Click here to Find a Contractor in your area.
In fairness to all, we don't discuss pricing on the Wall. Thanks for your cooperation.
Need to contact us? Visit

Dielectic unions in hydronic systems

Does anyone know where I can find proof that dielectic unions are not needed in hot water heating systems? My boss is a PE and insists that any change in metals needs a dielectic union. I have been putting copper into black fittings for years with no worries. Any help would be appreciated.


  • Kal RowKal Row Member Posts: 1,518
    cant proove, too many variables...

    from the water's acidity to the outside grounding of the pipe mounts - and if you pipe like me with both telflon paste and tape, you arent getting that good an electrical contact to begin with - such a joint is more of a capacitor than a battery ;)

    also use flanges with gaskets instead of dielectric unions, they last longer and leak less - and now that B&G makes a flange with a spring check in it - you might want to place that where you have your cast iron check valve today
  • hrhr Member Posts: 6,106

    The CDA address the issue in closed loop fire protection systems, close to a hydronic application as far as closed loop O2 free enviroment.

    Sounds like he would like to hear it from authority, put him in touch for some technical writing.

    I agree with you, if it has been problem free in your installs for that many years, that's proof enough for me. But I don't stamp drawings :)

    hot rod

    To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"
  • jackchipsjackchips Member Posts: 1,338
    Has he

    found where it says they are?

    Is he convinced yet that the world is round? :-))
  • Carl PECarl PE Member Posts: 203
    2000 IMC

    Section 1203.1.1

    "Joints between different piping materials shall be made with approved adapter fittings. Joints between different metallic piping materials shall be made with approved dielectric fittings or brass converter fittings."

    Nowhere can I find where a "brass converter fitting" is defined, so presumedly I can define it as a close nipple.

    The key word here is "approved".

    from section 202:

    "Approved - Approved by the code official or other authority having jurisdiction."

    So it mostly depends on whatever the inspector will let you get away with..

    I agree that the the dielectric unions are a waste of time. On most of the systems I've seen, that union is the most corroded part of the system.
  • Can't argue

    with a PE; just do what he tells you.

    What Carl says is true: dielectric unions are where you see all the corrosion. It's also where the leaks occur since the rubber gasket cannot take the high temperatures.

    If the PE is worried about corrosion, use minimum 4" long brass nipples.

    To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"
  • jim wallsjim walls Member Posts: 31
    dielectric connections

    One of the engineering firms we work with frequently, also requires dielectrics. However, they will not allow the use of dielectric unions. Instead the only acceptable installation must be made with either a Victaulic style # 47 or a Gruvlok # 7090 dielectric nipple,,,,,,,,,,,,,be careful, there are some manufactures who also make these type of nipples whose product may not be approved for hydronic heating systems
  • Brad WhiteBrad White Member Posts: 188
    All of the other postings

    have excellent points. Agreed that dielectric unions have a history of gasket failures, yet, I have had them installed for years with no problems. EPDM or Nitrile gaskets work well and I believe that is what the manufacturers use when they rate them for temperature.

    The other method (at the 2.5 inch size more or less where one decides copper versus steel in mains) I have seen gasketed flanges with the appropriate gasket used, so no metallic contact in presence of the electrolyte (water of course).

    Brass nipples are also used. One of these postings said 4 inch minimum and I never had any indication of length. Interesting, can you expand on that?

    My office specifies Victaulic Dielectric waterways as also posted in this thread, but also a bronze ball valve. With the disparate materials, most contractors around here seem to swear by that last one.
  • hrhr Member Posts: 6,106
    For small residential

    You can use a bronze sweat flange (Taco) and a threaded flange on the black pipe as a home made dielectric, of sorts.

    Make sure you don't use a galv coated anything in a glycoled system! The zinc doesn't like glycols, or vice versa!

    hot rod

    To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"
  • I keep hearing flanges isolate

    What do you use for non conducting bolts?

  • ScottScott Member Posts: 5,884
    I have been involved

    in installing heating systems for up to 27 years now.

    I have Never installed or seen installed dielectric fittings on a hydronic hsystem. I have never seen corrosion where the copper adapters connect to the cast iron manifolds on heating systems ! Any corrsion would simple be from a leaking fitting that was not properly tightened. I also do not beleive that pipe dope or teflon would have any dielectric capabilities. Properly tightend a fitting is mated thread to thread.

    My experience leads me to belive this is just another example of designers sitting in an office and not getting out into the real world. A closed loop system with out constant make up water would have no need of a dielectric fitting.

    Just a humble opinion of a simple contractor.


    To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"
  • Minimum

    4" of brass to separate the dissimilar metals.

    That length comes from local inspector requirements.

    To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"
  • hrhr Member Posts: 6,106

    I used to supply dielectric flange sets when I had a flange business.

    I coated the bolts with two layers of heat shrink tubing (using 3/8" bolts not 7/16) then some high temperature nylon type washers on both sides of the bolt.

    This isolated the bolts from both flanges.

    hot rod

    To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"
  • that would do it


  • Kal RowKal Row Member Posts: 1,518
    in deference to all of you...

    i have taken milli-ohm and milli-volt readings, there IS voltage generated in an avg system across dissimilar metals, there is a lot less with lots of tape/paste and still less across flanges even with non insulated bolts -
    the numbers are all the place - so i dont have enough science - just a comparison in the same system,
    but SM is right, in a system with no water loss - the so-called "batteries" neutralize in short order and wont corrode the system –

    however t would be more of a problem on a hot water tank that is an open system where if one side is copper and one cast, if the polarity is wrong - it could cancel the anodes action, - that is another reason why the same water heater with a similar load, lives a lot longer in some houses than others, "question is, do you feel lucky...." , I give them an explanation and a choice, and they usually choose to save the 50 bucks I want to isolate the system, oki-doki, I’ll be back in 7yrs with a new water heater – see if I care – 1000/7 or 1000/10, the 3yr diff is a lot more than I charge for dielectrics – and I can reuse them for the next 10yr cycle
  • Robert O'ConnorRobert O'Connor Member Posts: 688
    Dielectric Q?

    An all brass ball valve is an accepted means of protection..Been working with different PE's over the years, who wants the union and others want brass or bronze. I argued the all brass (Not composition) works best. I won!, after the unions (dielectric that is) failed and I recognized the extra $$ in T&M repairs..Robert O'Connor/NJ
  • Aidan (UK)Aidan (UK) Member Posts: 288
    proof that dielectic unions

    Prove it to him. Get a plastic bucket of water and dip the end of a piece of copper pipe in one side and the end of a piece of black steel pipe in the other. Connect the two piece of pipe to a multimeter, set to read millivolts.

    You may get a small voltage between the two pipes with plain water. The voltage indicates that galvanic corrosion is taking place. The current and the rate of erosion is dependent on the immersed surface area of the two pipes and the conductivity of the water.

    If you add something acidic (e.g., few drops of lemon juice) the voltage may increase by a factor of 3 or 5. This is what happens if you don't flush and neutralize the residues of acid/active flux (HCl) from a system.

    If you add something alkaline ( a pinch of caustic soda or any corrosion inhibitor) the voltage will reduce to 0. Ask the PE if he can explain why he thinks you'd need an electric insulator when there's no PD.

    The dielectric unions are unnecessary if the system is properly treated with corrosion inhibitors.

    I think that it's the inhibitors in the glycol solution (potassium diphosphate?) that react with the zinc galvanizing. I've found that knowledge about all things chemical is rare in the heating business, so I'm not certain about this.
This discussion has been closed.


It looks like you're new here. If you want to get involved, click one of these buttons!