Thread: connection of dissimilar metals

I've seen so many like this on water heater ( this case, electric water heater)
all those connections are between copper & galv.
Should I write them up?

Similar Threads:

• Equipment ground connection
• Waste Connection
• Dishwasher drain line connection
• condensate drain connection
• NM liquid Tight to EMT connection

Ben, I write it up if I see corrosion at the connection.

Hi Ben
Corrosion may be 1 issue but it looks like the relief vavle pipe has a lot of bends in it.

Brass to galvanized connections seem to have fewer corrosion issues than copper to galvanized. But I always note dissimilar metal connections, including galvanized straps on copper pipes, too. And I will make a bigger notation when corrosion is present.

When I see corrosion at water heater nipples, I always write it up. If the corrosion is mild on an older water heater, I'll tell the client that it is unlikely that the connection will fail before the heater needs replacing. It's a judgment call based on experience and I don't recommend doing that if you haven't been doing this for years.

Originally Posted by Ben Im

I've seen so many like this on water heater ( this case, electric water heater)
all those connections are between copper & galv.
Should I write them up?

Copper and steel period. I don't know why the plumbing code limits to galvanized steel pipe only. The fitting on the water heater in your first picture came with a plastic washer that separates the copper flare and the top of the nipple. Hope they were installed correctly.

Somehow the water heater manufacturers got their nipples approved as a dielectric fitting simply because they put a plastic liner inside the nipple. Or at least they make that claim. It doesn't do anything to prevent contact between copper and steel pipe.

Originally Posted by WilliamGreen

Hi Ben
Corrosion may be 1 issue but it looks like the relief vavle pipe has a lot of bends in it.

I agree. From the picture it looks like the TPR discharge pipe does a 180 so it is pointed up and then turns again and runs horizontal to the wall where it ties into a waste pipe.

We always write it up. Just did a 1930 house last night with typical lipstick on a pig kitchen and baths remodel with direct copper to galvanized connections all over the place. None leaking at the moment but I called them out as a potential long term issue.

Thank you for the answers,,

I understand like these,,,

The TPR - improper slope
galv.== brass - if no corrosion, it's OK
galv.== copper - write it up anyway

Am I following you guys?

Originally Posted by Ben Im

Thank you for the answers,,

I understand like these,,,

The TPR - improper slope
galv.== brass - if no corrosion, it's OK
galv.== copper - write it up anyway

Am I following you guys?

Well - kind of.
BRASS is an acceptable dielectric fitting. It doesn't react with steel, whether it's galvanized or not. If there is no sign of leaking at a copper -steel water heater connection, I don't right it up as a deficiency any more. I may "advise" that the connection may deteriorate and leak over time.

I can't beat the manufacturers who claim their pipe nipple is an approved dielectric fitting and I'm certainly not taking it apart to see if it is or not. Even though as an "approved" device the nipple is supposed to be marked as such, I seldom have enough of the nipple visible to see any marking. So, if it isn't leaking or corroded, it don't call it deficient.

Originally Posted by Ben Im

Thank you for the answers,,

I understand like these,,,

The TPR - improper slope
galv.== brass - if no corrosion, it's OK
galv.== copper - write it up anyway

Am I following you guys?

On the electrical ENT connection, if there is a locknut on it, it's hanging on by a thread, ( needs to be tightened ). Often there is a little code forgiveness on normal electrical support/strapping requirements at water heaters, but that excessively large loop should be supported. Some creative seismic strapping there.

Yeah, looks like they intentionally trapped the T&P discharge line - a big no-no.

For plumbing uses, brass is compatible with steel, and brass is compatible with copper ... (brass is an alloy containing copper and zinc) ... and there are two types of brass - 'white' brass and 'red' brass (okay, there are probably a lot more 'types' of brass, but those are common terms), either will serve as a dielectric between steel and copper.

Not all metals that are not similar are considered dissimilar as the term implies. Galvanic action depends on how active one metal is relative to the other, electrically speaking.
To save yourself from sounding like a foo.. carry a copy of the galvanic chart.

Determining what is a potential problem is not as simple as comparing aluminum and brass or stainless and copper for example. Allot depends on the type and/or alloy. Type 400's stainless (martinsetic [sp?]) is more of an issue than say type 300's stainless (austinetic) (sp? again)
Another example would be 1100 aluminum which is not alloyed vs 6061 in a T6 temper or a T6-651.

Originally Posted by bob smit

Not all metals that are not similar are considered dissimilar as the term implies. Galvanic action depends on how active one metal is relative to the other, electrically speaking.
To save yourself from sounding like a foo.. carry a copy of the galvanic chart.

Determining what is a potential problem is not as simple as comparing aluminum and brass or stainless and copper for example. Allot depends on the type and/or alloy. Type 400's stainless (martinsetic [sp?]) is more of an issue than say type 300's stainless (austinetic) (sp? again)
Another example would be 1100 aluminum which is not alloyed vs 6061 in a T6 temper or a T6-651.

Well,,, when residential plumbing gets that sophisticated, I'll probably be retired and writing a book on how to make a fortune in home inspections in 1 year or less. Then I'll start my own home inspector organization with instant certifications.

Originally Posted by bob smit

Not all metals that are not similar are considered dissimilar as the term implies. Galvanic action depends on how active one metal is relative to the other, electrically speaking.
To save yourself from sounding like a foo.. carry a copy of the galvanic chart.

Determining what is a potential problem is not as simple as comparing aluminum and brass or stainless and copper for example. Allot depends on the type and/or alloy. Type 400's stainless (martinsetic [sp?]) is more of an issue than say type 300's stainless (austinetic) (sp? again)
Another example would be 1100 aluminum which is not alloyed vs 6061 in a T6 temper or a T6-651.

Also, for piping, the closeness of the materials to each other on the galvanic charts, the closer to each other the better, sometimes gets critical for various chemicals being transported through the piping.

I'm seeing the term "dielectric" utilized in many answers here and can only surmise that the use of that word is from a galvanic action preventer as opposed to the actual ability to conduct electricity.

Wouldn't any dissimilar metal, with a difference of valance electrons between such, cause one metal to be an anode and the other a cathode?

If my understanding of that scientific principle is true, then the term "dielectric" is untrue, as utilized here, correct?

Richard, read J.P.'s comment above, very true.
Yes all metals are dissimilar if they vary or are not aligned in the chart, but it does not mean they will corrode by electron exchange unless there is an atmosphere conducive to such, and closeness in proximity as stated.

Dielectric strength is resistance to this electron exchange be it galvanic action or by voltage being applied. Works for both.

I just thought I'd give the fair warning in my first post as I experienced an Inspector attempting to apply the 'dissimilar metal' to everything touching another... I had to correct him

Thanks for clearing that up Bob.

Hi, ALL &

Good input and good to remember copper is the 'sacrificial' metal when in-contact with a steel strap or other improper contact such as sheet metal...

Using plastic or copper straps on copper is the ONLY way to go !


CHEERS !

To be picky in Ca:

1. a 30"x 30" working space required before the w/h.
2. Strapping incorrect ( as mentioned above): incorrect method - incorrect height positions.
3. Labelling obscured.
4. Lower element cover trapped.
5. Can the w/h be conveniently removed from it's in-situ position? Note the dryer venting.
6. First 5ft of cold and hot to be insulated.
7. Is the elec. isolator accessible?
8. Is the drain-off valve accessible?

I see:
1) Improper PRL
2) Is that a cut out in the fire wall ? If so Repair.
3) Improperly installed straps.
4) Labels not visible. _ Recommended
5 Lines not insulated. - Recommended.
Note dis simular piping materials used.

In my area, any galvanized supply pipe is bad. It frightens Insurance agents away, and that can make the whole deal crash. I am surprised it is OK for you guys in the Golden State.

The threaded portion of a galv pipe is not galvanized, BTW.

Originally Posted by Glenn Duxbury

Hi, ALL &

Good input and good to remember copper is the 'sacrificial' metal when in-contact with a steel strap or other improper contact such as sheet metal...

Using plastic or copper straps on copper is the ONLY way to go !


CHEERS !

I believe you have that backwards.

Copper is a more noble (cathodic +) metal than steel (less noble = anionic -), steel is more noble than zinc.

In your example the steel would be sacrificed (corroded), as electron transfer (current flow) from the more noble metal to the less noble metal (flow + to -) and the less noble metal would develop a greater current density (all other things such as ratio mass contact to penetration, still electrolyte, etc. being equal). If the steel were galvanized (zinc) in your example, the Zinc would be first metal to be sacrificed (develop a greater current density, and suffer corrosion), followed by the steel.

The "sacrificial anode" in a usual run-of-the-mill storage type water heater is comprised of a sacrifical anionic metal that is less noble than the galvanized steel tank it is installed in.

If the electrolyte is not still or recirculating but controlled circulation/directional flow (such as in plumbing pipes): having the less noble (anionic) metal upstream of the more noble (cathodic) metal (and not the reverse) can help to reduce the effects of bimetallic or galvanic corrosion.

There are a multitude of "recipes" for various types of steel, and Stainless steel has a likewise a multitude of "recipes" for various alloys (oftentimes referred to as a number) each of which have distinctly different characteristics and hold a different place on the "galvanic scale" or bimetallic charts.

Additionally the ratio of the cathodic area to that of the anode needs to be considered. The rate of penetration from corrosion increases as the ratio of the cathode to anode surface area increases.

The corrosion current that flows between the cathode and anode is independent of area, but the rate of penetration at the anode is dependent on the current per unit area, that is, current density. Therefore, it is undesirable to have a large cathode surface in contact with a small anode surface. The rate of penetration from corrosion increases as the ratio of the cathode to anode surface area increases.

When using a stainless steel plate with a zinc rivet, the ratio of the cathode surface area (stainless steel plate) to the anode surface area (zinc rivet) is large, and the rivet will fail rapidly because of accelerated corrosion. When combining a zinc plate with a stainless steel rivet, the area ratio between the cathode and anode is reversed, and although more surface area is affected, the depth of penetration is small; the fastener should not fail because of corrosion.

Here the Illinois Plumbing code states that a dielectric union is to be used to join pipes of differnt metals. Or a brass transtion fitting may be used.

Section 890.350 Unions

Unions may be used in the drainage and venting system when accessibly located above ground. Unions shall be installed in a water supply system within 5 feet of regulating equipment, water heaters, water conditioning tanks, water conditioning equipment, pumps, and similar equipment which may require service by removal or replacement. Where small equipment may be unscrewed, only one union shall be required.

a) Drainage System. Unions may be used in the trap seal and on the inlet and outlet side of the trap. Unions shall have metal to metal seats except that plastic unions may have plastic to plastic seats.

b) Water Supply System. Unions in the water supply system shall be metal to metal with ground seats, except that plastic to metal unions may utilize durable, non-toxic, impervious gaskets. Unions between copper pipe/tubing and dissimilar metals shall either be made with a brass converter fitting or be a dielectric type union.