Dielectric Unions - Copper to Steel Pipe
Dielectric plumbing connectors:
Tthis article describes the building code and practical requirements for a dielectric fitting when connecting copper piping to steel pipes or steel fittings such as at the top of a water heater.
We explain the effect of galvanic corrosion at copper to steel pipe connections and we include notes on the recommendation for a jumper wire across dielectric fittings or water meteres to assure that a plumging system is properly grounded for electrical safety.
The FAQs section of this article includes questions & answers about the dielectric fitting debate, concerns for fitting leakage, and notes that a brass nipple may be permitted to connect copper to steel in some building code jurisdictons.
Green links show where you are. © Copyright 2015 InspectApedia.com, All Rights Reserved.
In most jurisdictions plumbing codes require the use of a dielectric fitting when joining steel piping to copper or other metals; some plumbing jurisdictions permit use of a 6-inch brass nipple in this location as an alternative to diaelectric fittings. Why?
When connecting iron or galvanized iron pipes to copper in buildings, often galvanic corrosion and ultimately plumbing leaks will occur at the meeting of these two dissimilar metals.
Using a diaelectric fitting or an approved brass fitting to connect these two metals, or more commonly, using plastic or bronze fittings at the joint between these two metals will avoid future corrosion and leaks.
The photo (left) shows a galvanized iron union used to connect copper to galvanized iron. In the upper image you can just make out the black bronze ring built into this plumbing connector to avoid corrosion where the copper presses against the galvanized iron.
OPINION: in our experience having inspected thousands of homes, we find that original plumbing in older homes rarely made use of a dielectric fitting or even a brass nipple at copper to iron or steel piping - and corrosion is often found at one or several such connections.
In new construction where more careful code compliance and enforcement are applied, dielectric or brass fittings are more common. Some plumbers opine that "lined" nipples sold by some manufacturers for water heater hookup reduce the galvanic corrosion risk at those connections. - Ed.
How do we explain that in some buildings we see direct copper-to-iron pipe connections with no corrosion? Luck? Maybe.
But the corrosivity of the water is probably a factor in how rapidly copper-to-galvanized pipe connections will corrode and leak.
Spelling note that may help some searches: it's not dialectic pipe fittings, but dielectric pipe fittings.
Another reason we may not always see rapid galvanic corrosion at copper to steel pipe connections is the use of teflon tape (or less-so, pipe dope or sealant compound) that provides a thin non-metallic contact surface between the two dissimilar metals.
Nevertheless, both to comply with plumbing codes and to avoid leaky water pipes, use an approved diaelectric coupling or an approved brass nipple (if local codes permit) when connecting copper to steel pipes or fittings or when connecting copper pipe to the steel fitting at a water heater.
More about the galvanic scale and corrosion between dissimilar metals is at GALVANIC SCALE & METAL CORROSION.
Jumper wires are required to ground plumbing across dielectric fittings
At ELECTRICAL GROUND SYSTEM INSPECTION we discuss the need for a jumper wire around interruptions in the metal piping system, such as where a diaelect4ric fitting is installed or where a water meter interrupts continuity.
As Carson Dunlop's sketch shows at (A), if the building plumbing includes dielectric fittings or non-conductive sections of piping (such as plastic piping) then the building plumbing system may not be safely grounded.
A jumper wire is installed to assure that the building metal plumbing pipes are safely grounded to earth. The purpose of the ground jumper is to ground the building plumbing, not to use the building plumbing as a grounding system.
That is to say, we're making the plumbing safe, we're not using it to provide a ground for the electrical system.
Spelling note that may help some searches: it's not dialectic pipe fittings, but dielectric pipe fittings.
READER OPINION: On your web page ELECTRICAL GROUND SYSTEM INSPECTION, near the bottom, is a question "jumper across dielectric fittings is asking for galvanic corrosion." Your reply to the question concludes with the comment "Also I am nor sure which jumper you saw, but connecting a ground between similar metals ought not create the concern youncite."
That same page includes an illustration [shown just above in this article - Ed] showing this very connection: copper pipe joined to galvanized pipe through a dielectric union with a jumper wire around the dielectric union.
The unilateral recommendation that this connection should be jumpered is concerning: if the union is required to protect the integrity of the pipe and whatever fluid (or gas?) flows inside it, then jumpering around the union defeats that protection and may lead to accelerated failure of the pipe. I think additional consideration on this topic is warranted. Regards, G.H. 7/10/2013
Thank you GH for the comment, and for being interested in this episodic debate about using dielectric fittings and jumper wires around them as well as understanding the effects of galvanic corrosion.
I think we may be mixing two different concerns, so I summarize my understanding below and invite your comment if I'm mistaken.
The diaelectric union has the purpose of preventing corrosion that occurs between two dissimilar metals, such as copper pipe joined to galvanized pipe in a water supply system. In such circumstances the corrosion occurs at the point of contact between the two dissimilar metals. The diaelectric fitting includes an insulator, typically of plastic, to prevent the two metals from coming into direct contact.
A problem is that the diaelectric union may in some locations prevent the building's metal plumbing system from being safely grounded.
The jumper wire is installed to assure electrical continuity as a safety feature for the building's electrical system. E.g. if a toaster, plugged in, falls into a sink of water, a good electrical pathway between metal plumbing connected to the metal sink (or its metal components) and earth, assures that current flows to ground through that safe path, thus blowing a fuse or tripping a circuit breaker, thus protecting the building occupant from possibly fatal shock.
The ground wire connectors and ground wire do not themselves become points of corrosion in the piping system - at least not at any I've seen, most likely because the wire-to-pipe connectors use an intermediate metal, such as brass, between the copper pipe and (say) galvanized pipe.
We could and should include authoritative citations on a topic such as this - something for which we both should be alert and that I'd be pleased to add to the article.
I would add, though it's a bit too esoteric to imbed in the article, that stray electrical currents, combined with improper electrical grounding can have strange and damaging effects on a building's mechanical system. For example, improper neutral or ground wiring at a building can cause current to flow where it is not expected, causing damage. I found costly damage to an internal refrigeration coil in a water to air cooling system was ultimately traced to improper connections in the electrical panel.
While I'm certainly not an expert on the topic, I think the assumption that galvanic corrosion will occur only at the point of contact between the two metals is flawed. Consider for example the "lasagna cell" battery: an uncoated steel baking pan containing a lasagna (or other salty food serving as electrolyte) and covered with conventional aluminum foil. Holes will form through the aluminum foil in the places where it contacts the lasagna -- not around the edges where the foil is touching the steel.
I recently found myself in a sprinkler (irrigation) supply shop. A customer at the counter held a length of galvanized pipe with (part of) an elbow at one end. He explained that it was used in a water feature connecting underground piping to an electric pump above ground and had "rusted out" twice, each time lasting less than a year. There was basically nothing left of the sidewall of the elbow; it had been dissolved away from the outside (I could conclude this because in one area the elbow metal was gone, exposing the threads on the end of the pipe inside it).
There's no way it could have corroded so quickly without a boost from something like galvanic corrosion. But it didn't corrode at a connection between dissimilar metals; it was the side of the elbow that was eaten out. Anyway, I think he mentioned that beyond the elbow (continuing underground) was PVC pipe so there probably was no dissimilar connection in that area at all.
In the case of the wire jumper around the dielectric union, I expect it won't help any to use a brass or bronze clamp -- both are alloys of copper and are very near copper in the anodic index (see "galvanic corrosion" page on wikipedia). For that reason I believe the corrosion will still happen.
If not so, I suppose we'd be using brass or bronze couplers or unions to join the pipes and skip the business with the plastic isolator in the dielectric union. I believe your experience (not seeing corrosion at the location of a jumper around a dielectric union) supports my suggestion that the corrosion is likely to happen somewhere out of sight -- specifically, at a point where the steel pipe contacts an electrolyte (something wet, probably soil).
I do understand (and agree with) the concern about having the plumbing systems grounded for occupant safety. To that end, I'm trying to figure out how to accomplish both mechanical and electrical protection. It seems to me that the galvanic corrosion problem occurs when copper and steel are both in an electrolyte and a conductive path exists between them -- anything that looks conceptually like a copper ground rod and a buried steel water pipe, even buried apart from each other, but with some electrical bonding between them.
Given that we're talking about a plumbing joint of copper and steel, one of these is the supply pipe buried outside and the other is the plumbing through the building interior. Suppose it's copper in the dirt and steel through the building. I guess there's no problem here (and in fact no dielectric union needed) because there's no steel immersed in electrolyte with the copper. But if it's the other way round with buried steel supply, then I think it's imperative that the steel be electrically isolated from all buried copper. We can assume that the electrical system contains buried copper either in the form of a ground rod at the same building or at another installation nearby (and connected via the neutral wires and ground-bonding at the main service panels). So, in this case, I think I would isolate the buried steel with a dielectric union and install a bonding wire from the copper pipe back to the electrical panel -- I wouldn't connect wire to the buried steel at all. This building would have to get its grounding some other way, maybe through ground rods, plates, etc.
As I said in the beginning, I don't claim to be an expert on the topic. I'm genuinely interested in your thoughts, and if you have electrical inspectors you consult with regularly, it'd be interesting to hear their thoughts on the matter as well.
Regarding the damage to the refrigeration coil traced back to improper connections in the electrical panel -- that sounds fascinating. If you have the time to write, I'd love to read more about the details of the problem and especially how in the world you managed to find the cause. That must have been a treat to figure out! Ugh!
I appreciate the discussion but it seems we disagree about the locus of galvanic corrosion in plumbing systems and perhaps you didn't know that building codes require both dielectric fittings (or in some locales a brass nipple) connecting copper to steel pipes in a building as well as jumper wires around dielectric fittings when needed to assure safe electrical grounding of the plumbing system.
The jumper wire around a dielectric fitting, properly connected with approved (typically bronze) clamps to the copper and steel piping and a copper ground jumper will not itself promote corrosion at the pipe connection nor nearby.
Plumbing texts as well as field experience point out that corrosion occurs at and/or very close to (micros to millimeters) the point of contact between dissimilar metals that are further apart on the galvanic scale, such as copper and steel. Long-standing and successful plumbing practice was to connect via an intermediate metal such as brass - as the galvanic corrosion problem does not appear at, for example a copper to brass connection (there are hundreds of millions of such connections in both plumbing and heating systems today).
Diaelectric fittings, also referred to as diaelectric unions or DEUs, are required by the model plumbing codes adopted by most states and provinces, though in some U.S. states plumbing codes permit the use of brass fittings in those locations instead.
I would agree that there are episodic debates in the plumbing community about using diaelectric fittings, out of concern that some leak and others may not work as well as advertised.
Section 316 of the California Plumbing Code is amended by adding Subsection 316.2.5 to read as follows:
316.2.5 All connections between ferrous and nonferrous pipe shall be made with a six-inch red brass nipple or a dielectric union.
Ord. 559 § 2 (part), 2010: Ord. 543 § 2 (part), 2007: Ord. 499 § 2 (part), 2002: Ord. 463 § 2 (part), 1999. Formerly 15.16.170
Other building citation code sources where the requirement for a dialectric plumbing fitting is cited:
Continue reading at GALVANIC SCALE & METAL CORROSION or select a topic from the More Reading links or topic ARTICLE INDEX shown below.
Green link shows where you are in this article series.
Try the search box below or CONTACT US by email if you cannot find the answer you need at InspectApedia.
I have a home that is 7 yrs old in Burleson Tx. We are on a co-op water supply. I recently had a pin hole in my copper water pipe under the slap. I though it was the water but had the water dept lab test my water and all was normal. I then read about electrolysis could cause this. I do have a ground on my water pipe and a rod out side my main. I checked whit a meter and I do have continuity between the main panel and my plumbing and my gas pipe to the hot water. can this possible cause my plumbing to fail?
The water has been ruled out so it has to be something else the pin hole was from the inside out and my house water connections are green and blue. need some help bad. I did disconnect the one ground off the plumbing when I read about the electroylis. Please advise.
Donald McKinley 10/19/11
With just the info in your note I can only guess at some possible causes for the pinhole leak in copper water piping under your building slab.
If the pH is low <6.0, the hardness low generally<50ppm, the alkalinity low generally <40ppm, the water could be considered extremely “soft” and aggressive to the home’s metallic plumbing system. If the chlorides are elevated >100ppm this would only compound the problem. The water should be treated to make the water less aggressive by raising the pH, alkalinity or hardness. - CT DOH.
If you do nothing but fix the pipe and the problem never recurs I'd suspect the pipe itself. But if it were me, I'd also have a licensed electrician check that the home's grounding and neutral systems are properly wired, that the grounding electrodes are properly sized and installed, and that there are no stray currents on the neutral system nor shorts or leaks in the wiring system (an AFCI or GFCI can help detect these too).
Dan, concerning the pin hole pipe leak and all the green and blue plumbing from Oct. 19 & 21, 2011. I have seen this before and traced it to bad grounded (neutral) connections at the utility transformers or tap boxes causing all the neutral loads to be carried on the grounding system such as the copper plumbing.
I even seen it in one house but show the signs of trouble in a neighbors house because of a common city water pipe. This situation eats the copper water piping from the inside out and can cause green/blue water color, usually the first sign.
Also, these pin hole can develop because of excessive flux being used before sweating. The excess flux lays in the bottom of the pipe and corrodes the copper, hence pin holes on the bottom only. - Rod, electrical contractor, 8/1/12
thanks for the important and helpful comment - we agree completely. Bad electrical grounding and incorrect connections among grounding connections can cause a wide range of odd problems, including corrosion and leaks in plumbing, HVAC equipment, even equipment internal parts such as the coil in a water to air heat exchanger coil in a groundwater sourced heat pump.
The excessive solder flux corrosion problem is not one I'd realized -thanks for that tip. I suppose we could confirm that problem cause after the fact by noticing just where the copper piping leaks are occurring - all at solder joints - and then disassembling or cutting one of those joints apart to inspect the interior of the pipe. Water chemistry may also play a part in that corrosion problem.
Thanks again. Astute and helpful. If you want us to cite and refer readers to you in your area email me contact information.
You show a jumper wire across a dielectric plumbing connection (between copper & galvanized pipe). This will promote galvanic corrosion & make the dielectric connection pointless. Instead of at the connection the corrosion will now take place inside the galvanized pipe near the jumper wire clamp.
Grounding the plumbing does not make a house safer. It places half an electrical circuit through out the house. This increases the likelihood of connecting that circuit with some current. - Galvanic 9/20/12
The jumper wire across a dielectric fitting does not cause corrosion in copper to steel piping connections provided it is properly made using approved connectors (typically brass or bronze clamps).
If we don't jump across a non-conductive dielectric fitting on water piping then the water piping is not grounded. By current NEC, metal piping may not be used as a grounding conductor, but metal water piping in contact with the earth for a length of ten feet or more, that piping is indeed connected to the electrical ground system.
For protection from lightning and possibly leakage from a high voltage transformer, the current National Electrical Code (NEC) requires two grounding electrodes at a building. If one of these is water piping it is tested and must show less than 25 ohms of resistance to earth.
Typically, pinholing in copper piping that is traced to an electrical grounding problem (electrolysis) is, if we exclude neutral/ground wiring errors, traced to inadequate local grounding electrodes.
Thanks for the interesting comment. I'm not sure where your surmise takes us, since there are both code and basic safety reasons for grounding house plumbing. Also I am nor sure which jumper you saw, but connecting a ground between similar metals ought not create the concern youncite. Can you give us a citation or article to review?
The reason people ground in-building plumbing is not to provide an additional grounding conductor in a building but to ground the plumbing. Picture someone knocking a toaster into a stainless steel sink or into any sink with a metal drain and drain piping. If the sink and piping are grounded the fuse or breaker will blow. If not, the system is waiting to electrocute the building occupant when s/he touches the live water/toaster in the sink and perhaps a nearby metal faucet, radiator, or other component that is ultimately connected to earth.
Incidentally, as we discuss pinholing and bad neutral connections, keep in mind that the return path for current in a building's electrical system is not intended to be primarily through the building's local grounding electrodes. Rather it is on the neutral wire that is connected back to the pole transformer.
See LOST NEUTRAL Shocks Homeowner for details of what can happen when this connection is not made or goes bad.
(Aug 24, 2014) Eric Schemanske said:
My water supply from the city comes in through the basement. Some years back the city installed a new meter in close proximity to where the pipe comes through the basement wall, which is copper (one inch). They connected a electrical ground from the one side of the meter ( copper pipe) to where the pipe enters through the wall.
Now I have electroylis ( corrosion, big time) after
several years. The clamp they used appears not to be (?) compatible with code to prevent this from perhaps happening. Is this possible?
If the installer used a listed grounding system bonding clamp then its composition was intended to avoid corrosion due to electrolysis.
Usually the installer also includes a jumper around the water meter to assure continuity of grounding of the piping system.
I would be looking for a different electrical problem first as ruling it out is important. There may be stray voltage on your electrical system's neutral circuit or ground circuit from a short somewhere in the electrical panel or building wiring.
(Oct 28, 2014) Jason said:
Is there a need to provide a dielectric fitting in a steam system where steel piping connect to copper piping?
Is there any problem with galvanic corrosion at that point?
We have a situation where steel piping is being connected to a copper tube heating coil both on the steam supply and condensate return without any dielectric isolation.
Lucky, but still it'd have been good practice; steam condensate is in fact quite corrosive.
28 Feb 2015 Anonymous said:
can we connect bronze to steel without die electric union
Take a look at GALVANIC SCALE & METAL CORROSION for a clear explanation of this.
Use the "Click to Show or Hide FAQs" link just above to see recently-posted questions, comments, replies, try the search box just below, or if you prefer, post a question or comment in the Comments box below and we will respond promptly.
Search the InspectApedia website