Strange corrosion - time for new gas WH?

[phoenixmac]

Hi folks, new to the forums here. I'll DIY as much as makes sense but always call in the pros for anything to do with natural gas.

Just looking at this makes me think it's time to replace the WH (10+ yo?) but am curious as to exactly what the heck is going on here.

Just what kind of corrosion is this?

 

[havasu]

Looks like electrolysis of different metals has occurred. How old is the W/H? You say 10 years old? Most last 5-8 years, obviously depending on what you buy. I'm also wondering if movement from the rigid pipes leading into the water heater had caused some damage? I myself like flexible lines leading to the water heater, along with using dielectric fittings at the top of the tank.

 

[breplum]

I think Havasu means electrolysis. But what you had is a leak(s). And likely mineral content and good old corrosion.
nothing to worry about that a new water heater won’t solve. And I too agree flex lines for sure. Here in California it’s required due to well established level of seismic activity.

 

[phoenixmac]

Thanks havasu/breplum

I saw a posting from breplum in another thread about tankless WH's and I'm thinking of replacing it with a Navien NPE-150S if it (and the needed venting) will fit in my small and old basement. They're pricey but the local utility is offering a $700 rebate on the unit. After the extras, venting and installation cost, it will probably work out about the same as replacing with a standard tank unit.

 

[phoenixmac]

Interesting on the flex lines, that's the way the plumber installed it.

I'm curious about the electrolysis though as the joints are taped. Is that just the steel in the tank fitting working it's way through the tape deformations over the years?

 

[Jeff Handy]

Teflon tape does not create a solid electrical current barrier.

The threads cut through it in places, when being tightened.

 

[breplum]

phoenixmac, I only recommend the "A" series Navien.
The reason is primary: It has a mini-tank built in, along with a pump to keep it hot, so there is no confounding and wasteful delay for hot water exiting the unit.
I don't think there is any significant galvanic corrosion going on, just correcting the word.
Brass nipples are not generally problematic even though there is a differential between steel and brass.

 

[Jeff Handy]

There could be a loose or missing jumper wire at the water meter.

Or some other source of current leak.

Or maybe the water chemistry is really corrosive.
It should be tested.

 

[havasu]

I think Havasu means electrolysis.

Thanks, darn auto-correct!

 

[phoenixmac]

phoenixmac, I only recommend the "A" series Navien.
The reason is primary: It has a mini-tank built in, along with a pump to keep it hot, so there is no confounding and wasteful delay for hot water exiting the unit.
I don't think there is any significant galvanic corrosion going on, just correcting the word.
Brass nipples are not generally problematic even though there is a differential between steel and brass.

RE 'A' series: that will teach me to pay attention to the details!

Thanks breplum

 

[havasu]

I too agree with TWH, Navien is the only way to go. I had the 240A (if I remember correctly), with the controller inside and the unit outside. PVC venting, auto recirculating. Water was a constant temp. Can't go wrong, until you get a trouble light.

 

[fixitron]

I agree on replacing the tank. I don't agree that it is corrosion from electrolysis. If it were, there would be signs of leakage where the copper piping is connected to the tank at the threaded fittings. Copper corrosion would be greenish-blue and iron would be reddish. That appears to be yellowish, such as with sulpher compounds. It appears to me that the water connections on the tank are leaking, enough to leave mineral deposits when the water evaporates, from what I would guess would be hard water, but not leaking enough to form puddles. Note that the yellow stuff heads towards the draft hood. The draft would be pulling air, and moisture, up the flue.
The PT valve is leaking as shown by the rust (iron oxide) streaks on the side of the tank. In my opinion, that is an unusual amount of corrosion for a 10 year old tank, and I suspect that it had been leaking for quite some time.
I disagree that most water heaters last 5-8 years, at least in this part of Vermont. Most are warranted against tank leaks for 6 years, some as much as 10 years. I have seen many propane (we don't have natural gas available in most of this state) water heaters that are far older, often lasting 10-20 years, and those that have leaked from manufacturing defects tend to leak in the first few years.

 

[phoenixmac]

Thanks all. Water chemistry test (private well) is slated for next week actually but I'm fairly certain there is no sulphur. I've had water heaters that go 15 years and some that go 6. The last one was in a duplex we have that lasted 8.

Fixitron - I did notice the yellow stuff heading toward the flue, to the point I wondered if some possible backdraft of various gases was causing the corrosion. Can't imagine what that would be though.

As for auto-correct, I've turned it off everywhere and every time I get the chance. I prefer any embarassing miss-statements to be my own.

 

[phoenixmac]

phoenixmac, I only recommend the "A" series Navien.
The reason is primary: It has a mini-tank built in, along with a pump to keep it hot, so there is no confounding and wasteful delay for hot water exiting the unit.

I'm thinking that using re-circulation external to the unit would involve a fair amount of costly piping, or do I misunderstand? Does not having external re-circulation conflict with the essential design of the A series?

 

[CT18]

Make sure you have the gas volume for the tankless. Many houses we did needed a meter upgrade which blew the rebate out the window. I am not a tankless fan at all, but that is your choice

 

[SHEPLMBR]

You have to make sure you do regular flushing with a tankless, and that you have minimum flow rate for one.