Dielectric unions and water heaters

Users who are viewing this thread

Terry

The Plumbing Wizard
Staff member
Messages
29,942
Reaction score
3,459
Points
113
Location
Bothell, Washington
Website
terrylove.com
Edit: Still curious, how did you decide how long of a brass nipple is "sufficient separation"?

The City of Bellevue prefers 6" brass nipples too.

Using these are kind of pointless.

dilectric_union_2.jpg


This is what a dielectric union looks like in service. Is there a good reason to use something that closes off with rust and corrosion? I think not.

dielectric-union-4-years-old.jpg
 
Last edited:

Cass

Plumber
Messages
5,947
Reaction score
7
Points
0
Location
Ohio
Here the only insulated nipples that I see fail are the ones that some plumber attatched a copper FIP adap. to and then soldered tubing to it causing it to melt the plastic lining of the nipple...personaly I solder pipe to the adapter first then install it on the nipple...I have no dialetric problems when doing it that way...
 

Master Plumber Mark

Sensitivity trainer and plumber of mens souls
Messages
5,537
Reaction score
357
Points
83
Location
indianapolis indiana - land of the free, home of
Website
www.weilhammerplumbing.com
why not use flex connectors????

Better look again. The Bradford heater I just bought has steel nipples. No liner on the nipples, just the heat trap retainers. I don't know if the nipples are separated from the rest of the tank and it doesn't matter.

The nipples are steel and they sit in water. If I connect copper fittings to them, there WILL be galvanic cells and the nipples WILL corrode.

Edit: Still curious, how did you decide how long of a brass nipple is "sufficient separation"?

I know that this is falling on deaf ears, but
why not use BLACK SS Brass craft flexible water heater connectors??

female to female ...

connecting to the dielectric nipples.
on the top of the heater and then onto the copper male adaptors soldered
to the copper pipe about 18 inches away???

.isn't that about 10 times better than
just shitty old dielectric unions??
 
Last edited by a moderator:

DX

General Contractor
Messages
198
Reaction score
5
Points
18
Location
Michigan
Mark, funny you should mention those.

My plumber ended up using a version of those, only better. He used ss flex hoses with a threaded brass female at the heater end

BUT

they have brass sharkbites at the other end so you can go to copper directly. No adapters, no soldering.
 

Master Plumber Mark

Sensitivity trainer and plumber of mens souls
Messages
5,537
Reaction score
357
Points
83
Location
indianapolis indiana - land of the free, home of
Website
www.weilhammerplumbing.com
those are sweet but expensive

Mark, funny you should mention those.

My plumber ended up using a version of those, only better. He used ss flex hoses with a threaded brass female at the heater end

BUT

they have brass sharkbites at the other end so you can go to copper directly. No adapters, no soldering.

the ones you are talking about are real nice but they cost about
19 bucks each... and don't seem too flexible to me


the black ones are about 10 each and you can almost tie them in a knot .... and no future call backs for any leaks in the handful of copper fittings going to the heater...

no stress on the pipes if the heater begins to sag or their is movement in the foundation of the home...

we have probably gone through about 1000 of them

I carry 24 inch, 18 inch and 12 inch with me and they
probably save me a good hour on the install


some people like to see straight pipes going into the heater
but these definitely work much better for the union effect
 

Couch-Tuber

New Member
Messages
15
Reaction score
0
Points
0
Location
Texas
My plumber ended up using a version of those, only better. He used ss flex hoses with a threaded brass female at the heater end

BUT

they have brass sharkbites at the other end so you can go to copper directly. No adapters, no soldering.

I thought this thread was about dielectric unions and ways to isolate any current, dissimilar metals? If you use flex tube which has stainless steel jacket, you still have a conductor running the length and depend on that little plastic T flange inside the brass nut. That is about 1/32" separation wihch could form rust in bad water and create contact.

I saw the Brasscraft black pipe Mark mentions but shy'd away as I'm afraid of plastic thinking copper is stronger. But they're website says it is made from as polymer tube with polymer braid. Like PEX perhaps? THey don't say cross linked polymer. But still, no conductivity the length of the tube. Sounds like what I was looking for (see my other thread on dielectric unions).
 

Dajguy

New Member
Messages
7
Reaction score
1
Points
3
Location
Florida
Dielectric nipples have a plastic liner that is supposed to provide the break between the parts of the tank and the pipe in contact WITH the water. WATER is the electrolyte which powers electrolysis, not the mere fact that two materials are contacting each other. You could screw a steel pipe into a copper fitting and set it on your workbench, then come back 100 years later and they would be the exaCTLY same, except dirtier.

So when tightening down a fitting to the top of the dielectric nipple, is it important that the fitting compress down on the plastic that extends above the top of the nipple so that it acts as a seal to keep water from contacting the threads of the nipple and fitting (where the dissimilar metals would be making contact)? Does that protruding plastic act as any long-term seal to keep water away from the threads? Or does a proper amount of Teflon tape on the threads act as the separation?

Also, is there a greater deal of insurance in attaching a brass fitting to the nipple than a copper one? I've read in this thread about using flexible pipes to come off of the nipples to further reduce problems but this particular instance involves a solar hot water heater where I need to use copper for the solar loop and I'm having to tie in a mixing valve between the cold in and hot out between tank and house, plus an thermal expansion tank on the cold side. It seems like it would be impractical to do this in other than copper. I don't want the weak link of the system to be the connection to the tank (in this case, four ports: solar in, solar out, cold in, hot out) so I want to do it in whatever way is deemed best by those of you who have had years of experience in observing system failure at this point (although it sounds like the dielectric nipples have only been out a few years).

Thanks.
 

THE rocketman

New Member
Messages
1
Reaction score
0
Points
1
Location
TEXAS
Reading all these replies to the original question, I see that some corrosion and galvanic cell information is really needed.

Galvanic corrosion put simply is a battery with a completed circuit allowing current to flow around the completed circuit. An unconnected battery sitting on a table is half of a complete circuit. The battery is the corrosion part. Electrochemically dissimilar materials are connected thru an electrolyte. The equivalent in a fluid system (in this case, a water heater and plumbing), are the dissimilar metals used and the water as an electrolyte. A battery sitting on the table with no connections (no completed circuit) will not react inside. The dissimilar metals in the water heater and plumbing will not corrode rapidly either if they are not electrically connected. Another way that to slow corrosion is to increase the resistance of the electrolyte. There are two ways to increase the electrolyte resistance. One is to make the conductor "longer" by separating the dissimilar materials by putting more distance between them. In a battery, for example, you want just the opposite so that corrosion can take place rapidly. The dissimilar materials are only separated by a thin electrolyte film. The other way to increase the electrolyte resistance is to make the water a poorer electrolyte, making it less reactive with the materials and less able to carry a current between the two materials. A dead car battery has almost pure water in it with high resistance as opposed to a fully charged battery that has very acidic water in it that is highly reactive and conductive. So this is why soft water is more corrosive in water systems. It is more reactive and a better electrolyte. In a home water system, the softness or reactivity of the water coming to your house may not be something that you can control. But if you use if a water softener, you can dial in how soft (reactive) you make your water.
So to summarize how to reduce corrosion in the water heater and system. Use less reactive metals, separate them as best you can in the electrolyte (water), use less reactive water (hard, less acidic water is better), and above all, try not to connect the dissimilar metals to complete the circuit (don't let them touch or be connected).

So to correct a misconception, the plated steel nipples supplied on many domestic water heaters are absolutely NOT dielectric. They may have plastic inserts for masking some of its surface area or holding heat trap devices. But they are not dielectric couplings. Dielectric means NON-conductive. A dielectric coupling is made to connect between two plumbing pipes without letting the two touch electrically. The use of a dielectric coupling is like disconnecting a battery. To prove to yourself that these nipples are not dielectric, get an ohm meter. Measure the resistance between the cold water inlet nipple and the hot water outlet nipple. A dead short (less than an ohm) means there is no dielectric coupling at all. In contrast, a dielectric coupling will have more than 20 Mohms measured between its ends.
 
Messages
152
Reaction score
15
Points
18
Location
Berkeley, CA
Consider this https://www.opp.psu.edu/sites/opp/files/3.pdf
Penn State Technical Bulletin-OPP TB- 232116.100
which recommends dielectric nipples, not dielectric couplings. Though, that said, I don't see how those actually work, as there still is metal to metal connection.

There's also the
Charman Transition Pipe Nipple, a brass nipple with threads on one side, and sweat fitting on the other for copper.

--------------------------------------------
I see corrosion at pipe transitions all the time, but has anyone done any real side by side building science testing?
 
Last edited:

Reach4

Well-Known Member
Messages
38,847
Reaction score
4,427
Points
113
Location
IL
Penn State Technical Bulletin-OPP TB- 232116.100
which recommends dielectric nipples, not dielectric couplings. Though, that said, I don't see how those actually work, as there still is metal to metal connection.
The way I see it, by insulating the inside path, you reduce the voltage gradient for the ions.

Currents can still pass through the metal pipe due to powerline flows when you have a dialectric nipple, because they will pass through the steel. It's the ions in the electrolyte (conductive water) that will be given less drive.

I am thinking a dialectric nipple blocks the flow through the metal, doesn't it?
 
Top
Hey, wait a minute.

This is awkward, but...

It looks like you're using an ad blocker. We get it, but (1) terrylove.com can't live without ads, and (2) ad blockers can cause issues with videos and comments. If you'd like to support the site, please allow ads.

If any particular ad is your REASON for blocking ads, please let us know. We might be able to do something about it. Thanks.
I've Disabled AdBlock    No Thanks