|
Water Treatment Non‑chemical Device
Device
System Data
History In an effort to reduce blowdown and address the scaling issues, the chemical program was discontinued during February, 2002, and four (4) NCD were installed, one in the sump of each cell in the cooling tower. Shortly after installation, operators found scale deposition in the conductivity probe, strainer, and return line. A fifth NCD was installed in the return line to address this problem. Blowdown is automatic using the conductivity controller, setting at about 3000 mmhos, resulting in 1000 gallons average discharge per day. Operators report that system conditions are better now, with the NCD, than when the chemical treatment program was in use.
Claims Made
Review Steve Morton, an experienced water treatment expert (30+ years), visited the site on October 15, 2002, to examine the cooling tower and obtain water samples. During the visit, plant personnel reported that the system was running better with the NCD in place than when the chemical program was used. Visual inspection of the cooling towers showed that there was a moderate amount of gritty calcium based scale on the drift eliminators. Shutting down one cell of the cooling tower and close examination of the tube bundles revealed about 1/16" of hard scale buildup. This type of scale buildup would not be evident on casual examination. No white rust was noted. Samples of makeup and cooling water were obtained for laboratory analysis by Analytical Services, Inc., Brockway, PA, a state certified laboratory, #33‑411, with the following results reported: Analytical Results
A Fluidtron Systems control box was examined and found to contain a high voltage transformer and rectifier. Steve remarked that it was similar to many other such units he has examined in the past 30 years. IntegraTech literature indicates a power consumption of about 0.035 milliamps by the NCD.
Discussion
When cycled up, the water does not become scaling until about 3.7 cycles and chemical treatment is not required to prevent scale formation until this cycle value is exceeded. At 10.5 cycles, close to the 10.8 cycles at which the cooling tower is actually being operated, our calculations project an SI of 2.53, which indicates severe scaling potential. The low SI, 1.2, actually found in the cooling water is caused by precipitation of calcium and alkalinity from the cooling water, which also lowers the conductivity. This gives us a "Catch 22" situation, where precipitation lowers the conductivity, resulting in lowered blowdown, further concentrating calcium and alkalinity which then precipitate, lowering the conductivity and repeating the cycle. Thus, we have high cycles, low SI, and scale. On review, the most startling item in the laboratory analysis is the disappearance of calcium from the cooling water. Based on operation at 10.8 cycles, there should be 155.5 mg/l calcium present in the cooling water, the laboratory reports just 12.4 mg/l (analysis by atomic absorption). This, coupled with the substantial change in Ca:Mg ratio between the makeup and cooling water, and the observation of scale on the tube bundle, documents that scale formation is taking place within the cooling tower. While one would expect more scale to be present than reported by Steve given the substantial calcium deficiency, some chemical control of scaling is taking place due to the high level of phosphate in the makeup water. The 1.2 mg/l total phosphate in the makeup cycles up to 16.1 mg/l, a level higher than that carried in many cooling water chemical treatment programs. Additional chemical control of scale is also occurring via carbonate precipitation of calcium in the bulk water due to development of an alkaline pH via cycling. This is confirmed by the finding of 24 mg/l suspended solids in the cooling water as well as noting that the makeup alkalinity has not cycled up, a good indication of removal via bulk precipitation as calcium carbonate and bicarbonate. Alkaline calcium precipitation within the bulk water as a means to control scale is a well known industry water treatment practice and commonly used in many boilers as the "carbonate cycle". Typically, we see white rust at operating pH values above 8.2 su. The absence of visible white rust in this system can be explained by noting the low developed alkalinity and high scale potential. White rust requires a pH above 8.2 su, substantial alkalinity, and clean, no scale, galvanize surfaces to become a visible problem. The literature from the IntegraTech Associates WEB site on the technological basis of the NCD was reviewed. Basically, it was found to be a mix of pseudo‑scientific ramblings, none of which are consistent with accepted scientific principals. The key claim made is that the device establishes an "electric field" in the water being treated, thus "charging particles" so that subsequent precipitation behavior of calcium, and the water surface tension, is changed. A simple test, checking to see if any voltage differential exists in the water being treated, will show no voltage and thus the electric field claimed does not exist. It is a fact that it is impossible to establish an electric field, shown as a voltage differential, across a conductor, and cooling water is a good conductor. In addition, to create a "charged" particle, electrons must be added, or removed; since the NCD does not supply any current (electrons) to the water, it is obvious that it cannot be "charging" anything.
Conclusions
Reporter
|