Leak leads to corrosion?Corrosion leads to leak?

[John Kogel]

Which is it?
Could galvanic corrosion have caused the leaks? Would bonding the hot pipes to ground help stop corrosion?


These are 6 year old gas water heaters, 80 gallon units on the 5th floor of a condo unit. Both are dripping from the same junction.

Shouldn't they have used couplings to separate the steel from the copper?

[Bruce Ramsey]

The friction of water moving through a pipe causes static electricity. Even if the pipes are bonded, the dissimilar metals form galvanic action.

Yes, there should have been brass between the steel and copper or all one metal.

[David Bell]

The friction of water moving through a pipe causes static electricity. Even if the pipes are bonded, the dissimilar metals form galvanic action.

Yes, there should have been brass between the steel and copper or all one metal.


Sure looks like a brass 90 and nipple to me.

[David Bell]

Be willing to bet the anode rods are gone.

[John Kogel]

Be willing to bet the anode rods are gone.If they're gone in 6 years, that's some serious corrosion, no?

[Vern Heiler]

Sure looks like a brass 90 and nipple to me.

It says it's "China":D

[Jerry Peck]

Sure looks like a brass 90 and nipple to me.

Looks that way to me too.

I'd say it was leaking because someone did not make up the joint properly.

Also (hard to tell from the photo - looks like there may be upon a second look) is there a proper sediment trap in that gas line?

There should also be enough space to remove and replace, but at least to operate, that T&P relief valve.

[Markus Keller]

The 90 and union are brass. The nipple looks like brass also. My understanding from talking with the guys at the plumbing supply house is poor metal content and lots of contaminants in brass from China, which is likely where those fittings are from. The color difference between older american brass and the new chinese brass is quite noticeable.
I've seen that same condition around here, so I would rule out your water as a problem. Problem may be a combination of things posted by all. 6 years is pretty quick.

[Roger Frazee]

Brass elbow to galvanized steel nipple is what I see. Likely not made up with pipe dope or not tightened properly and seeping from the first day of installation. Surface moisture bridged the two different metals and as the corrosion accelerated electrical resistance to current flow also accelerated. The galvanic corrosion attacks the galvanized steel (anode) as one would expect and the steel protects the brass or copper. Six years I would say would be reasonable for the results posted in that image.
A dielectric union between the steel and brass elbow would be a good bet to have prevented that corrosion shown in the picture.