer mixing valve.
My guess is the hot water accelerates any corrosion. But interesting no corrosion on the propane fired tank

I know, I just don't know what else to attribute it to other than an electrical potential/current. The nipple looks like it's part of the Titanic, but it's not even 4 months old, and I know the last tank that I threw into my truck to get rid of it 4 months ago looked exactly the same way.

 

[hillbilly1]

Throw the old amp clamp around it and see if there is any current flowing on it. I’ve done that on copper water lines before, and was surprised at how much current flows on it. Had five amps on a water main at a hotel.

 

[hbiss]

The tank nipples are dielectric.

So, if they are dielectric why would you expect current flowing? UNLESS it's defective and not dielectric. I believe that's why they are there, to prevent that kind of corrosion.

-Hal

 

[synchro]

Dielectric nipples are typically galvanized steel with a plastic inner liner. So unlike a dielectric union it doesn't isolate the metallic piping from each other, it just forces the path for electric currents through the water itself to be somewhat longer. So the path for galvanic currents in the water would be longer between the copper pipe and the steel water heater, which supposedly will help reduce corrosion of the water heater. In the OP's case and perhaps Infinity's, I think it's just corrosion of dissimilar metals at the direct connection of the copper fitting and the end of the galvanized nipple (and is not related to the nipple being dielectric type, notwithstanding it could be lower quality). Hot water would accelerate corrosion. Using a brass or bronze nipple (typically 6 inches will do) in order to isolate the copper from the galvanized pipe usually works well. With some building codes a brass nipple is sufficient, otherwise you may have to include a dielectric union/nipple as well.

 

[RAKocher]

Throw the old amp clamp around it and see if there is any current flowing on it.

I'll have an amp clamp and meter on Monday.

So, if they are dielectric why would you expect current flowing? UNLESS it's defective and not dielectric.

Like I said, the manufacturer's rep immediately offered a new nipple and the labor to replace it, which sounds like a common defect, but that doesn't explain the water heater before this one.

 

[RAKocher]

Dielectric nipples are typically galvanized steel with a plastic inner liner. So unlike a dielectric union it doesn't isolate the metallic piping from each other, it just forces the path for electric currents through the water itself to be somewhat longer. So the path for galvanic currents in the water would be longer between the copper pipe and the steel water heater, which supposedly will help reduce corrosion of the water heater. In the OP's case and perhaps Infinity's, I think it's just corrosion of dissimilar metals at the direct connection of the copper fitting and the end of the galvanized nipple (and is not related to the nipple being dielectric type, notwithstanding it could be lower quality). Hot water would accelerate corrosion. Using a brass or bronze nipple (typically 6 inches will do) in order to isolate the copper from the galvanized pipe usually works well. With some building codes a brass nipple is sufficient, otherwise you may have to include a dielectric union/nipple as well.

I don't know how they consider a steel nipple with a plastic liner to be dielectric.
Two things I didn't mention before, for what it's worth, a water softener is in use, which would raise the resistance of the water, and the water heater hasn't been turned on for the 4 months. The electric water heater is just in case the propane boiler fails, so in the meantime it's just gravity heated by the main tank.
It'll be interesting to see what the plumber thinks/does.

 

[Dennis Alwon]

Before reading the thread my first reaction was current on the water line but why on one pipe and not the other. I would still check that to eliminate the obvious.

 

[synchro]

I'll have an amp clamp and meter on Monday.



Like I said, the manufacturer's rep immediately offered a new nipple and the labor to replace it, which sounds like a common defect, but that doesn't explain the water heater before this one.

If your clamp meter can measure DC current then measure that too. DC current is much more likely to cause corrosion than AC, and even at relatively low current levels.
There aren't many sources of DC current in a typical home, but it could be present if there's a load with half-wave rectifiers somewhere in the electrical system.
None of this is likely to be the cause of the problem, but it's still worth checking while you're there.

 

[RAKocher]

Before reading the thread my first reaction was current on the water line but why on one pipe and not the other. I would still check that to eliminate the obvious.

Here's a better shot.
I had the tank fully insulated (it came w/ it), including about the 1st foot of both pipes. The insulation looks to be fiberglass or mineral wool. When I removed the insulation it was sticking to the tank housing with moisture, but, like I said, there's no noticeable leak, and since the insulation's been removed there's no moisture either. If you notice the yellowish residue in the picture, that's what was the damp sticky stuff. It looks more like the foam insulation under the outer housing.

 

[StarCat]

Without being able to see whats is above that remale adapter, it looks like the adapter was threaded on tight then soldered, and the flux was not cleaned off.
This , by appearance is incorrect practice for connections of this type. There is not way to do it correctly unless there is a union above that fitting.
A Corrugated Copper, SS, or SS flex hose is a better and faster connection, and easier to service.
NPT joints that have been fully tightened are not supposed to the be heated up. Also any flux of any type is supposed to be wet wiped off the joint to finish the job.
Acid type flux like C flux will cause that kind of corrosion. If the factory DE fittings are found to be lame, one can always install DE unions directly above.

 

[Dennis Alwon]

I still think you should check for voltage on the pipes. You on a well?

 

[RAKocher]

Without being able to see whats is above that remale adapter, it looks like the adapter was threaded on tight then soldered, and the flux was not cleaned off.
This , by appearance is incorrect practice for connections of this type. There is not way to do it correctly unless there is a union above that fitting.
A Corrugated Copper, SS, or SS flex hose is a better and faster connection, and easier to service.
NPT joints that have been fully tightened are not supposed to the be heated up. Also any flux of any type is supposed to be wet wiped off the joint to finish the job.
Acid type flux like C flux will cause that kind of corrosion. If the factory DE fittings are found to be lame, one can always install DE unions directly above.

Although I've done a lot of plumbing since I was a teenager I don't consider myself a plumber, so I always give it the extra TLC to make up the difference. The 3/4" copper female adapters were both sweated onto about 12" of pipe while held in a vise and then washed in a laundry tub. I then used both a couple of wraps of Teflon tape and newly bought pipe dope.on the tank nipple threads and using a 12" crescent wrench made them tight, real tight. After sliding the tank over to the existing pipes and cutting them both to match, I attached them using repair couplings, with the torch never closer than 10" to the threaded connections. And I always wet wipe my sweat joints while inspecting the amount of solder.in the joint.

 

[RAKocher]

I still think you should check for voltage on the pipes. You on a well?

The bag that my clamp on and meter are in is on the job, but I'll have them tomorrow afternoon.
I'm on a municipal water system, their sources are wells and the water is hard, so I use a softener.
All of the domestic water plumbing in the house is copper (no plastic). There are 4 one piece brass tub/shower controllers w/ sweat fittings and 3 one piece faucets w/ hard supply chromed brass tubing and compression fittings in the house connecting hot to cold. All other faucets use flexible supply lines.

 

[RAKocher]

I still think you should check for voltage on the pipes. You on a well?

My clamp on and meter are in a bag on the job, but I'll have them tomorrow afternoon.
The house is connected to a small municipal water company whose source are wells and the water is hard, so I use a softener.
Also, all of the domestic water in the house is copper, no plastic, including the underground water service, which is part of the house's GES.

 

[curt swartz]

This nipple is NOT a dielectric nipple. It simply a plastic lined nipple. The copper is screwed directly to the galvanized steel, there is no isolation.

What you are seeing is very common and expected when copper is connected directly to steel. Most of the time water heaters are connected with flexible connectors which have dielectric connectors. If you want to connect rigid copper pipe directly to the water heater you need to use a section of brass between the steel and copper.

Dielectric unions exist and are frequently used for this application but they typically disintegrate almost as fast as a direct copper to steel connection. The thin rubber washer inside the union gets compressed and does not allow enough separation between the steel and copper.

 

[hbiss]

If you compare your picture to what Infinity posted, Infinity had a leak. With yours, the oxidation clearly covers the entire female adapter and even up onto the tubing where it was apparently wiped with flux. The corrosion on the nipple itself is severe. This isn't galvanic corrosion, it's almost like the nipple and fitting was exposed to some corrosive while possibly wet from condensation.

Also, how does this setup work if that unpowered water heater is plumbed in parallel with your boiler mate? Won't cold water be diverted through it or do you have it valved off? Water sitting in that thing is ripe for cultivating Legionaries by the way.

Something is very wrong here but I would be surprised if it was because of stray current.

-Hal

 

[K8MHZ]

This nipple is NOT a dielectric nipple. It simply a plastic lined nipple. The copper is screwed directly to the galvanized steel, there is no isolation.

What you are seeing is very common and expected when copper is connected directly to steel. Most of the time water heaters are connected with flexible connectors which have dielectric connectors. If you want to connect rigid copper pipe directly to the water heater you need to use a section of brass between the steel and copper.

Dielectric unions exist and are frequently used for this application but they typically disintegrate almost as fast as a direct copper to steel connection. The thin rubber washer inside the union gets compressed and does not allow enough separation between the steel and copper.

A long time ago I had a water heater inspection done. The inspector gigged me for not having flexible lines, possibly for the reason you are stating. It was too cheap of a fix to ask for an explanation.