glycol leak in dielectric union

Dave_61

Can I attach brass nipple to steel threaded connection?

Dave_61

glycol leak in dielectric union

I had my system repiped last winter with copper pipe. At the time, the plumber placed a couple dielectric unions where the copper met other metals.
I noticed some water on the floor yesterday, and it looks like there is a slow leak at one of the dielectric unions (past the rubber gasket....not at the threads). We have a 30% mix of glycol/water. I will try to cinch the union tighter after I buy a larger wrench. Is there any way to stop the leak without replacing/disassembling the union?
Thanks.
Dave

Brad White_9

Die-a-leak-trick Unions

Dielectric unions are prone to leaking even without glycol. Add glycol and the propensity increases as you are sadly discovering. The glycol could well be dissolving the resilient gasket. If that is the case, forget tightening of course.

Short of buying a glycol resistant gasket, I would replace the union with a brass nipple or better yet a brass ball valve. You may still have thread leaks but you can deal with those conventionally with the proper thread dope and prep.

Glycol, despite apparent viscosity, has a low surface tension. It finds more cracks and crevices than a plumbers convention.

Unknown

I don't use

I don't use die-elctric unions at all, they not worth the trouble and always leaks.. I always use copper male adts into fittings, never female adts... Gonna needs a lot of expando for plumbers convention!

hr

I've read that glycol and zinc

are not a good mix. If in fact the dielectric is a plated type, I'd get rid of it.

Even if it is not, I'd get rid of it

hot rod

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

Brad White_9

Yes you can

While dis-similar (and an interesting previous thread coverd this some months ago) it is thought that something else in the system is more anodic (sacrificial) than the joint in question.

Key is to have the nipple go into a female iron fitting, which it will by definition of course.

Iron to brass is more forgiving than steel to brass, steel being the pipe itself. You would not want a female copper coupling over a threaded steel end for example. That would set the steel up as an anode and begin corrosion. (Yes, some have existed for years but why tempt fate?)

So it is a combination of the brass nipple and cast/malleable iron coupling or elbow that is passive.

A bronze ball valve is even better.

Ken_40

In all lilihood...

Cathodic action is probably NOT the problem anyhow!

After doing all that soldering, I wonder if the system was purged of all the flux residuals? The acids in soldering flux are legendery and difficult/impossible to simply "purge" out.

Glycol and acidic residuals is a leak-city formula!

Flush the system with TSP or some similar alkaline detergent and watch the system "junk" come out the hose!

Once the system is assured to be "clean," put your fresh glycol in and test annually for the glycol properties. Old glycol has a habit of suddenly becoming a major corrosion contributor, once the inhibitors get hot and mix with other water chemistry.

hvacfreak

another option

http://www.perfectioncorp.com/water/clearflow.html

I think Grinnell or Victaulic ( one of those ) make this as well.

ALH_4

Potential

Having seen aluminum pipe nipples with female copper adapters threaded directly onto them provide service for decades, I have to think this issue is stemming from small leaks that rust the pipe near the fitting. Yes there was corrosion everywhere on the aluminum, but not just near the fittings. I'm no expert on this topic, but galvanic corrosion appears to take place in localized areas, without much reason to them that I can discern. The pipe nipple would have to corrode completely through the threads to create a leak.

Personally, I think people are reaching for a complicated solution to a simple problem. The fitting developed a leak. I think it is much more likely that the differing expansion rates are a much more probable explanation for a leak than galvanic corrosion. However, throw a single steel pipe nipple in a system with a stainless boiler and copper piping, and you might be in for problems. That nipple will have everything else in the system attacking it.

Why would brass break the continuity?

-Andrew

Mitch_6

From what I have seen

if the system is closed like a forced hot water boiler the mix of metals is not as critical for electrolysis.

Still never hurts to put brass between copper and steal.

expansion and contraction rates of different material comes more into play.

Open systems like domestic water corrode real fast.

Mitch S.

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