Current causing corrosion?

[jerjwillelec]

A customer called and said they've been replacing an abnormal amount of (natural gas) water heaters over the last several years (2003 to present)...like every three to four years. Their plumber of course said this isn't normal and compared it to another time he found that there was current flow causing the tank to rust out and leak. He told my customer it could be the same thing. I don't necessarily disagree so I immediately went to the water main, which is not bonded to the grounding system. My thinking is that sometime before they bought the house, someone may have bonded a receptacle to a water pipe which could cause current flow. I told them I would ground the water main to the electrical grounding system to equal out the potential in hopes that would take care of the issue. Problem is, I won't know if that does it until either the tank rusts out again or it doesn't. Has anyone else ran across this before?

 

[tom baker]

Typically grounding the water lines will not cause corrosion and this bonding is required for safety.
For the grounding and bonding, this is a normally non current carrying connection, but there may be something else going on.
Corrosion can be complicated to tackle but some of the issues to look at are:
Are there dielectic unions on the water heater?
What type of plumbing galv or copper
Look at you water suppliers consumer confidence report, often this can be a water quality issue, IE the water can have natural chemicals that accelerate corrosion. What is the pH? Do they have corrosion control in place?

 

[drcampbell]

It's simple enough to put a clamp-on ammeter on the hot & cold water pipes and measure how much current is flowing, before & after bonding the water main.

It couldn't hurt to bond the hot & cold water pipes to each other.

If the corrosion is mostly on the flue side, the chimney could be [partially] clogged, causing acidic flue gasses to be deposited. (and also causing a carbon monoxide hazard) A smoking piece of incense will enable you to see the airflow, which should be into the draft hood and up the chimney.

If a gravity-vented gas appliance (water heater?) and a fan-forced-exhaust gas appliance (high-efficiency furnace?) exhaust into the same flue, the result could be the same backflow as a clogged chimney.

 

[kwired]

A customer called and said they've been replacing an abnormal amount of (natural gas) water heaters over the last several years (2003 to present)...like every three to four years. Their plumber of course said this isn't normal and compared it to another time he found that there was current flow causing the tank to rust out and leak. He told my customer it could be the same thing. I don't necessarily disagree so I immediately went to the water main, which is not bonded to the grounding system. My thinking is that sometime before they bought the house, someone may have bonded a receptacle to a water pipe which could cause current flow. I told them I would ground the water main to the electrical grounding system to equal out the potential in hopes that would take care of the issue. Problem is, I won't know if that does it until either the tank rusts out again or it doesn't. Has anyone else ran across this before?

What is failing? Heat exchanger, combustion components or the tank? Maybe just where water inlet/outlets attach to the tank?

Heat exchanger or combustion component failures may be because exhaust or make up air issues or poor fuel/gas mixtures and nothing to do with electrical problems.

 

[Jraef]

...
Are there dielectic unions on the water heater?

One would hope eh?

I had all of my pipes replaced a few years ago by a licensed plumber. No dielectric unions on the water heater. I caught it and told him he needed to add them, he said he had no idea what I was referring to! Maybe he was just acting dumb to see if I would flinch, but I didn't and refused to pay him until the job was done. He left and came back with some and put them in. Still, I was disturbed to see him try that stunt and wrote an email to the Contractors License Board. Turned out I wasn't the only one complaining.

 

[Barbqranch]

There is more than one type of anode rod for water heaters, depending on the qualities of the water. Also, if a plain zinc one (the most common type) is called for, you can get an additional one that goes in the hot water outlet. It serves both as a zinc rod and an nipple on top of the water heater.

 

[Jraef]

There is more than one type of anode rod for water heaters, depending on the qualities of the water. Also, if a plain zinc one (the most common type) is called for, you can get an additional one that goes in the hot water outlet. It serves both as a zinc rod and an nipple on top of the water heater.

They are called a "sacrificial anode". But that's only done if the water has corrosive elements, right? I've always lived in places where we've never needed one, I've just heard about them.

http://www.castanet.net/news/About-the-House-Hugh-Cairns/55912/Sacrificial-anodes

 

[GoldDigger]

AFAIK off the shelf water heaters normally contain a magnesium sacrificial anode assembly screwed into the top of the tank.

 

[jerjwillelec]

Typically grounding the water lines will not cause corrosion and this bonding is required for safety.
For the grounding and bonding, this is a normally non current carrying connection, but there may be something else going on.
Corrosion can be complicated to tackle but some of the issues to look at are:
Are there dielectic unions on the water heater?
What type of plumbing galv or copper
Look at you water suppliers consumer confidence report, often this can be a water quality issue, IE the water can have natural chemicals that accelerate corrosion. What is the pH? Do they have corrosion control in place?

I'm not sure on the dielectric unions as I'd not even thought of it until you mentioned it. I'll check.
Mostly copper mixed with some PVC
Water quality is unlikely because he's done some neighborhood research and no one else is struggling with this.

 

[jerjwillelec]

It's simple enough to put a clamp-on ammeter on the hot & cold water pipes and measure how much current is flowing, before & after bonding the water main.

It couldn't hurt to bond the hot & cold water pipes to each other.

If the corrosion is mostly on the flue side, the chimney could be [partially] clogged, causing acidic flue gasses to be deposited. (and also causing a carbon monoxide hazard) A smoking piece of incense will enable you to see the airflow, which should be into the draft hood and up the chimney.

If a gravity-vented gas appliance (water heater?) and a fan-forced-exhaust gas appliance (high-efficiency furnace?) exhaust into the same flue, the result could be the same backflow as a clogged chimney.

All very interesting and helpful...thank you, I'll pass it on

 

[jerjwillelec]

What is failing? Heat exchanger, combustion components or the tank? Maybe just where water inlet/outlets attach to the tank?

Heat exchanger or combustion component failures may be because exhaust or make up air issues or poor fuel/gas mixtures and nothing to do with electrical problems.

I'm not exactly sure except there was a leak. I'll pass this on as well

 

[FionaZuppa]

I'm not sure on the dielectric unions as I'd not even thought of it until you mentioned it. I'll check.
Mostly copper mixed with some PVC
Water quality is unlikely because he's done some neighborhood research and no one else is struggling with this.

the dielectric unions dont solve the issue of current passing through, mineral content in the water can still allow current to flow. if it is suspected, then 1) is the neutral from xfrmer to panel 100%? if it is then 2) bond both in & out pipes of heater to the bare gnd side in panel (or a known nearby bare gnd that is 100% back to panel), then amp clamp the wire to see if there is current coming off of the water pipes, you may need to turn things on though (washer, stove, other) as the src of the current may be something specific to a circuit. a neighbor's bad wiring could also cause some of their current to traverse the home to get back to xfrmer, etc.

and to note, mA of current is good enough for electrolysis to happen.

 

[kwired]

Electrolysis needs DC current to make it work. AC current in the pipes doesn't cause this action. Dissimilar metals and an electrolyte however creates a battery and DC current will flow in that circuit.

 

[ggunn]

Electrolysis needs DC current to make it work. AC current in the pipes doesn't cause this action.

Sure it does. It just switches from corroding to plating 120 times per second.

 

[Tony S]

AC does cause electrolysis as we found out on the cooling water system on an induction furnace that used brass fitting for the final coil connections.

The leakage current leached the zinc out of the fittings and deposited it in the heat exchangers. Eventually the fittings became porous. The only way we sort of cured it was by increasing the length of the insulated water passages and using SS coil fittings, not ideal for 4000A.

The water was deionised and so of relatively low conductivity