di-electric unions good or bad? 3

[EnergyProfessional]

I've searched multiple threads in these forum and still don't know if i should use di-electric unions. I understand the issue of galvanic corrosion, and also understand that if I have an open or domestic water system I introduce new oxygen all the time (I know galvanic corrosion doesn't need that oxygen), and also that it depends on pH value. i know this has been discussed before, but there still is no conlcusion beyond that the theory says they are needed, and the fieldstaff says they are evil

but I'm more looking for advice on closed loop HVAC piping with copper/steel pipes. I know theory (and code) prescribe di-electric unions. but I talked to many contractors and also to our maintenance crew and all said:
[ul]
[li]they never saw a corroded pipe where copper/steel met (at least not to the point of failure beyond some light surface corrosion)[/li]
[li]they have seen many di-electric unions leak[/li]
[li]many devices with internal copper / steel transition don't have di-electric unions from the factory[/li]
[/ul]
Obviously this makes me think I would spend money on something that would make systems worse. I usually include a good air removal device, and we use annual chemical treatment. I wouldn't mind spending the money on di-electric unions, that is not the issue, but they seem to bring more problems than they promise to heal. They seem to be something someone sitting in a desk likes, but someone maintaining systems hates them.

Should I look for adding di-electric unions/ and if so, what brands would be good? I believe the issue with leaking di-electric unions also is related to the cheapness.

thanks for any insight.

 

[Coredordelasolas]

I would use the dilectric union my self for ease of use. The other option is use a brass fitting inbetween the copper and iron. Problem solved.

 

[MJCronin]

Bewasre of GALVANIZED steel mating up to copper or brass.

While your crew may have never witnessed the slow corrosion beween carbon steel and copper ( TIME is a factor in corrosion), the galvanic coupling between steel and copper if far, far weaker than the couple that exists between zinc and copper.

I have seen galvanized/copper joints degrade in a few weeks.....its a better battery !

My opinion only.....

 

[EnergyProfessional]

Coredordelasolas: i looked up some electric charts and it seems copper and brass are very similar, but they still have a similar electric potential to steel (and zink). So would that even help on the steel/zink side?

what are good brands for di-electric unions? I see there are 2 major types. For example Hart has them threaded with a seal. And many other manufacturers have threaded/soldered ends and have them in one piece. I would think the latter is less prone to leaks. which one is better? and what would be good brands?

 

[bimr]

Closed loop HVAC piping will not typically experience significant corrosion because the systems are not open to the atmosphere.

Di-electric unions are generic items (like pipe) that are sourced from a piping distributor and are not something that you shop around for. If di-electric unions are leaking, it means that they were not properly installed.

Corrosion at the connection between copper and steel is different than the general corrosion that you may see in a closed loop HVAC system.

I agree with MJCronin, installl di-electric union and be done with it.

 

[EnergyProfessional]

bimr: thanks for the response. I think I'm sold on the di-union.

The problem is that when I just specify di-electric union, I will get the cheapest crap and that is more likely to leak. Sure installation plays a role. But the initial quality makes a difference as well.
typically for valves etc. I specify a few brands that are good (based on what maintenance tell me etc.). The same way I'm looking for a few "good" brands for di-electric unions.

I guess I go with the manufacturers that make the vales and other items we like, like Watts etc.

 

[moltenmetal]

Your OP post was dead wrong on one key, important point: galvanic corrosion DOES need oxygen, or some other process to discard the electrons which are a by-product of corrosion.

Your system is closed loop with limited to no make-up water, so no problem with galvanic corrosion. Same with treated cooling water loaded with corrosion inhibitors and antioxidants- to the extent that the water is properly maintained.

Potable or fresh/river water? External water with high ionic strength? BIG PROBLEM!

My 80-90 yr old house's hydronic heating system has plenty of copper- and brass- to carbon steel joints and all of them are 100% fine. No oxygen and no food for the bugs of rust = no problem.

 

[EnergyProfessional]

Moltenmetal: your statement made me google more and it seems we don't need oxygen to corrode.
I'm sure not having oxygen, and having steel/copper which are not too far apart help. Also the chemicals we use keep pH up and the teflon tape and paste also may help a bit to be an insulator.

But in the end I may go over to the dark side and prescribe di-electric unions just to be safe.

 

[TomDOT]

Yes, you do need oxygen (or a reasonable substitute) to corrode.

 

[EnergyProfessional]

I found this explains it well. what we will end up with are H-ions and the oxygen from water. i guess since we de-aerate constantly and the chemicals keep the pH value neutral for a while, we kind of slow down the galvanic corrosion. But it still will happen. Does it happen quickly? Probably not when we have steel and copper, which explains wehy we don't see many problems in properly maintained systems.

Still better safe than sorry, even in closed systems.

 

[moltenmetal]

Sorry HerrKaLeun, you're dead wrong. You need two half reactions to make an electrochemical cell. You need something to accept the electrons from the oxidation of the metals to metal ions, i.e. Fe =>Fe+2 + 2 e-, which is an oxidation half reaction. You need either oxygen, or another oxidant, or microorganisms and something to feed them, to provide an electron accepting (reduction) reaction to close the cell, or else there will be no electron flow and hence no corrosion. If your voltage difference is large enough, you could in theory reduce H+ to H2, but in a closed loop system the pH will rapidly rise to shut down hydrogen evolution.

Galvanic unions are NECESSARY in open systems where there is a continuously refreshed source of either oxygen or food for MIC organisms (organic matter). Such systems include potable water, river water, external corrosion due to weather/seawater exposure etc. Galvanic unions are unnecessary in closed loop water systems unless you're concerned more about water purity than about corrosion of the metallic parts.

As you noted, the galvanic unions are also not without problems. And their intent is easy to defeat, say, by means of misplaced electrical grounding wires, poorly installed pipe supports etc.

 

[EnergyProfessional]

Moltenmetal: it looks like you know this pretty well. So the O2 from dissolved OH (with the obvious O) and H doesn't help corroding? OK, assume no corrosion on the inside of the pipe. but what about the outside where the 2 metals meet? sure we miss the water as electrolyte. but if it is chilled water we easily could have condensation and O2 from the air?

OM, in closed loop systems with steel-copper i install good de-aerators, and use annual chemical treatment and call it a day.
In open systems, or domestic water (or system with zinc etc.), I use di-electric unions.
does that sound like a good policy?

 

[PEDARRIN2]

I specify the following for dielectric couplings, which are different from dielectric unions.

Dielectric coupling with non-conductive polymer liner, Victaulic Style 47, Gruvlok “Di-Lok” and Lochinvar Corp. “V-Line” Dielectric fitting on services 200 degrees and less, and pressures less than 300 psi.

 

[TomDOT]

HerrKaLeun,

The exterior of the pipes can certainly galvanically corrode if you have moisture present. Once the "necessary" portions of our corrosion cell are satisfied (oxygen, metallic pathway) and ignoring for the moment confounding mechanisms such as crevice corrosion and surface passivation/depassivation, the major factors which influence speed of galvanic corrosion are:

1) Time-of-wetness. How often and how long the surface is wet.

2) Voltage difference between the metals (more difference = faster corrosion)

3) Salts, particularly chlorides and to a lesser extent sulfates. The more corrosion-accelerating salts are present, the faster the corrosion goes. Some salts inhibit corrosion - Pb3O4 (also known as "red lead" or lead tetroxide) is the classic example of a corrosion inhibitor. Different salts also absorb moisture (hydrate) at different relative humidities. The effect of salts on corrosion is pretty darn complex.

4) Anode to cathode ratio. Is it galvanized bolts on brass, or brass bolts on galvanized?