Coffee with Caleffi: Piping Connection Options in Hydronic Systems

Training and education manager leads technical webinar for plumbing contractors, designers and wholesalers.
Coffee with Caleffi: Piping Connection Options in Hydronic Systems

Interested in Residential Plumbing?

Get Residential Plumbing articles, news and videos right in your inbox! Sign up now.

Residential Plumbing + Get Alerts

Bob Rohr, training and education manager with Caleffi Hydronic Solutions, leads an online webinar, “Piping Connection Options in Hydronic Systems,” from noon to 1 p.m. CT on Thursday, April 21.

There is no cost for the complimentary education series, Coffee with Caleffi, but registration is required.

Topics include:

  • Methods available for connecting copper tube to Caleffi components
  • Special precautions for soldering low-lead brass
  • The advantages of press-style fittings
  • BSP thread
  • How NPT threads differ from BSP and compatibility
  • Tips for assembling threaded connections
  • Tips for using green fiber gaskets
  • Concerns for solar thermal connections, high temperature operation
  • How PEX fittings differ from brand to brand

The following is a question-and-answer session from March’s webinar, “Water Quality in Hydronic Systems,” led by John Siegenthaler (pictured), mechanical engineer with Caleffi Hydronic Solutions:

Q: When it is necessary to join two dissimilar metals such that the potential for galvanic corrosion is present, I have heard that it is advantageous to use dielectric unions. However, I have also heard that these unions are a source of failure, and that the problem is bad enough that some plumbers refuse to use them (I realize, of course, that code requirements may force their use). What is your experience?


A: My experience is that dielectric unions that have been in place for several years on heating systems often fail when they are reopened for service. The plastic collar used to separate the two metals in the union seem to become brittle with age, and the stress of opening the union with wrenches causes it to crack. The only solution when this happens is to put in a new union.

When rigid joining “nobility diverse” metals such as copper to ferrous metal, I use a brass transition fitting, this keeps the metals of the two connection points fairly similar in nobility. But this is largely insurance because if TDS is less than or equal to 30 ppm, the potential for galvanic corrosion is mitigated.

Q: If the system is "all" copper, do we still get oxygen influx?

A: Yes, O2 molecules can enter systems in a number of places. Seals on valves and pumps, and pump flange gaskets are a few common areas.

BTW, the Germans learned of O2 ingress through plastic pipes and developed a DIN standard 4726 back in 1987 to define and deal with the concerns.

Q: Are the terms demineralized and deionized synonymous?

A: Yes. Three methods to produce deionized water are reverse osmosis, distillation and ion exchange. They all yield deionized water. RO and distillation are relatively expensive and yield ultra-pure water containing essentially no trace of materials in solution. Beverage manufacturers, for example, will employ these. The Caleffi HYDROFILL is an ion exchange method using a “mixed bed” media that is much more cost effective and produces a grade of water ideally suited to hydronic systems. It removes minerals in solution. This includes the salts calcium bicarbonate and magnesium bicarbonate, ingredients for “hard water.” Unlike RO or distillation, the ion exchange process does not remove, say, iron in solution.

Q: When using propylene glycol that has a Dipotassium Phosphate inhibitor (Univar- Vanfrost), do you need additional inhibitors?

A: Hydronic glycols come with inhibitors blended in. They should be checked every few years. A pH tester will tell you the condition of the fluid, or send a sample to the manufacturer.

Q: Can the propylene glycol mix be run through the Caleffi HYDROFILL?

A: We do not recommend running glycol through a HYDROFILL. I have been to glycol recyclers and they use a four-step process, which included demineralized water blending at the last step.

Q: Instead of using de-ionized water for filling a hydronic system with Propylene glycol, could reverse osmosis water be used?

A: It can but RO water is very expensive. See above answer. Also RO water is slow to produce and wastes maybe 3 gallons per gallon produced. However, if the water is known to contain high amounts of non-minerals in solution such as iron or even say phosphates or nitrates, it should be considered. Check the spec of the generator (boiler, chiller or heat pump) – they often drive the fluid requirement.

Q: Does it (HYDROFILL, the de-ionizer system) take out iron if the on-site water happens to be high in iron?

If the water on site is already softened, would you put softened water through the de-ionizer, or look for a raw water source

If the de-ionizer is drained (i.e. open the bottom tap) will freezing temps hurt it, or damage the resin? (Thinking about transporting the unit to and from site in the winter)

A: No, the HYDROFILL does not remove dissolved iron. When dissolved iron enters the system and is exposed to oxygen, iron oxide will form. A magnetic dirt separator would capture this just as it does the iron oxide formed from the ferrous components like black pipe or pump casings. But if you have known high levels of iron, it’s best to test and treat it. In my case, I use the Hellenbrand Iron Curtain system on my house before it runs through the water softener, for example. I believe it is designed to remove iron bacteria as well. You don’t want iron bacteria growing in your resin – nor your system.

Water that is first run through a water softener is OK to use. (Assuming the water does not have other problems like bacteria). Using softened water will prevent scaling (changes the

chemical structure of the calcium and magnesium), but does not lessen condition for galvanic corrosion. The Hydrofill removes all dissolved minerals (the ingredient for water to be conductive) thus scaling and galvanic corrosion are both addressed.

By design, the resin beads come moist in the buckets they ship in. They stay moist in the units as well and freezing will not damage them. But you may need to thaw the beads before adding them to the tank, or using the HYDROFILL. I would suggest taking care to prevent the unit from freezing, if at all possible – even when drained.

Q: Very interesting, this is a subject I have a lot of interest in, but I have a question: Why do you suggest adding inhibitors to stabilize? If you have demineralized the water and if the pH is correct, water will create a natural layer of protection on the metals, no need for false ones. If the conductivity of the water is low, then oxygen ingress can be handled as low conductivity means that corrosion is unlikely to occur, plus adding inhibitors actually increase the conductivity of the water, so you have undone some of the good work by demineralizing and reducing the conductivity. I think we need to be moving away from adding chemicals to heating water.

A: Thanks for the excellent questions from the UK! Here goes:

Site water that is run through any deionizing process will result in a lowered pH regardless of the incoming pH. Distillation and RO will produce the lowest pH (ultrapure) and in addition to being expensive as noted above, is one of the reasons it is not typically recommended. The HYDROFILL treated water pH will typically measure in the 6 range regardless of the incoming pH level. However, over time, the pH of the water will naturally rise and stabilize (this process is referred to as self-alkalizing). The resulting pH typically settles around 8 or a bit more. This natural process can take days or weeks to occur depending on the system, component materials and temperature of the water. Heating the water will accelerate the conversion of the carbonic acid to carbon dioxide, which will with its temperature elevated, come out of solution and if coalescing air separators are used like ours, the gas will be automatically vented out. We recommend taking a pH measurement shortly after the system is fired and the running system a bit. Even knowing that it will continue rising to about 8 naturally, if desired to force adjust the pH up or down, use one of the pH adjuster additives available. But often times a pH adjuster is not needed, depending on the pH spec of the boiler or chiller – the two pieces of equipment that usually dictate system pH because they are the most expensive.

OK, pH is within spec, so now back to your question. The demineralized water will prevent scaling and because the conductivity is low, galvanic corrosion is prevented too. But what about oxidation corrosion? Yes, it is true that a ferrous film will form (but at pH 11 or above it won’t), but if there is an ongoing source of ingress, oxidation corrosion of the iron or steel will still continue with one result being the formation of magnetite that is usually microscopic and can cause havoc with ECM circulators, seals, and heat exchangers (check out our archived webinar “Magnetite Separation in Commercial Systems”). A good coalescing air separator will minimize any oxidation, and for the magnetite, a good magnetic dirt separator will capture that too. So to your point, in the end, installing high performing components will largely minimize (or in many cases even eliminate) the need for inhibitor additives. But we recommend them as another safeguard for designers to consider adding. In some jurisdictions though here in NA, adding any chemicals will require a more expensive backflow preventer to be used (RP type) in many residential applications.

BTW, the systems installed in the U.S. and Canada often include multiple metals. We see copper, steel, iron, stainless steel and multiple aluminum alloys, and various plastics including some without 100 percent O2 barriers. Boiler manufacturers here are really clamping down on warranties if the failure was deemed to be fluid related. There is not as much knowledge about water quality here compared to Europe. Thus, I like that hydronic conditioners can provide multiple functions such as oxygen scavengers, pH buffers, film providers, and will even lock up some hardness.

Perhaps most important is the system is cleaned well with a hydronic cleaner. We see oils, assembly lubes, pipe dope, solder flux residue, dirt, and other jobsite debris in new and old systems. No doubt a clean system with conditioned water is much better than systems we currently see: purged and filled with jobsite water, and that is it!

Q: What can you do to treat a DHW system with multiple SS heat exchange failures due to extremely hard water?

A: Not knowing your peculiars I offer:

Best: Demineralize the water and add an inhibitor. (Pre-blended glycols will suffice but this is an expensive way if you don’t need freeze protection).

Next best: Soften the water and add an inhibitor. Softening changes the ion structures of the minerals and prevents them from coming out of solution and scaling up but does not lessen galvanic corrosion. However, SS is noble and not susceptible to galvanic attack.

Better than nothing: Add inhibitors … but realize, without a zero baseline that demineralized water delivers, the proper amount of inhibitor dosing is not easy for the untrained.

Maybe the manufacturer of the HX has published a water quality spec? From what I read, chlorides are not friendly to SS. A major SS boiler manufacturer, for example, specifies less than 25 ppm chlorides required.

Q: Would it be OK to use cheapie, supermarket distilled water if your TDS meter indicated it had the required TDS range?

A: Not knowing how that water is processed, it would be tough to answer. I suspect it is run through a DI tower. You could buy some and test the TDS and pH. I imagine it’s better quality than most job site water. I know much of the water labeled distilled is just filtered water; my local water supplier showed me the bottles and the process they use to fill them!

Q: Would using soft water prolong the use of the resin beads in a HYDROFILL?

A: Prolong? Most likely no. The actual mineral content of softened water does not change from the source water – only the chemical structure of the salts (calcium and magnesium) change. See question above.

Q: Are the resin beads used in HYDROFILL irreversibly consumed or can they be restored?

A: The resin beads do not wear out, but it would take a strong caustic solution to clean them. In larger cities, you may find a water condition shop that is licensed to perform that cleaning.

The Caleffi bags are easy to replace to avoid this, and are recyclable.

Q: What would be recommended for an older schedule 40 steel heating system that is only flushed every 3 years, no chemical treatment? Will adding treatment create more problems?

A: If the system isn’t giving you any problems, I would check the fluid pH and TDS, then add a good hydronic conditioner. With no chemical treatment, probably the less you change the water, the better at this point. Every time you add freshwater you may be adding more minerals and certainly fresh O2.

Q: What about using RO water instead of demineralized water?

A: Yes, RO is another way to treat water and a viable consideration if the water has problematic dissolved matter besides minerals. The initial fill water pH is likely to be very low though and thus aggressive (acidic). So adjusting the pH with a buffer rather than wait through the self-alkalizing period might be good. Also, generally it takes about 3 gallons of water to generate 1 gallon of RO water as it is squeezed through a semi permeable membrane. It is a slow process also. For large quantities of hydronic fill water it may not be economical.

Q: What about large commercial systems such as apartments and condos?

A: No different than residential systems. It really comes down to the quality of the water on-site. There are many areas where city water is acceptable for hydronic system out of the tap – especially if the source water is largely comprised from runoff (like a river) versus ground sourced. Really, it only takes a few inexpensive tests to determine if your water meets the spec of the boiler and component manufacturers.

Q: My radiant floor (PEX) was filled with tap water. Not long afterwards, the water turned grey. There is iron pipe in the system. What might be the cause of this and should I pursue flushing/filtering and installing the recommended refilling system with a protective film? Any long-term issues by doing nothing?

A: It is not uncommon for the water to take on a bit of grey color from the ferrous components in the system. This is typically magnetite (a form of iron oxide). Worth checking though if the system was cleaned and flushed when first filled. Running a good hydronic cleaner in the system first is a good practice. Take a sample of the fluid. Check the pH and TDS if you can. Compare that to what the boiler manufacturer recommends. Keep in mind, if any chemicals were added, that will change the testing and readings. The supplier of the chemicals would be able to give you testing information.

Back to that grey. If it starts turning dark grey or black, magnetite is forming excessively. A magnetic dirt separator (Dirtmag) is meant to address this (and prevent it from incurring to begin with).

Here is a job site that had such a problem – check out the results of the Dirtmag purge in this short video:

Q: Is the water going through the cartridge system potable?

A: I would be concerned about water sitting inside the HYDROFILL for long periods and becoming stagnant. It may depend on the source of the water you are running into the HYDROFILL unit. Is it treated city water, for example, or well water from a private untreated source? The resin beads pull out the hardness and other salts in solution. Microbes and solids such as silica are not.

Q: If you are using an antifreeze, is demineralization still recommended prior to adding the antifreeze?

A: If you are buying pre-blended glycol, those are mixed at the plant with DI water. If you buy glycol in full strength and mix it at the jobsite, yes – you should always use DI water and the Hydrofill is good for that purpose. One of the inhibitor chemicals in premixed glycol will lock up some hardness, but starting with top quality water is critical. If you use hard water in mixing with glycol, the inhibitors will precipitate out and render the anti-freeze properties ineffective. Also a good clean and flush first to get rid of all the oils, flux residue, etc. is important.

Q: What is the only good thing about anti-freeze?

A: Properly mixed, the fluid doesn’t freeze! And as previously mentioned, most have inhibitors. Beyond that, it does bring along some penalties and challenges, regarding heat transfer and maintenance.

Q: How would you add glycol to the system with a HYDROFILL?

A: Possibly the easiest way would be to pre-blend the glycol with the HYDROFILL water to the proper percentage.

Unless you know the actual system volume, it would be tough to fill with water then add the appropriate amount of glycol. As you fill the glycol you would need to constantly check the percentage after it circulated. You would end up dumping some fluid down the drain maybe, until you acquired the proper percentage.

It may be easiest to just buy the pre-mixed glycol. Clean and flush the system well. Blow out the fluid, and add the pre-mix.

If you are filling the system for the first time, you could put a water meter on the fill. You would need to account for any water you purge out.

Q: Approximately how many gallons of water can go through the HYDROFILL before the beads need to be replaced?

A: The bead replacement depends on how hard the water that is going into the unit. The HYDROFILL does have a TDS meter on top. Keep an eye on that as you are filling. In absence of a spec from the system equipment manufacturers, we recommend changing out at 30 ppm. This value is within the vast majority of boiler and chiller specifications.

This table lists the capacity of the Hydrofill.

Q: Apart from galvanic corrosion, is there another reason to use demineralized water? i.e. Hydronic system that is completely plastic piping with a few brass adapters and stainless steel heat exchanger.

A: A primary benefit of DI water is the elimination of scale formation. Regardless of the piping materials, hard water scale can coat out the heat exchangers, boilers, chiller HX, pumps, control and balancing valves, really all the components. Starting out with a top quality fluid assures you optimum conditions inside the system.

Q: After the addition of fluid protectant, the TDS rises to about 400 ppm within the system. Is this a new basis point for checking the system in the future? Or is the chemical analysis of the water better?

A: You are correct. Once you add any conditioner or glycol to the system the ppm will increase and you need to test that fluid according to the manufacturer’s direction. I know Rhomar has a small test kit for their conditioners. I’ll bet other brands do also. I see the Rhomar conditioner kits include some test strips. Rhomar also will accept fluid samples for a more complete analysis. They have a full-time chemist at their Springfield, Missouri, location. With glycols, a basic pH and refractometer are a good way to keep an eye on the fluid. DOW also offers fluid analysis, as I expect other brands of glycol do.

Last month’s webinar, “Water Quality in Hydronic Systems” can be viewed below:


Comments on this site are submitted by users and are not endorsed by nor do they reflect the views or opinions of COLE Publishing, Inc. Comments are moderated before being posted.