Water treatment question, aluminum heat exchangers

RPK

I'm dealing with a water-quality issue with a water-source heat pump loop in a building I service. The company I work for installed the heat pump loop, 2 boilers (Bosch GB312, 733mbh), and a closed-circuit cooling tower back in 2013. The majority of the piping is steel, but there is also copper, brass and bronze. The water quality was okay for 2 years (water quality reports look okay, water looked clear, no plugged strainers, etc.) In 2015, another contractor added some piping and additional heat pumps to the system, shortly thereafter we received a water quality report showing that the pH had risen above acceptable limits (it was at 10.4) and the inhibitor was gone. The chemical treatment contractor was supposed to add chemical as necessary to get the pH under control, but a couple months later the water was black and the strainers and heat pump heat exchangers were full of black sludge. The chemical treatment contractor recommended a "bleed and feed" to clear up the water and also installed a 1 micron bag filter (bypass filter feeder). We thought everything was okay at this point and hadn't been monitoring things all that closely, but last week we had black sludge causing heat pumps to lock out again. The strainers were clean, but backflushing the heating exchangers revealed that they were full of black stuff again. After backflushing, the water-side delta Ts were better, but still higher than normal, so I'm thinking the surfaces of the heat exchanger may be fouled with this stuff. The chemical treatment contractor says they change the bag filters every two weeks (and they are always full of black sludge). They also say that the pH keeps rising no matter how much inhibitor they add. They are using a molybdate inhibitor. The most recent testing included installation of metal coupons (aluminum, steel and copper) for corrosion analysis and the aluminum coupon showed a noticeable loss of mass over a 200 day period. The water looks clear, but the bag filters are consistently loaded with some kind of black material every two weeks, and dT on some of the heat pumps we tested suggests that heat exchange is being effected.

As a rule, we leave the makeup water valve closed on this system so we would know if large amounts of water were being added (and they're not).

Does anyone have any recommendations or thoughts to share on this? The chemical treatment contractor told us they're not sure what is going on.

Jamie Hall

Step 1 -- find out what is in that sludge. I'm going to go out on a limb and suspect that you may find quite a bit of aluminium compounds, quite possibly aluminium hydroxide. Also, check you pH and pH control; aluminium corrodes quickly at both low pH and high, and that 10.4 you mention is lethal, but not at more nearly neutral pH. Also check for chlorides; you don't want any. Another possibility is galvanic reaction with steel -- that will happily attack the aluminium, but not the steel.

hot_rod

Take a magnet to the black sludge. If it sticks you have magnetite forming as ferrous metals break down. O2 ingress is a common cause.

The sludge needs to be analyzed.

On critical component of using chemical treatment is the correct dosage. if you under treat it you end up with a more aggressive attach on the under protected metals.

Two opinions on fluid, one is demineralized or RO water watching ph, buffer if necessary.

Another is the constant use and maintenance of treatment chemicals, I call it a chemical romance, once you start it may be impossible to stop.

The German water standard VDI 2035 seems to lean towards adjusting the water before it goes into the system DI or RO for example, the use of chemicals only to fix other system problems.

Find the entire standard online, it is an interesting read, a must read for all dealing with hydronic systems.

The treatment contractor should know the warning signs and be able to adjust the chemical balance as needed.

I'd also look for any changes that could be allowing air (o2) into the system all of a sudden. Piping changes, additional air vents added, expansion tank located on the discharge side of a large circulator.

It is very possible to be drawing air into a system, fresh o2 and the air is expelled at a separator but the o2 corrodes the ferrous components. When you have magnetite formation, expansion tanks are one of the first failure points, the thin gauge metal can pin hole.

The Amtrol Engineering Handbook is another good read concerning o2 ingress cause and prevention. Posted by permission from Amtrol.

Canucker

@Jamie Hall is right, aluminum is destroyed quickly with either end of the pH scale. Case in point, the company I work for purchased aluminum bodied, diaphragm air pumps with an interior teflon coating for solution and slurry transfers. We used 1brand new pump to transfer a slurry of celite and water, later followed by a 1000L solution with a pH of less than 3. The celite slurry scoured the teflon coating down to the aluminum and it did not survive the 1000L transfer that took less than 15 minutes. There was a black sludge, similar to what you described, coming out of the leak points. Hard, expensive lesson for the engineering department to learn.

kcopp

You may want to bring in a fresh set of eyes for this one.... Dwight from Rhomar is very good w/ this sort of thing.
You have a short window of opportunity as this boiler may have already been compromised.

RPK

Thank you everyone for your responses.

I can't believe we hadn't thought of checking with a magnet. The metal coupon analysis suggests that steel corrosion is being inhibited pretty well, so we're thinking it's aluminum, but a magnet test would help rule O2/air issues. To the best of my knowledge, the system has only one air vent (on the air/dirt separator on the suction side of the pumps), and the expansion tank is properly located. I'm thinking it would definitely be worthwhile to triple check all this. It's amazing how often we overlook the basics when a problem starts to seem unusual.

Hot Rod: thank you for those resources, that Amtrol book looks great, all kinds of great information in there! Regarding chemical treatment and RO/DI water.. it's interesting to hear that the use of chemical inhibitors is not the norm on closed-loops everywhere. Of course I've seen plenty of smaller systems, mainly residential, that don't have regular chemical treatment or testing, but standard practice on systems we install is chemical cleaning, bleed and feed flush, testing to verify all the cleaner has been removed, and addition of a corrosion inhibitor (based on the type of metals in the system). We have always used city water to fill closed-loop hydronic systems -- which I understand can be problematic depending on hardness, chloride content, etc.

kcopp: I've never worked with Rhomar, but I have definitely seen the name (referenced specifically in a few boiler installation manuals). Do you think Dwight would be willing to advise us or can we send sample to Rhomar for analysis? Unfortunately, we don't have a chemical treatment contractor around here that we have complete faith in (maybe one, but they're too busy to get involved).

I'm curious to know more about how the molybdate inhibitor is supposed to work. As I understand it, the molybdate is supposed to control/buffer the pH and (I read somewhere) create a protective film on metal surfaces as well. My understanding of acid/base reactions is that they can be "quenched". So in an acidic solution, the free hydrogen ion that makes the solution acidic gets used up as the reaction proceeds. It sounds like this may not be the case here. They add chemical to get the pH under control only to return and find it too high again.

hot_rod

Yes that is probably what is going on, the inhibitors and treatments are trying to correct the fill water issues.

Even before the use of quality water or chemical additives you want to assure a clean system.

The products used to assemble hydronic systems need to be flushed out, it usually takes a good cleaner to do the job. Most solder fluxes contain chlorides, so you want to flush all that junk out.

You have oils, flux, assembly lubes, loctite, pipe dope, all sorts of stuff that reacts with the metals.

Dwight retired from Rhomar, Daniel is the chemist that can help with water sample analysis. They have a menu of testing options at their website, www.rhomarwater.com

hot_rod

Here is some of the info at the Rhomar site.

Canucker

RPK said:

Thank you everyone for your responses.

I can't believe we hadn't thought of checking with a magnet. The metal coupon analysis suggests that steel corrosion is being inhibited pretty well, so we're thinking it's aluminum, but a magnet test would help rule O2/air issues. To the best of my knowledge, the system has only one air vent (on the air/dirt separator on the suction side of the pumps), and the expansion tank is properly located. I'm thinking it would definitely be worthwhile to triple check all this. It's amazing how often we overlook the basics when a problem starts to seem unusual.

Hot Rod: thank you for those resources, that Amtrol book looks great, all kinds of great information in there! Regarding chemical treatment and RO/DI water.. it's interesting to hear that the use of chemical inhibitors is not the norm on closed-loops everywhere. Of course I've seen plenty of smaller systems, mainly residential, that don't have regular chemical treatment or testing, but standard practice on systems we install is chemical cleaning, bleed and feed flush, testing to verify all the cleaner has been removed, and addition of a corrosion inhibitor (based on the type of metals in the system). We have always used city water to fill closed-loop hydronic systems -- which I understand can be problematic depending on hardness, chloride content, etc.

kcopp: I've never worked with Rhomar, but I have definitely seen the name (referenced specifically in a few boiler installation manuals). Do you think Dwight would be willing to advise us or can we send sample to Rhomar for analysis? Unfortunately, we don't have a chemical treatment contractor around here that we have complete faith in (maybe one, but they're too busy to get involved).

I'm curious to know more about how the molybdate inhibitor is supposed to work. As I understand it, the molybdate is supposed to control/buffer the pH and (I read somewhere) create a protective film on metal surfaces as well. My understanding of acid/base reactions is that they can be "quenched". So in an acidic solution, the free hydrogen ion that makes the solution acidic gets used up as the reaction proceeds. It sounds like this may not be the case here. They add chemical to get the pH under control only to return and find it too high again.

Ahh, the joys of pH adjustment. You'll find that most solutions that read the same on a pH meter, 3, for example, may require very different amounts of neutralizer to return them to a neutral pH.One could be adjusted with very little and another might require much, much more. The ones that require a lot to adjust might start to drop precipitate out of the solution during adjustment, as well.