[Sneap] De-ionized water

Thu Jan 10 21:32:41 EST 2008

Alan,

We have seven closed-loop chillers, but not all of them require non-conductive water. The ones that can tolerate a little conductivity are routinely filled either with bottled drinking water (essentially tap water) or city tap water, which is quite conductive. The ones that require non-conductive water are flushed as needed (when the conductivity increases to the point where the systems they cool quit working) and re-filled with distilled water from the same supplier as the bottled drinking water. We have a hand-held water conductivity tester, but no one uses it as a preventative maintenance tool. It just verifies the problem when something fails. I'm going to add it to my list of "things to do in my 'spare' time."

I assume our non-conductive water loops become conductive because the distilled water is attacking the copper, brass, and aluminum fittings. I like Nathan Jones solution of using a sacrificial anode. These are commonly available as replacement parts for hot water heaters. Most home owners are unaware that their hot water heater has a sacrificial anode or two, so when the anode eventually wears out in ten or fifteen or twenty years the water heater develops a leak and has to be replaced.

Howard

****************************************************** 
Howard B. Evans, Jr. 
Engineer, Materials Laboratory 
UES, Inc. 
4401 Dayton-Xenia Road 
Dayton OH  45432-1894 

937-426-6900 ext. 116 (office and voice mail) or ext. 121 (lab) 
937-426-5718 fax 

hevans at ues.com e-mail 
http://www.ues.com <http://www.ues.com/>  web site 

"Things should be explained as simple as 
possible, but not simpler." -- A. Einstein 

******************************************************* 

-----Original Message-----
From: sneap-bounces at tunl.duke.edu [mailto:sneap-bounces at tunl.duke.edu]On Behalf Of kerna
Sent: Thursday, January 10, 2008 4:33 PM
To: sneap at tunl.duke.edu
Subject: [Sneap] De-ionized water

Dear SNEAPer's,

A few years ago, after being in service for almost 30 years, a heat exchanger in our closed loop chilled water system sprung a leak.  Because the system was woefully energy inefficient, the entire system was replaced (It was calculated that the energy savings over 5 years would pay for the system replacement costs which was $125,000).  We were instructed by the project engineer to add a water treatment solution to the system.

Last year a problem developed in our magnet power supplies.  Some of the brass fittings (where the rubber hoses are attached that route the water from heat sink to heat sink) began leaking.  After some investigation, we discovered that the water treatment solution, which contained Sodium nitrate, was making the water very conductive.  This conductivity was causing oxidation and reduction to take place between heat sinks in the power supply where there was a voltage difference.  The fittings were being etched away wherever current flowed through the water.  After this discovery, we flushed out all of the water in the system, replacing it with water from our municipal water source.  Fortunately the damage to the power supplies could be repaired before this etching phenomena ruined them.  

At the last SNEAP Symposium I was talking to Nathan Jones about chilled water in the equipment that he works with.  He told me that they used de-ionized water in their equipment.  I found an article on the internet that talked about advantages and disadvantages regarding using de-ionized water.  Here is an excerpt from that article that addresses disadvantages:

"The disadvantage of de-ionized water is that it is very corrosive to metal. Since it has no dissolved solids in it, water will seek equilibrium with whatever it contacts.  So water with a pH of 7.0 can dissolve metal pipe.  Especially yellow metals like copper and brass.  It is also very aggressive to mild steel or "black" iron, and forget about galvanized pipe.  Piping is best as PVC or glass."

I ask Nathan Jones to review and comment about the article.  His comments were:

"We have a simple go-nogo cell in our system that lights a neon glow lamp as long as the conductivity is less than 1micro mho / cm. Our system also contains a sacrificial anode of zinc rod to prevent galvanic corrosion due to dissimilar metals (the zinc corrodes instead). The only time that I have heard of one of the sacrificial anodes being attacked was when a customer let their deionizer run for three years without replacement (supposed to be replaced annually) and then over-temped the system resulting in the deionizer dumping all that it had collected back into the system at once. They had a zero ohm short on everything and gray slime circulating in the system. The anode was 50% consumed in just a few hours. Luckily for them they were able to flush the system, install a new deionizer, and return to operation in just a few days with no damage. I would have volunteered to go help them recover - it was at Queen's Hospital in Honolulu!"

Additionally, I talked to several local companies that are in the water treatment business.  They hold the same view with regard to the corrosiveness nature of de-ionized water

My question to the SNEAP community is:  What do you do regarding maintenance of the water in your closed loop chilled water systems.

Thank you taking the time to read this very lengthy email.

Allan Kern

Western Michigan University, Kalamazoo, MI

269-387-4958

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://daytona.tunl.duke.edu/pipermail/sneap/attachments/20080110/e858b39e/attachment.html 