Dielectric Nipples

Stokehold

I know this has probably been discussed, but I am asking again:

Do plastic lined dielectric nipples on water heaters work better than brass?
I've read a lot of controversy regarding this and realize there are other factors such as mineral content in the water.
I have seen the dielectric unions fail many times but have also seen dielectric nipples and copper adapters live a long time.
So, how do the pros pipe a hot water tank?
Again, advice is much appreciated!

kcopp

It all depends on the water quality. Most all glass lines steel tanks come w dielectric nipples installed.
I have uses brass nipples from time to time in projects w/ good results. You best bet is to make sure they are installed w/ tape and dope and make sure your anode in checked and replaced on a regular basis.... or add the second anode when you install a new water heater.

hot_rod

Also, some times those nipples have dip tubes and or anode rods attached. If you replace them with brass make sure you have the same functions.

Dielectric unions are more of a concern, skip those, use the nipples.
Flex connectors may be required in seismic areas for gas and water connectors.

L Thiesen

What are the failures of dielectric unions? I have used them for 45 years now and don't have problems with them.

hot_rod

The problems I have seen when removing older tanks with dielectric union is they are often clogged up. Certainly the minerals in the water and operating temperature adds to the mineral precipitation.

I see HTP adds this caution with their tanks now.

Mark Eatherton

L Thiesen said:

What are the failures of dielectric unions? I have used them for 45 years now and don't have problems with them.

With all due respect, I've not seen a dielectric union that DIDN'T have a problem. They are their own worst enemy.

I prefer the use of a yellow brass nipple, or valve (M X C or M X Union).

I used to keep a box of bad ones on my truck so when the inspector called me out due to my lack of use, I could show him why.

ME

L Thiesen

Maybe its a water quality thing but I see brass nipples, brass drain valves and Pressure and Temp relief valves filled with crud, but not the dielectric unions.

hot_rod

I guess we all have different experiences to share.

I don't see how brass nipples and valves would crud up, but not a galv nipple or union in the same tank and condition?

No doubt drain valves plug due to the low spot where sediment falls.

The dielectric nipples have been around as long as I can remember. I like the combination of the insulating seal and also the slippery plastic lined inside. I've yet to see a dielectric nipple plug or close off with sediment deposits. They seem to be standard equipment even on large commercial 2" tank connections.

This manufacturer provides some third party testing to back up their claims.

Dielectric nipples and flex connectors are required by some AHJ in seismic areas.

Stokehold

Thank you everyone for your input!
There is going to be a reaction when any two dissimilar metals are connected in water. The quality of the water, I think, is the main consideration provided the anode is good.
However, generally speaking, I must concur with Hot Rod. The dielectric nipples that have a lipped plastic liner will help prevent buildup; but then again, if water can breach that liner at any point, you may have an issue because galvanized steel is far lower on the galvanic index than brass. That is probably why the dielectric unions fail so frequently. The gasket and plastic sleeve break down allowing a minute amount of precipitation that worsens over time.
Thanks again, everyone!

Leonard

I never understood why dielectric nipples (lined pipe) would work. Any 2 dissimilar metals in electrolyte ( water) will make an electric corrosion cell. If those 2 metals are in electric contact ( threaded together) or jumpered then there will be an DC corrosion current thru the electrolyte (water.)

All the dielectric lining does is lengthen the path of the water electrolyte, which may increase it's resistance to lower the DC current.

Seems the real solution is to install sometype of non-conductive pipe (plastic). But the tank would have to be electrically isolated from the water pipe,........ but gas lines are typically electrically bonded to the water pipes somewhere , NEC electrical code.

But I would think 3-20 ft of plastic pipe might be more effective at increasing water's resistance than only 2-3 inches of plastic nipple. Easy enough to measure the change in resistance. separate 2 peices of copper pipe with 4 inches, then 10 ft of plastic pipe, fill it all with water. and measure resistance between copper pipes, Use this method to measure milliohms.
https://www.smokstak.com/forum/showthread.php?t=149217

Len
MSME

steamfitter

"All the dielectric lining does is lengthen the path of the water electrolyte, which may increase it's resistance to lower the DC current."

Leonard,
Would you say that there is a significant difference between the dielectric nipple and the dielectric union due to length of the path? Could this be why the unions fail?

I have seen cor rosin ONLY with the union and never with the nipple in my experience. However, I have not worked on HW heaters, just HVAC.

I have attributed the failure to improper installation. Many installers over-tighten the
union, virtually crushing the rubber gasket and making the dielectric union useless.

I always try to install the dielectric nipple and if I need a union, then I add a brass union on the steel side. I avoid copper unions.

Electrical knowledge is a weakness for me and I have tried to understand the concept of dielectric use and galvanic corrosion for a long while. As I read CLEARFLOW's literature. I was under the impression that the electrical current running through the water was broken at the dielectric, thereby preventing electrolysis or galvanic corrosion. But you mention resistance and the length of the electrolyte path, which forces me to re-think this entire concept. The chemistry and physics is a bit confusing to me. Can you elaborate on the electrolyte path?
Thank you!

Larry Weingarten

Hello, There is a nice book by Fontana and Greene, called Corrosion Engineering. You can find it at www.addall.com, in the used book side of things. One of the things it discusses is how distance between dissimilar metals really matters in how much of a corrosion reaction you'll get. Dielectric unions might put 1/8" between different metals. A lined nipple puts 3". Unless you have really clean water, dielectrics are more trouble than they are worth... in my experience.

Yours, Larry

JUGHNE

Larry, do you think there are any consequences of using the WH furnished/installed 3" die nipples then using die electric unions to connect to the existing copper...don't like FPT X C fittings. And sometimes need the union capability as the copper can not be raised in a tight situation.

Leonard

I think Larry nailed it, separation distance counts! ( it should increase resistance thru the water). In general I'ld guesstimate that when using plastic pipe doubling separation distance between the dissimilar metals doubles the water resistance. So from ohms law ( V=IR) the current in the water would be cut in 1/2 , and so would the corrosion rate.


Steam fitter ...... galvanic corrosion
To put it in it's most basic form a galvanic corrosion cell can be thought of a glass of electrolyte ( water, acid , or other) with 2 strips off metal in it , each strip being a different metal. This is VERY similar to one cell , of the 6 cells that make up a car 12V battery. ( a battery is just defined as more 2 or more cells in series)

Now back to the glass of water. As long as there is no electrical connection between the 2 strips of metal then no current flows thru the water, just like in a car battery. So there is no corrosion of the metals. But as soon as there is an electrical connection between the 2 strips , current starts flowing in the water, and you start to dissolve one of the metal strips. ( the steel one if other is copper)
-----------------

Problem with real world equipment is I suspect there is frequently a connection between copper piping and the steel tank. Either thru the METAL dielectric nipples, ground wire of electrical power, or nat gas pipes ( gas metal pipes and power panel ground wire must be connected/bonded to water pipes for electrical safety, NEC code).
Don't know if your allowed to put dielectric union on gas line , but if you could then if you also put one on the water pipes to a hot water tank heater and you electrically insulated that tank from the cement floor and did not connect it to electrical ground then tank might not corrode, due to copper pipes. But separately, steel tank can rust slowly on it's own also just by being wet, just like a car does.


Dielectric unions.....I'm not fully in the heating business but have taken one apart that serviced a gas line to Roof HVAC unit. Looks like dielectric is so short that the distance they would separate the steel and copper/brass pipes is insignificant to add much resistance to the water ( due to increased separation). So I'ld guess they are next to unless for stopping corrosion, if the metal pipes connected to them are dissimilar metals.

JUGHNE

Our gas system uses die electric unions or nylon reducing bushings to prevent the buried UG pipe from being grounded. The UG steel piping is plastic coated and the pipe is cathode protected.
If the UG pipe gets grounded unintentionally it will lose that protection.
At the house entrance point the nylon bushing is used and at the meter there is a dielectric union. These are never supposed to be inside a building.
All the bushings I have removed during remodel or relocate have been cracked and always the suspect for a gas leak.
Once the steel pipe is inside the house it is required to be bonded to the electric system and gets that accomplished by the gas furnace being grounded. For the rare occasion where there is only a NG WH and no NG furnace, then the pipe could be electrically floating, grounded only by the tank sitting on concrete....again very rare situation. Code wise it should be bonded to electric service grounding system.....that won't happen in most places.

Leonard

Restaurants typically only have a gas HW heater tank, heating is by HVAC on roof. We have some old HW heaters that are passive ( no electric power) and tenant used the new PEX ??? plastic flexible piping for water lines


Thinking about my dealings with gas co I'ld suspect dielectric unions on gas lines INSIDE a building are likely not allowed. My theory is during a fire the plastic would melt and gas would leak, fueling the fire more.

Before gas co would turn on the gas they made me cap off gas lines with MALIABLE or FORGED not CAST steel caps and elbows, tenant just hacked off the gas pipes to sell off the equipment when went bankrupt.

Larry Weingarten

Hello, To answer JUGHNE, the method I like best is to leave the lined nipple in the tank and use a copper or stainless corrugated flex connector to get to the piping. I use flexes that have true dielectrics built in. This gives me both a way to work on it and not much to go wrong. Also, flexes come up to two inch. I have found that flex line washers shrink with heat, so need to be re-tightened after about six months to stay watertight.

Yours, Larry

Leonard

That made me think of an interesting solution. Attach steel pipe maybe 1 foot length to the steel tank, then use a dielectric nipple or plastic pipe.. That way the steel that corrodes will be an easily replaceable thick steel pipe that is CHEAP compared to a tank. I assume tank fittings are steel.

Or would the pipe corrode so fast that I said something stupid??

JUGHNE

Leonard, your gas company and probably code require all gas openings to be solid plugged or capped......even if there is a stop/valve on the line.

I recall reading of a house fire where the unused gas supply for the clothes dryer was not being used. The washer went out of balance on the spin cycle and walked across the room, as it did one of the water supply hoses opened the flipper handle on the 1/4 turn ball gas valve. The ball valve was not plugged. Room filled with gas and found some source....maybe the washer controls etc. Fire dept got it out quickly enough that the obvious source was easy to identify........not an urban legend....write up was in my Fire newsletter. FWIW....stuff happens.

Stokehold

Let's review this...
If the voltage potential is kept at a minimum by utilizing metals close to each other on the galvanic index (.25v or under difference) for the connections AND there is a sacrificial anode in good condition, corrosion should be minimal. As Leonard stated, the plastic lined dielectrics increase the resistance, lowering potential by lengthening path of water electrolyte.
Again, the water and it's electrolyte composition is important
The tank is steel, the dielectric nipples are galvanized steel and lined with plastic. These metals are not optimum, but close to each other at .35v difference between the two, the galvanized nipples being the highest at 1.20v and the steel tank at .85v if I am reading the scale correctly. The magnesium anode is at the bottom at 1.75v.
Now, we connect copper unions with a potential for .35v and wind up at a difference of .85v between the copper and nipples and a 50v difference between copper and steel tank.
If it wasn't for the plastic liner and a strong anode rod, I am supposing things could go bad fast. Am I understanding this so far?
Again, I have seen good results using three inch lined dielectrics and copper unions on various water ranging from wells, pond systems and city supplies. I just wanted to clear up any misunderstanding I may have about using brass.
As far as grounding, I have always grounded my gas lines. In my case, I installed black pipe throughout and it is grounded to the service entrance so the heater is grounded by its connection to the gas line. If a water heater gas supply is not grounded, would not the heater also pickup a ground through a circulator if used? I would also think all electric tanks are grounded.

steamfitter

Leonard,
Thank you for the explanation!

hot_rod

correct you are. The conductivity of the water is a big part of the equation. In a boiler or closed loop system, if you strip the minerals out wit DI or RO water, the corrosion by electrolysis drop to near zero.

With potable water we often see the opposite, a water softener is added which increases the waters conductivity, more bluish green residue at faucet and drains can be a symptom.

A basic TDS meter is one way to check the conductivity.

Here are some simple tests Jeff Persons shared, you can perform these with the meter you probably already own.