CTI Journal, Vol. 30, No. 2 · 4 CTI Journal, Vol. 30, No. 2 View From The Tower Denny Shea Welcome...

CTI Journal, Vol. 30, No. 2 1

The CTI Journal(ISSN: 0273-3250)

PUBLISHED SEMI-ANNUALLYCopyright 2009 by The CoolingTechnology Institute, PO Box 73383,Houston, TX 77273. Periodicalspostage paid at Houston, Texas.

MISSION STATEMENTIt is CTI’s objective to: 1) Maintain andexpand a broad base membership ofindividuals and organizationsinterested in Evaporative HeatTransfer Systems (EHTS), 2) Identifyand address emerging and evolvingissues concerning EHTS, 3) Encour-age and support educationalprograms in various formats toenhance the capabilities andcompetence of the industry to realizethe maximum benefit of EHTS, 4)Encourge and support cooperativeresearch to improve EHTS Technologyand efficiency for the long-termbenefit of the environment, 5) Assureacceptable minimum quality levelsand performance of EHTS and theircomponents by establishing standardspecifications, guidelines, andcertification programs, 6) Establishstandard testing and performanceanalysis systems and prcedures forEHTS, 7) Communicate with andinfluence governmental entitiesregarding the environmentallyresponsible technologies, benefits,and issues associated with EHTS, and8) Encourage and support forums andmethods for exchanging technicalinformation on EHTS.

LETTERS/MANUSCRIPTSLetters to the editor and manuscriptsfor publication should be sent to: TheCooling Technology Institute, PO Box73383, Houston, TX 77273.

SUBSCRIPTIONSThe CTI Journal is published inJanuary and June. Complimentarysubscriptions mailed to individuals inthe USA. Library subscriptions $20/yr.Subscriptions mailed to individualsoutside the USA are $30/yr.

CHANGE OF ADDRESSRequest must be received atsubscription office eight weeks beforeeffective date. Send both old and newaddresses for the change. You mayfax your change to 281.537.1721 oremail: [email protected].

PUBLICATION DISCLAIMERCTI has compiled this publicationwith care, but CTI has not Investi-gated, and CTI expressly disclaimsany duty to investigate, any product,service process, procedure, design,or the like that may be describedherein. The appearance of anytechnical data, editorial material, oradvertisement in this publicationdoes not constitute endorsement,warranty, or guarantee by CTI of anyproduct, service process, procedure,design, or the like. CTI does notwarranty that the information in thispublication is free of errors, and CTIdoes not necessarily agree with anystatement or opinion in thispublication. The entire risk of the useof any information in this publicationis assumed by the user. Copyright2009 by the CTI Journal. All rightsreserved.

ContentsFeature Articles8 Progressing the Frontier of Cooling Water Process

ControlKevin Milici and Gary Geiger

20 Intermittent Feeding of Aseptrol® Tablets Redefines theRole of Chlorine Dioxide in Small and Mid-sized Cool-ing Water SystemsKeith Hirsch, John Byrne and Barry Speronello

28 Cooling Tower Basin Evaluation and RepairThomas R. Kline

38 Recent Developments In Motor Technology AllowDirect Drive Of Low Speed Cooling Tower FansRobbie Mcelveen, Bill Martin and Ryan Smith

48 Inspection of Pultruded Cooling Tower ComponentsDustin Troutman and Jess Seawell

Special Sections58 CTI Licensed Testing Agencies

62 CTI Certified Towers

68 CTI ToolKit

Departments02 Meeting Calendar

02 Multi Agency Press Release

04 View From the Tower

06 Editor’s Corner

...see page 16...see page 30

...see page 43

CTI Journal, Vol. 30, No. 22

CTI JournalThe Official Publication of The Cooling Technology Institute

Vol. 30 No.2 Summer 2009

Journal CommitteePaul Lindahl, Editor-in-ChiefArt Brunn, Sr. EditorVirginia Manser, Managing Editor/Adv. ManagerDonna Jones, Administrative AssistantGraphics by Sarita Graphics

Board of DirectorsDennis P. Shea, PresidentJess Seawell, Vice PresidentFrank L. Michell, SecretaryRandy White, TreasurerJon Bickford, DirectorHelen Cerra, DirectorTim Facius, DirectorGary Geiger, DirectorChris Lazenby, DirectorKen Mortensen, DirectorAddress all communications to:Virginia A. Manser, CTI AdministratorCooling Technology InstitutePO Box 73383Houston, Texas 77273281.583.4087281.537.1721 (Fax)

Internet Address:http://www.cti.org

E-mail:[email protected]

REDWOOD

DOUGLAS FIR24 Hour Service on Your Lumber and Plywood Requirements

Redwood Lumber • Douglas Fir Lumber & Plywood • • • • • Fiberglass DeckingFiberglass Structural Components ••••• Corrugated Fiberglass Panels ••••• 24 Hour Service

GAIENNIE LUMBERCOMPANY

BOX 1240 • OPELOUSAS, LA 70571-1240800-326-4050 • 337-948-3067 • 337-948-3069 (FAX)

Member

FUTURE MEETING DATESCommittee Workshop Annual Conference

July 12-15, 2009 February 7-11, 2010Marriott Hotel The Westin Galleria

Colorado Springs, CO Houston, TX

July 11-15, 2010 February 6-10, 2011Marriott Albuquerque Pyramid N Westin Riverwalk

Albuquerque, NM San Antonio, TX

July 17-20, 2011 February 5-9, 2012Amelia Island Plantation Hilton Hotel

Amelia Island, FL Houston, TX

For Immediate ReleaseContact: Chairman, CTI Multi-Agency Testing Committee

Houston, Texas, 1-May-2009

The Cooling Technology Institute announces its annual invitationfor interested drift testing agencies to apply for potential Licensingas CTI Drift Testing Agencies. CTI provides an independent thirdparty drift testing program to service the industry. Interested agen-cies are required to declare their interest by July 1, 2009, at the CTIaddress listed.


View From The Tower

Denny Shea

Welcome the CTI Summer Journal. I knowthat you will find the articles in this issue bothinteresting and informative. I learned very earlyin life from my father that keeping referencearticles was a practice that would be invalu-able to me in my career. His advice was verysimple “You can’t remember everything sodon’t try, just remember where it is writtendown.” His words have proven to be very soundover the years. He retired long before theinternet. We do not have to tear pages fromjournals and put them into binders for refer-ence anymore. We have the internet and all the resourcesit brings. The CTI website is one such resource. It is avail-able to anyone with access to a computer and the internet.The website is a door to vast amounts of information. Itlinks everyone to CTI publications, technical papers, a di-

rectory of manufacturers, suppliers, and owner/operators, and updated news on CTI activities.The CTI “Ask the Expert” provides contact toexperts in the fields of Water Treatment, Me-chanical Equipment and Cooling Tower Perfor-mance.

The resources of CTI are wide and vast. All some-one needs to do is start-up a computer and re-sources become a mere key stroke away. A re-source that CTI holds, which is not accessibleon-line, is meetings. Meetings invoke the sharingof information between CTI members on a per-

sonal level. This week a friend asked me what the differencewas between a member and a participant. The answer issimple. A member is someone who pays dues. A participantis a member that takes an active role in the organization. Iunderstand that distance, money constraints, and time are all

valid reasons why members do not activelyparticipate in CTI committees. However, welive in the electronic age and CTI is expand-ing to meet the challenge that age. Our com-mittees use conference calling, on-line meet-ings and e-mail as avenues to reach mem-bers who want to participate but cannot al-ways attend meetings. If you wish to partici-pate, please contact a committee chair andexpress your interest. The committee chairsare always in need of assistance for commit-tees, reviewing papers, new ideas and justplain outside the box thinking. We welcomeall the help we can get.

Denny Shea

CTI President 2008-2009


Editor’s Corner

Paul LindahlEditor-In-Chief

Dear reader,

“May you live in interesting times”

Like many, I’ve had the impression that this isa translation of an ancient Chinese curse.Doing a little research, it turned out that it isprobably the creation of British diplomats inChina in the 1930’s or before, perhaps adaptedfrom vaguely similar Chinese proverbs. TheChinese historians don’t claim it.

The thought, however, seems to have enoughmeaning to people that the quote is used veryoften. (I got 41 million results on Google™).

Right now is a good example of that meaning.

Interesting times is a very ironic way of describing theconditions we all have experienced since the economicinstability hit home. We, who are lucky enough to live(and still work) in these times, will have some stories totell our grandchildren. Hopefully not like those I heardfrom my grandparents and parents. As I’m writing this,Chrysler and GM are in or near bankruptcy, and many

leading names from the financial industry are nowhistory. I just read that the government is ex-pected to hold about 75% of GM after its bank-ruptcy. That would be difficult even to imaginefrom not so long ago. We and our children havesome serious challenges ahead to understand anddeal with all of the changes that have happenedrecently.

A common theme heard from many of the econo-mists and would-be-economists that seem to beeverywhere right now, is that a critical driver ofrecovery is consumer confidence. Seems logi-

cal. At a basic level, we all need to do the best we can tocontinue to transact our personal and business lives normally,which will cumulatively increase that perception of confidence.More caution than normal will likely be the rule rather thanthe exception, but it has to start somewhere.

The businesses that touch our industry no doubt have enoughdiversity to have experienced the whole range of side effectsof the economic times. Nonetheless, we have many reasonsfor confidence in the future as a group. Careful consider-

ation and patience will serve us well, but we do needto move forward.

The Cooling Technology Institute has its own chal-lenges, requiring our careful consideration and supportto navigate toward a successful future. We look for-ward to working with many of you to meet the chal-lenges.

Best regards,

Paul Lindahl, Journal Editor


Kevin MiliciGary GeigerGE Water & Process Technologies

ABSTRACTFor the past several decades, industrial andnon-industrial users and operators of opencooling water systems have realized a steadystream of developments and innovations inthe automation and control of cooling watersystem chemistry. This paper discusses theperformance of a new innovation for the mea-surement and control of polymeric dispers-ants used in cooling water systems. Theadvance moves the industry towards the in-

The goals and require-ments are not neces-sarily new, but thechallenges in obtain-ing them are. Evertightening budgets, in-tolerance for failuresor unplanned forcedoutages, efficiency de-mands and relentlessdemands for improvedefficiencies and opti-mization, increased

Progressing the Frontier of CoolingWater Process Control

evitable desire and efficiency of the direct measurement of func-tional treatment chemistries for the control and optimization of scale,deposit and corrosion because of it’s implicit purity, simplicity andlogic. The technology takes advantage of new, patented reagentchemistry, senses spectral response to proprietary polymers atmultiple wavelengths, uses one-fifth the reagents of conventionalon-line wet chemistry monitors due to advanced solid state designand miniaturization, and has a built-in self-zeroing function thateliminates interference from background color and turbidity. Mea-surement technologies, while innovative and elegant, must be de-livered with simplicity, for the rugged industrial world. Both dy-namic laboratory data and field applications that demonstrate theefficacy of the technology are described.

Keywords: Automation, deposit control, corrosion control, stabi-lized phosphate, polymer, phosphate, cooling water, cooling tower,direct measurement, and stressed water condition.

INTRODUCTIONThe owners and operators of businesses that depend on opencooling water systems to enable their processes, have a set of com-mon goals that to a large extent, haven’t changed in decades. Theyinclude …

· Production: throughput and yield

· Asset protection; equipment integrity; meeting its planneduseful life

· Operating efficiency: optimizing resource consumption perunit of production

· Total Cost Out: removing unnecessary cost (labor, energyloss, etc.)

· Human Productivity: getting more or the same done withfewer or less resources

· Environmental Stewardship: moving beyond just the mini-mum of compliance

environmental reporting to name a few. In addition, deterioratingwater qualities and availability are forcing practices of water re-useand recycling to reduce the required water footprint creating oper-ating conditions that require the cooling water program to be al-ways be on it’s “A-game.”

While the economic, environmental and human health benefits ofeffectively managing heat transfer surfaces, corrosion and biologi-cal activities are clearly known, the challenges continue to mount.As a result, the advent of automated process control for coolingwater systems has progressed accordingly. It is these advancesthat help owners and operators navigate and find solutions to thenew challenges.

Cooling water treatment programs must control corrosion, deposi-tion and microbial activity to maintain heat transfer efficiency andavoid premature corrosion failures of process equipment. All threeareas of concern are interrelated and must be simultaneously ad-dressed. Excessive steel corrosion will result in flow restrictionsand the accumulation of iron corrosion products that impede heattransfer. Soluble iron released at the corroding surface can stresspolymeric dispersants, resulting in diminished efficacy and the for-mation of iron phosphate deposits. Deposits not only impede heattransfer, but also are responsible for premature corrosion failuresdue to under-deposit corrosion mechanisms. Deposits provide asafe haven for anaerobic bacteria that are responsible for microbio-logically influenced corrosion (MIC). Both sessile and planktonicbacteria control are necessary to prevent slime forming bacteriafrom establishing biofilms that restrict heat transfer and provide aprotective environment for the colonization of anaerobic bacteria.

Inorganic phosphate is the most widely used steel corrosion in-hibitor for open cooling systems. Orthophosphate suppressesboth the anodic and cathodic corrosion reactions. High dosagesof orthophosphate (12-20 ppm) catalyze the formation of a protec-tive oxide film, that retards the anodic corrosion reaction.1,2 Ortho-phosphate restricts electron transfer at the cathode through the

Gary Geiger Kevin Milici


formation of an insoluble calcium phosphate film. Orthophosphate-based corrosion programs may include the addition ofpolyphosphate (pyrophosphate or hexametaphosphate) and/or zincto enhance cathodic corrosion inhibition and guard against local-ized corrosion. The use of polyphosphate is favored where dis-charge restrictions limit or preclude the use of zinc. Like ortho-phosphate, polyphosphates form insoluble calcium salts at the cath-ode due to the localized high pH.

The use of inorganic phosphate for steel corrosion protection re-quires an effective calcium phosphate precipitation inhibitor tomaintain the phosphate soluble in the bulk cooling water and pre-vent deposition at heat transfer surfaces. Effective polymeric in-hibitors/dispersants for calcium phosphate were first developed inthe late 1970’s. Since that time, a wide variety of copolymers andterpolymers have been introduced that have expanded the rolefrom calcium phosphate inhibition to particulate fouling control.However, the primary role of the polymeric dispersant in an inor-ganic phosphate-based program is to prevent calcium phosphateformation. Calcium phosphate demonstrates retrograde solubilitywith both pH and temperature. At any given level of calcium hard-ness and phosphate, the dosage of the dispersant is dictated bythe temperature of the hottest process equipment and the operat-ing pH range, if scaling is to be avoided. Precise pH control isrequired to minimize high pH swings and avoid exceeding the con-trol capabilities of the inhibitor.5

The measurement, control, and verification of polymeric dispersantlevels in cooling systems has long been of importance to the indus-try. Analytical methods to date, both off-line and on-line, havelacked sensitivity and reliability, and often require the addition oftracing materials. 6

This paper discusses the performance of a new innovation for themeasurement and control of polymeric dispersants used in coolingwater systems. The advance moves the industry towards theinevitable desire and efficiency of direct measurement of functionaltreatment chemistries for the control and optimization of scale, de-posit and corrosion control. Validation of this new technology wasfirst demonstrated in dynamic laboratory systems. Subsequently,through a series of field evaluations, the performance was furtherverified.

TECHNOLOGY OVERVIEWDirect Polymer MeasurementFundamental to this technology advance is the ability to measurethe polymer concentration in cooling water, apart from any poly-mer which may be adsorbed onto suspended solids or other sur-faces. A focal point is patent pending reagents that provide supe-rior polymer detection capabilities. The spectral response to thenew reagent chemistry is sampled at multiple wavelengths to pro-vide superior accuracy and response. A self-zeroing function elimi-nates most sources of interference including variable backgroundturbidity and color, ensuring that the unit provides accurate andreliable “hands-off” response necessary to confidently control criti-cal cooling water applications.

The direct measurement of polymer concentration in cooling waterhas long been limited by the availability of reagent chemistries thatare sufficiently sensitive and specific to enable reliable control.Detection reagents have progressed from early turbidimetric meth-

ods in the early 1980’s, to the grafting of fluorescent functionalgroups onto the polymer8, to metachromatic dye methods based onpinacyanol chloride (PNC). 9,10 All of these approaches have beensuccessful to some degree, but lacked the degree of sensitivity andspecificity required for a sensor that would be used to control a keycomponent of the cooling water program. Turbidimetric methodsrely on the principle of precipitating the polymer from a filteredsample with a quaternary amine surfactant and measuring the in-crease in turbidity. Although turbidity sensors are inexpensive andreliable, the method requires a well-filtered sample and the responseis affected by the somewhat variable size of the particulates formedin solution. The PNC dye method in many respects represents animprovement over the turbidimetric method. However, PNC lacksspecificity to the polymer being detected, providing a strong re-sponse to background humic compounds and other charged or-ganic species. Grafting a non-functional tag onto the dispersantpolymer that would make it easier to detect was likewise rejectedbecause it would add to the cost and interfere with the efficacy ofthe polymer.. Fluorescent groups in particular are very sensitivefrom a detection standpoint, but are subject to interference frombackground fluorescence from both naturally occurring compoundsand organic contaminants (process leaks)..11

The initial focus of the project was to develop a new reagent chem-istry that was both highly specific in interacting with the functionalgroups and structure of the target polymer , and sensitive enoughto be detected at very low levels. Among the many specifications,the new chemistry was required to be accurate over the pH range of6.5-9.0, have a stable shelf life of at least 1 year, be effective overthe range of 32-110°F, and be non-hazardous. Fundamental re-search into the basic chemistry of cationic dyes in the presence ofa water soluble polymer identified a specific material that met all thedesired criteria. The result of this effort is a new patent pendingpolymer detection method that is far more specific and sensitivethan current methodologies.. The new reagent chemistry func-tions over a wide pH range and is essentially interference free up toa hardness of at least 2,000 ppm. A unique buffer solution contain-ing specialty additives to further reduce interferences from non-targeted polymeric substances and other anionic species was for-mulated to improve the robustness of the method. 12

MonitorThe monitoring instrument itself is shown in Figure 1. It was de-signed to operate trouble free and hands free in an industrial envi-ronment, with a simple and infrequent, and brief maintenance pro-cedure. The unit outputs a 4-20 mA linear analog signal that can beinterfaced with an appropriate controller compliment. Highlightsof the major sub-components appear in Figure 2 and are describedas follows:

Electronics Module: This module is the unit’s nerve center. Itcontrols all key functions of filtration, sample preparation, signalprocessing from the optical measurements polymer concentrationcalculation, simple single-point calibration, diagnostics and alarm-ing. In addition, a simple touchpad interface to enable fast naviga-tion through all functionalities.

On-Board Filtration Module: Historically, the Achilles heal of manyanalyzers, both “wet” and “non-wet,” is sample quality, and spe-cifically, fouling of the instrument due to suspended materials and/or biological growth. With that in mind, the technology develop-


ment had a major focus of ensuring reliable operations, in an essen-tially hands-off mode. The function of the filtration module is toremove particulate matter which might clog the unit, as well as toremove polymer which might be bound to suspended particles.The filtration module is designed to be continuously self-backwashing, using a pair of elements, one of which is always inbackwash mode using filtered water. This unique design allows theunit to function reliably even in very turbid cooling waters. Fil-tered water from this module progresses downstream en route tothe Detection Module.

Detection Module: The Detection Module includes the opticalcomponents that detect the reaction color change. It consists of anLED array, photodiode sensing array (PD), an optical detectionzone, and a LED/PD holder. The optical path geometry has beenspecifically designed to maximize fouling resistance. The LEDsand photodiodes are installed in channels mounted perpendicularto the central optical cell. The LEDs generate incident light atmultiple wavelengths, and the corresponding photodiodes detectthe respective transmittance on the opposite sides. The LEDs andphotodiodes cover the spectral range where the reagent colorchange occurs in proportional to the free polymer concentration.“Outliers” in the correlation of the photodiode array to polymerpulses are screened out by a filtering algorithm. The unit featuresan automated self-zeroing function between each reading in orderto cancel out the effects of varying sample turbidity, color, or foul-ing of the optical cell. Response is also compensated for tempera-ture variations. The compact size, sensitivity to multiple wave-lengths, and fouling-resistant optical cell design provide excellentsensitivity with low maintenance requirements and minimal regentconsumption.

System ConfigurationsSituations in which cooling system owners or operators have notalready made substantial investments in controllers, feed and dos-ing equipment, preconfigured plug and play systems are available,including complimentary array of other sensors and capabilities.

DEMONSTRATED PERFORMANCEDynamic Laboratory StudiesLaboratory evaluations were conducted using a dynamic simulatorprotocol. Shown in Figure 3, the simulator is a recirculating systemincorporating a heat transfer surface to assess deposition, and hasbeen described in prior papers.6 Dynamic study protocol is de-signed to simulate a severe system condition. Bulk water and heattransfer surface temperatures are controlled, pH is monitored andcontrolled, and corrosion rates of low carbon steel (LCS) and admi-ralty brass (ADM) are measured with test coupons. The device isnot an evaporative system, so it does not simulate an open recircu-lating system. The BTU simulates a heat exchanger that is con-stantly recirculating hot water (120OF) for cooling. The unit uses aheat exchanger tube fitted with an electrical heater to adjust thesurface temperature and a cooling coil to remove excess heat andmaintain the temperature of the recirculating water. The systemretention time (holding time) is controlled by adjusting the rate ofmakeup to the water reservoir. Samples are taken from the sumpdaily for water chemistry characterization including turbidity, ortho-PO

4 and ICP analysis of both filtered (F) and un-filtered (UF) samples.

At the end of each test run, corrosion coupons and heat exchanger

tube are visually inspected for deposits. Coupon corrosion ratesare determined by a weight loss method.

Figure 4 shows the results from a test run, whereby the actualmeasured polymer concentration is compared to the theoreticalpolymer concentration that was fed, using an established tracer ofmolybdate, that was formulated along with the polymer product.The two concentrations track consistently over the time frame. Thedirect polymer concentration is slightly less than the theoreticalfed, due to the minor stress placed on the polymer by the testconditions. Effectively, all the polymer is detected and measuredby the technology, and is consistent with what would be expectedto be found.

Figures 5-9 are a series of laboratory tests that demonstrate theperformance of the polymer monitor/controller when operated purelyas a monitor and as a controller. To evaluate both the monitoringand control capabilities under highly stressed conditions, solubleiron was added during the tests. Soluble iron is known to interferewith the functioning of polymer dispersants and will result in theprecipitation/deposition of both iron and calcium phosphate. Pre-cipitation of either of these salts reduces the polymer concentra-tion in the circulating water, resulting in accelerated fouling. Alltest runs involved a neutral pH, inorganic phosphate-based con-trol program. The control parameters were as follows:

pH: 7.2

Calcium: 600 ppm as CaCO3

Orthophosphate: 15 ppm

Pyrophosphate: 3 ppm

Skin Temperature: 135°F (57°C)

Retention Time: 3 days

Figures 5 and 6 are without automated control of polymer, andrepresents monitoring mode only. The polymer was being appliedat a carefully controlled rate of 10 ppm based on the system’s mate-rial balance. Following a baseline period, the feed of a stressor inthe form of soluble iron was introduced, gradually building to afeed of 6 ppm soluble iron solution. The measured polymer de-creases due to the stress caused by iron, and the efficacy of thetechnology to respond to stressed conditions is clearly demon-strated. In Figure 6, prior to introducing iron stress, the solubleorthophosphate measurement was approximately 16 ppm.Orthophospate and pyrophosphate were fed at 15 ppm and 3 ppmrespectively. Approximately 50% of the pyrophosphate was re-verting to orthophoshate, which accounts for the 16 ppm ortho-phosphate measured during this portion of the test. Following theintroduction of stress, the orthophosphate concentration decreasedto ~12 ppm, representing a 4 ppm loss in orthophosphate. The lossof orthophosphate and inability corresponds to the reduction inthe measured polymer concentration documented (Figure 5)

Figures 7, 8 and 9 repeat the same test parameters protocol. Theonly difference is that in this study, the technology is in a monitorand control mode. In Figure 7, again, polymer is initially applied ata rate theoretically resulting in 10 ppm. Prior to the introduction ofthe soluble iron, measured polymer concentration is being main-tained at 6.0 +/- 0.6 ppm. At the bottom o f the chart, the polymerdelivery pump status shows whether is on (1) or off (0), and theduration of polymer feed. Following the introduction of stress to


the system, the measured polymer continues to be controlled withinthe 6.0 +/- 0.6 ppm as shown by the upper and lower control limits.There was no reduction in the measured polymer concentrationafter iron was introduced as observed in the previous study. Asillustrated in Figure 8, the polymer delivery pump was clearly ap-plying additional polymer in order to maintain the 6 ppm target, asevidenced by the change (increase) in the duration of the pumpbeing on. In Figure 9, the baseline orthophosphate level of 16 ppmis again seen. After the introduction of stress, there was no ortho-phosphate loss since control technology enabled the target mea-sured polymer concentration of 6 ppm to be maintained, despite thehigh iron levels.

Field PerformanceThe polymer control technology has been demonstrated in numer-ous l field applications. As an example, a fully integrated refineryhas a major cooling system that serves an entire complex of operat-ing units that are the core of the production facilities. Maintainingeffective polymer residuals and overall cooling water parametercontrol are critical to maintaining heat transfer efficiency and hencemaximum process throughput.. Collectively, these units convertcrude oil to gasoline, jet fuel, diesel fuel, liquid sulfur and petro-leum coke. The tower has a recirculation rate of 39,000 gpm (148 m3/h), and a system volume of 300,000 gallons (1,136m3). Typical towersupply and return temperatures are 89p F and 101pF (32pC and38pC) respectively.

Critical Heat Exchangers – In this system, there is a multitude ofcritical heat exchangers. One of the most critical is a trim cooler.This exchanger cools diesel product before storage. It is shell-sidecooling, with the diesel product on the tube side. Cooling waterexchanger outlet temperatures reach as high as 170pF (77pC), andvery high skin temperatures. Loss of cooling water flow due tofouling increases the temperature of the cooling water and dieselproduct. In the past, there was a history of fouling due to poorcontrol over pH and cycles in the tower. Avoiding the need to bringdown and clean this heat exchanger, particularly during hot sum-mer months, avoids significant per cleaning cost as well as the lossin revenue due to production downtime.

A neutral pH, phosphate-based treatment program was in place,.

Key cooling water operating parameters were as follows:

Calcium: 500 to 1,000 ppm as CaCO3

Phosphate: 10 to 12 ppm as PO4,

Conductivity: 4,000 to 6,000 umhos

Cycles of Concentration: 4 to 6

pH: 6.8 – 7.3.

Figures 10, 11 and 12 show key operating cooling water parametersand results over an extended period. Figure 10 shows cycles ofconcentration control and calcium hardness for the cooling water.Wide swings in calcium hardness (800 ppm range over the period)due to a highly variable make-up water quality can be seen. All elsebeing constant, this creates significant surges in polymer stress,threatening fouling of the critical heat exchangers by phosphatescales if not compensated for. Figure 11 shows the consistentpolymer management achieved with full polymer control activated.The threshold polymer concentration numbers are as measured bythe polymer control technology. The target was 8 to 10 ppm. The

polymer application was optimized, and the variation significantlyreduced. The standard deviation of the polymer concentration im-proved 3.5x once full control was established. Over the entire time,monitoring of the approach temperature of the aforementioned criti-cal cooler (Figure 12) revealed no deterioration in performance. Insum, the polymer control technology optimized the product appli-cation, and at the same time, maintained optimum equipment per-formance in spite of wide variations in system stress.

CONCLUSIONSA new polymer measurement and control technology, based on thedirect measurement of functional polymer chemistry, has been de-veloped and validated in both dynamic laboratory studies and fieldapplications

· The new technology directly measures the targeted polymerand overcomes many of the limitations associated with tracertechnology and indirect monitoring techniques.

· Development of a unique dye reagent chemistry and buffersystem eliminates background interferences from naturallyoccurring polymers and other anionic organics.

· Development of a trouble free monitor/controller ensuresaccurate and reliable performance with highly colored andhighly turbid waters.

ACKNOWLEDGEMENTSThe authors would like to acknowledge the support of GE Water &Process Technologies’ Cooling Research and Engineering teamsaround the world for their exceptional expertise leading to the inno-vation described within this paper.

REFERENCES1. G.E. Geiger, C. Sui, “Improved Calcium Phosphate Control

for Stressed Systems,” TP08-09, Cooling Tower InstituteAnnual Meeting, February 2008.

2. R. C. May, G. E. Geiger, D. A. Bauer, “A new Non-ChromateCooling Water Treatment Utilizes High OrthophosphateLevels Without Calcium Phosphate Fouling,” Corrosion/80, Paper No. 196, 1980

3. W. F. Beer, J. F. Ertel, “Experience With High PhosphateCooling Water Treatment Programs,” Corrosion/85, PaperNo. 125, 1985

4. G.E. Geiger, C. Sui, “Improved Calcium Phosphate Controlfor Stressed Systems,” TP08-09, Cooling Tower InstituteAnnual Meeting, February 2008.

5. Ibid

6. R.M. Post, GE.Geiger, C.Xiao, “Advances in On-Line Moni-toring and Control of Dispersent Chemistry in Power PlantCooling Systems”, The 28th Annual Electric Utility Chem-istry Workshop. May 6-8, 2008.

7. Ibid.

8. Hoots, J. E. (1993). Tagged Polymer Technology for Im-proved Cooling System Monitoring and Control, CORRO-SION/93, Paper No. 397, Houston, TX: NACE Interna-tional.

9. Buentello, K. E., et al. (2001). Method of stabilizing dyesolutions and stabilizing dye concentrations. U.S. Patent6,241,788, June 5, 2001.


Figure 2: Filtration, Sample Preparationand Detection Modules

Figure 3: Dynamic Laboratory Unit

Figure 4: Dynamic Laboratory Study

Figure 5: Dynamic Laboratory Study without Polymer Control

Figure 6: Dynamic Laboratory Study without Polymer Control(Continued)

Figure 1: Polymer Monitoring Device

10. Buentello, K. E. Zhang, E. Y., Scattolini, J. L. (2001. AQuantitative Polymer Method For Cooling Water Systems,NACE Corrosion 2001, Paper No. 01447.

11. R.M. Post, G.E. Geiger, C.Xiao, “Advances in On-Line Moni-toring and Control of Dispersent Chemistry in Power PlantCooling Systems,” The 28th Annual Electric Utility Chem-istry Workshop. May 6-8, 2008.

12. Ibid.


Figure 6: Dynamic Laboratory Study without Polymer Control(Continued)

Figure 7: Dynamic Laboratory Study with Polymer Control

Figure 8: Dynamic Laboratory Study with Polymer Control(Continued)

Figure 9: Dynamic Laboratory Study with Polymer Control(Continued)

Figure 10: Calcium and Cycles

Figure 11: Measured Polymer Concentration

Figure 12: Critical Heat Exchanger Approach Temperature


Keith HirschBASF CorporationJohn Byrne, Barry SperonelloBASF Catalysts, LLC

BackgroundChlorine dioxide is a powerful oxidizing biocide. Addition-ally, it is a gas and is water-soluble, which enables it topenetrate entire sections of a water treatment system.

Intermittent Feeding of Aseptrol®Tablets Redefines the Role of ChlorineDioxide in Small and Mid-sized CoolingWater Systems

Chlorine dioxide can oxidize some inorganics in solution.Fe+2 and Mn+2 in solution can oxidize to higher valenceinsoluble forms (e.g., Fe+3, Mn+3), so it may be necessaryto increase chlorine dioxide dose rates in systems that con-tain high concentrations of those metals in dissolved form.2

Chlorine dioxide does not react with phosphates.

An initial evaluation of BASF Cooling Water Biocide(BCWB), based on chlorine dioxide-generating Aseptrol®

Chlorine dioxide provides several advantages over traditional chlo-rine. For example, the efficacy of chlorine dioxide is independent ofpH (up to 10). Chlorine dioxide also does not chlorinate organicsand does not produce trihalomethanes (THMs). As a biocide, chlo-rine dioxide exhibits broad-spectrum kill, showing efficacy, for ex-ample, against Bacillus anthracis (Anthrax) and Legionellapneumophelia. Furthermore, it is effective against microorganismsnot controlled effectively by chlorine, such as Cryptosporidiumand Giardia. Of great importance in all areas of water treatment isthe fact that chlorine dioxide is a very effective biofilm penetrant,enabling the complete removal of biofilm that harbors microorgan-isms.1-3

Among the oxidizing biocides, chlorine dioxide is recognized asbeing the least reactive with organics and many inorganics. It iswidely used for bleaching wood pulp in paper manufacturing be-cause its relative lack of reactivity with wood fibers produces thehighest quality paper products.2 As a result, chlorine dioxide iscompatible with wooden components in industrial water systems.Chlorine dioxide also is unreactive towards ammonia, and is thepreferred oxidizing biocide for use in industrial water systems thatcontain even low levels of dissolved ammonia. Chlorine dioxide isunreactive towards paraffinic hydrocarbons and only very slowlyreactive with olefins.2 It is also unreactive towards biguanide-based biocides, such as polyhexamethylene biguanide hydrochlo-ride (PHMB).4 Laboratory work at BASF has found that chlorinedioxide is unreactive with acrylates and with all quaternary aminestested to date. It reacts slowly with aldehydes, such as glutaralde-hyde.4

Conversely, chlorine dioxide is highly reactive with the functionalgroups of key classes of organic compounds. It reacts quickly tooxidize and deodorize reduced sulfur groups including H

2S and

organo-sulfur compounds. It also oxidizes the amine group of ter-tiary amines, and phenolic hydroxyl groups. In those reactions,however, little chlorine dioxide is consumed by unselective reac-tions with the hydrocarbon molecular backbone. As a result, chlo-rine dioxide is a highly effective deodorizer at low concentrations,as well as a powerful biocide.2

tablets, demonstrated the effectiveness of the technology as a cool-ing water biocide.5 Subsequent tests of BCWB were carried out attwo BASF manufacturing sites to optimize the chlorine dioxide dos-age. One location was a relatively dusty facility that manufacturedclay products and clay-based fluid cracking catalysts in southernGeorgia (Attapulgus), and the other was in a plant with a much lessdusty environment that manufactured a variety of inorganic cata-lyst and pigment products in northern Ohio (Elyria). This papersummarizes the results of these two cooling tower trials of BCWB.

ExperimentalBaseline levels of microbial contamination were measured usingdip slides (either Biosan SaniCheck® BF or Biotrace BTM2) withthe standard treatment regimen that was in-place at the time. Mea-surements were made of both suspended planktonic (bulk water)and sessile organisms on tower surfaces. The microbiological (MB)counts were measured after the slides were allowed to develop atroom temperature for the recommended 24 to 36 hours. Bulk waterresults were reported as CFU/ml (colony forming units/ml). SurfaceMB results were taken by contacting the entire surface of the dipslide plate at the same spot on a cooling tower slat. Surface countswere reported as CFU (per 11.5 cm2).

The original treatment regimen was suspended, and the towerswere then dosed with BCWB at varying rates and frequencies, andMB measurements were continued. After addition of the 8.33-gramBCWB tablets, the blow down circuit was shut off for about sixhours to prevent any of the chlorine dioxide generated from beingdischarged. The tablets completely exhausted their supply of chlo-rine dioxide within the 6-hour time period. Chlorine dioxide concen-trations were measured from samples taken at the discharge of therecirculation pump up to 120 minutes after biocide addition usingthe DPD method (Hach Pocket Colorimeter™ II, Product # 5870051).

Most MB counts were taken just prior to dosing with biocide (theworst-case situation), but some measurements were made at inter-mediate times to illustrate rates of re-growth. These trials demon-strated the viability of chlorine dioxide for the control of microbialgrowth in cooling towers.

Keith Hirsch


ResultsAttapulgus, Georgia Site5,500-Gallon Cooling Tower:

This trial was conducted over about a 20-week period at the BASFmanufacturing plant in Attapulgus, Georgia. The cooling towerwas in a relatively dusty location and the basin routinely containeda layer of solid sediment several inches thick. Such sediment pro-vided a protected environment for microbial growth.

Eight (8) weeks of baseline data were collected while the tower wastreated with bromine/chlorine briquettes that were applied by thewater treatment service company. This was followed by another 8weeks of treatment using BCWB. Water pH averaged 8.4 through-out the trial. Additionally, the bulk water and tower surfaces weresampled for microbes periodically. Basic information on the cool-ing system at the Attapulgus site, such as cycles of concentrationand water chemistry, was not revealed before or during the trial.The source of the make-up water was chlorinated lake water.

BCWB was added to the tower in the form of multiple 8.33-gramtablets enclosed in a polyester fabric mesh bag. The chlorine diox-ide was introduced into the unit by periodically adding a bag oftablets to a 3-gallon feeder and passing water through the feeder tothe tower at a rate of 5 gallons/minute. The solution concentrationof chlorine dioxide in the feeder was set to a maximum of 4,000 ppm.

Figure 1 is a graph of surface and waterborne microbial counts (asthe base 10 logarithm) versus time during the trial. Bromine/chlo-rine briquettes were used from day -55 through day zero. At dayzero, the unit was switched to weekly addition of BCWB and wastreated with BCWB through day 56. Figure 1 also shows the addi-tion rate of BCWB during its use.

Figure 1 shows that the tower was under only fair control duringthe baseline period, with four of the seven surface data pointsbeing at or above the maximum target of 105 CFU and with one ofthe seven bulk (waterborne) measurements over the maximum tar-get of 104 CFU/ml. Variability of microbial counts was high duringthis period, and this was attributed to microbial contamination ofthe system from organisms growing in the sediment layer (whichwe believe were not well controlled by the standard biocide).

Treatment with BCWB begins on day zero in Figure 1 and contin-ues through day 56. The dose rate was optimized throughout thetrial to control microbial counts. The initial weekly dose rate was180 grams/1,000 gallons and was increased to 275 grams/1,000 gal-lons on day 14, and further increased to 360 grams/1,000 gallons onday 35. The last dose, on day 56, was reduced to 180 grams/1,000gallons per week. During this time the average microbial countswere similar to the baseline period, but there was less variation withBCWB. As noted in Figure 1, only one surface reading and onebulk reading out of 19 of each were above the maximum targetlevels. It was interpreted that the improved consistency resultedfrom better control of organisms in the sediment layer, which re-duced the severity of microbe release from the sediment into thewater.

Figure 2 is a graph of BCWB dose rate and chlorine dioxide residu-als at 30 and 60 minutes after addition throughout the course of thetrial. Water samples were collected at the circulation pump dis-charge. The figure shows that there was little effect of dose rate onchlorine dioxide residuals except at the highest dose rate of 360

grams/1,000 gallons/week, where the chlorine dioxide residualdoubled from 0.15 ppm to 0.30 ppm. The residual values wererelatively constant 30 and 60 minutes after BCWB addition. Chlo-rine dioxide residuals were measured up to 90 minutes after dosage,at which point readings were no longer taken. Average chlorinedioxide residuals were slightly above 0.1 ppm after 90 minutes.Weekly dose rates averaged ca. 275 grams/1,000 gallons over thetrial period.

Elyria, Ohio Site5,500-Gallon Cooling Tower:

This test was conducted on a 5,500-gallon cooling tower at theElyria, Ohio plant of BASF Corporation. Two months of baselinedata were collected during which time the tower was operated withsodium hypochlorite that was applied by the water treatment ser-vice company.6 This was followed by replacement of the chlorinebleach with BCWB.

BCWB was added to the tower in the form of multiple 8.33-gramtablets enclosed in a polyester fabric mesh bag. The chlorine diox-ide was introduced into the unit by periodically adding a bag oftablets to a 6.3-gallon feeder and passing water through the feederto the tower at a rate of 10 gallons/minute. The solution concentra-tion of chlorine dioxide in the feeder was set to a maximum of 4,000ppm. BCWB was added weekly for the first eight weeks, then dailyfor the next three weeks, and weekly for the final three weeks.

The cooling system was run at ca. 2.5 cycles of concentrationduring the baseline and trial periods. The source of the make-upwater was city water. A molybdenum-based corrosion inhibitorwas used at levels between 2 and 4 ppm Mo6+. The water chemistrychanged little during the trial relative to the baseline period. Forexample, during the trial, conductivity was 690 mmhos vs. 740 mmhosduring the baseline period; calcium, as CaCO

3, was 255 ppm vs. 305

ppm during the baseline period; total hardness, as CaCO3, was 335

ppm vs. 380 ppm during baseline period. The pH of the water wasfairly high, at 8.5 to 8.8 vs. 8.3 to 8.9 during the baseline period.

Figure 3 is a graph showing the effect of time since treatment onmicrobial counts for six separate time intervals during the trial. Itshows that, as expected (with a few exceptions), counts are typi-cally very low immediately after treatment and then rise with timebetween treatments. In only one case (+ sign in Figure 3) is anearlier count higher than a later count, and in that case both read-ings are relatively low. Consequently, data taken just prior toretreatment represents a worst-case evaluation, and those are thesame data presented in Figure 4.

Figure 4 is a plot of microbial count and BCWB dose rate as afunction of time during the two periods of weekly addition. Initialweekly dose rates were 180 grams of BCWB/1,000 gallons of water,and microbial counts fell to zero. However, that result was laterfound to be invalid due to a valve failure that caused the tower tobe flooded constantly with fresh, city water. When that was cor-rected, the weekly dose rate of BCWB was varied between ca. 90and 180 grams/1,000 gallons until it was determined that acceptablemicrobial counts (<104 CFU/ml) could be achieved at an averagedose rate of 155 grams/1,000 gallons/week.

Figure 5 is a graph of microbial counts and BCWB dose rate (ingrams for the entire 5,500-gallon tower) as a function of time duringthe period of daily additions. It shows that dose rates may be


reduced relative to weekly addition and microbial counts can stillremain within an acceptable range (< 104 CFU/ml). Optimum controlwas achieved at an average dose rate of 110 grams of BCWB/1,000gallons/week using daily additions. This represents a 29% reduc-tion in BCWB dosage versus the optimal weekly dosage.

Figure 6 is a graph of chlorine dioxide residual measured either 30minutes or 60 minutes after dosing with BCWB during periods ofweekly addition. Water samples were collected at the circulationpump discharge. The graph shows that chlorine dioxide residualincreased with BCWB dose. It is also noted that the residual after60 minutes was consistently lower than it was after 30 minutes,indicating loss of chlorine dioxide from the system after the 30-minute interval. However, chlorine dioxide residuals were mea-sured up to two hours after dosage, at which point readings wereno longer taken. Average chlorine dioxide residuals were slightlybelow 0.1 ppm after two hours. The data in Figure 6 also show thatin this tower at the optimum weekly dose rate of ca. 130 grams/1,000gallons, the chlorine dioxide residual after 60 minutes was in therange of 0.1 ppm.

Tests were also carried out to assess the impact of chlorine dioxidedosage on corrosion. Rates were measured using test coupons ofcopper and mild steel during the period of BCWB use. Observedcorrosion rates were 0.25 mpy (90-day basis) for mild steel and 0.09mpy (90-day basis) for copper, both excellent values. A baselinecomparison was run on two mild steel coupons after the trial whenthe addition of chlorine bleach was resumed. The average corro-sion rates were 0.33 mpy (180-day basis).

Conclusions1. BASF Cooling Water Biocide (BCWB) was able to control

planktonic microbial counts to less than or equal to thetarget value of 104 CFU/ml at dose rates between 110 and275 grams/1,000 gallons/week.

2. A 29% reduction in the dosage of BCWB was achievedthrough daily addition versus weekly addition. Based onthis observation, an automated daily feeder has been de-veloped.

3. Microbial control was achieved with a chlorine dioxideresidual of 0.1 ppm or greater measured at the pump dis-charge at a time between 30 and 60 minutes from dosing.Thus, for effective control of microbiological growth, it isrecommended that a chlorine dioxide residual of at least0.1 ppm be achieved within 60 minutes of BCWB addition.

4. Microbial control levels were at least equal to thoseachieved using the prior commercial biocides at their rec-ommended dose rates.

References1. Gates, D. 1998 The Chlorine Dioxide Handbook. Denver:

Ammerican Water Works Association.

2. Simpson, G. D. 2005 Practical Chlorine Dioxide , Vol. I:Foundations. Colleyville, Texas: Greg D. Simpson & As-sociates.

3. Simpson, G. D. 2005 Practical Chlorine Dioxide , Vol. II:Applications. Colleyville, Texas: Greg D. Simpson & As-sociates.

4. BASF Corporation internal data, unpublished.

5. Puckorius, P. R., Puckorius, D. A., Speronello, B. “NewSolid Chlorine Dioxide Tablet as Cooling Water Micro-BioControl – Case Histories / Application Data”, paper pre-sented at 2005 AWT Annual Conference & Exposition,Palm Springs, CA, September 20, 2005.

6. Service of the Elyria, OH cooling tower was provided byCrown Solutions, Inc. of Dayton, OH.

Figures


Thomas R. KlineSTRUCTURAL GROUP, INC.

INTRODUCTIONMechanical Draft Cooling Towers arecritical fixed Assets that need to bemaintained and remain in service inorder to cool various plant operationsand systems. Essentially, the CoolingTower Basin in mechanical draft tech-

are located – and what it will cost to repair them – arrangedfrom highest to lowest priority.

Figure No. 1 – Condition Survey Flowchart identifying theinter-relationship between Field Investigation, Laboratory

Tests and Documentation.

Once adequately characterized, a thoughtful and detailed repairapproach can be developed addressing thermodynamic, chemicaland construction material properties of the structure operatingwithin the Cooling Tower process service environment - optimallyresulting in a long-term repair program.

STRUCTURE DETERIORATION TRENDS

Depending on process requirements, reinforced concrete is theconstruction material of choice for subsurface structures involvedwith partial or direct burial. However, as in all site-built construc-tion projects, construction defects can be significant dependingon the effort exercised with on-site Quality Control/Quality Assur-ance (QC/QA). Unfortunately, due to the aggressive operating ser-vice environment, containment structures with even small construc-tion defects (e.g. honeycomb concrete, misplaced embedded rein-forcing steel, waterstops, etc.) can greatly diminish the anticipatedservice-life of the subject structure.

It’s not unusual when reviewing prematurely deteriorated contain-ment structures that the original designer omitted provisions in thebuilding code, specific to environmental structures. These codeprovisions take into account corner cracking and require an in-crease in embedded reinforcing steel to address fluid containment1.

While in operation, Cooling Tower Basin Structures are continu-ously exposed to elements that are detrimental to the integrity ofreinforced concrete which leads to shortened life-expectancy. Nu-merous types of aggressive deterioration mechanisms exist withinCooling Tower Basin Structures and these mechanisms need to beaccurately identified and mitigated effectively. Initially, all reinforcedconcrete is exposed to long-term material shrinkage which is actu-

Cooling Tower Basin Evaluation andRepair

nology serves a two-fold purpose, one as containment for coolingwater and two as the foundation for supporting overlying “fill”structures. Almost all of these Basin Structures are constructed ofconventionally reinforced concrete, either partially or totally placedbelow grade. Their service environment subjects them to variousaggressive deterioration mechanisms including embedded metalcorrosion, original construction defects, environmental degrada-tion (i.e., freeze-thaw, algal growth, etc.) as well as chemical attackresulting in erosion and structural section loss. Since these struc-tures contain process water and function as support foundations,the existing condition of these structures should be evaluated regu-larly and repairs performed, should conditions warrant, on a timelybasis.

CONDITION SURVEY/FORENSICINVESTIGATIONConcrete deterioration comprises both obvious and latent charac-teristics that are not easily understood without gathering furtherinformation through investigation. Like forensic efforts in otherprocess units, the Cooling Tower environment can be hostile for anin-process evaluation. However, techniques have been developedto assess concrete exterior surfaces On-Line and interior surfacesquickly during short duration outages in attempts to determine thecauses and effects of concrete deterioration. Employing a combi-nation of Non-Destructive and Semi-Destructive Testing (SDT)techniques, characterizations as to the physical and chemical char-acteristics of the reinforced concrete structures can be determinedquickly. Using cutting-edge analytical and diagnostic tools, theevaluator establishes these repair parameters:

• An evaluation that investigates further, and qualifies causes& effects;

• A quantification of the problem that expresses its extent inconcrete terms (e.g. square feet, cubic feet, linear feet, etc.);and

• Documentation describing where the distressed conditions

Thomas R. Kline


ally desiccation of the “moisture-of-convenience” which makesthe concrete placeable, but is not part of the hydration/strength-gain of the concrete. This mechanism subjects the concrete massto volumetric shrinkage stresses that can result in concrete cracks.Concrete as a construction material is strong in compression butweak in tension. Therefore steel reinforcing systems, generallyembedded into the concrete mass, resist applied tensile forces.Structure designers generally understand this phenomenon andspecify low Water/Cement ratio (W/C) concrete (i.e., concrete mix-tures with an extremely low “moisture-of-convenience” amount)with sufficient embedded reinforcing steel to accommodate these“tensile” stresses. However, many times designers apply the wrongprovisions of the Building Code and the result is insufficient steelreinforcement.

Figure No. 2 – Corrosion-Induced Deterioration caused byembedded metal corrosion.

Cracks, acting as conduits, allow deleterious substances to deeplypenetrate the concrete mass as shown in Figure No. 2, above.Treated water, chemically altered for process considerations, canbe detrimental to concrete. Chemicals in the water can serve aselectrolytes (e.g. chlorides added to control algal growth) initiatingpremature deterioration of the concrete via corrosion processesassociated with the embedded steel reinforcing systems or attackthe concrete matrix (e.g. sulfuric acid added to modulate pH) andessentially erode the concrete mass by Sulfate Attack.

Sulfate Attack can be insidious within Cooling Towers as concen-trated Sulfuric Acid is used to modulate the pH of some coolingwater systems. The chemical injection point can concentrate theacid adjacent to unprotected concrete surfaces. Expansion forceswithin concrete, as shown below, are detrimental to the durabilityof concrete in service. The reaction causes significant surface ero-sion on concrete Cooling Tower Basin surfaces, sometimes onlyobservable during a Unit outage.

Sulfate Attack of Concrete

• Sulfate & Calcium Ions form Gypsum(CaSO4•32 H2O) - expands 124% in volume

• Sulfate & Calcium Aluminate form Calcium Sulfoaluminate(ettringite) (3CaO•Al2O3•3CaSO4•3H2O) - expands 227% involume

View of Cooling Tower Basin Wall Deterioration – Embeddedmetal corrosion and chemical attack of the concrete matrix

affected the concrete wall section approximately 5" in depth.

COOLING TOWER REPAIR PROCESSHistory and experience have shown that each Cooling Tower BasinStructure poses unique challenges to a Repair Contractor. Regard-less of whether the required repair involves partial or full-depthwall/floor section repairs, foundation stabilization, containment liner,crack repair or simply stopping cooling water egress, it’s imperativeto utilize an engineered solution. A proper repair strategy shouldconsist of the following elements:

• Identifying and determining the root-cause of the failed con-crete;

• Employing proper materials in construction and repair tech-niques; and

• Using a qualified, experienced contractor who can provide asolution, as well as a well-planned quality control and assur-ance program (QC/QA), for the repair.

These three steps will assure the Owner that the repair-failure-repair cycle is eliminated and a sound structure put back into op-eration. A more comprehensive view of how these steps translateinto the Repair Process is shown in Figure 3.

Figure No. 3 – Concrete Repair is a Team Process involvingthe Owner, Engineer and Repair Contractor working toward a

common goal of an enduring repair.


REPAIR SCENARIOSAs each Cooling Tower Basin Structure is unique in constructionand service, so to, many repair opportunities exist for structuralrestoration. Repair, based on the results of the Condition Survey/Forensic Investigation discussed above, can take many forms in-cluding, but not limited to, repair of leaking cracks (expansive groutinjection, polymer crack stabilization, battened membranes, etc.),repair of structural components (e.g. walls, floor slabs, fill supportcolumns, etc.), foundation stabilization (e.g. subsoil modification,compaction grouting, pier support, etc.), corrosion mitigation sys-tems (e.g. passive cp with embedded anodes, active cp with im-pressed current, corrosion inhibitors, etc.), and establishment ofnew protective interior surface liners (e.g. fluid applied coating,sheet applied membrane, spray-applied lining system, etc.). Expan-sion joint and embedded waterstop repairs reestablish integrity tothe details of construction, critical in process fluid containment.

When repairing concrete elements “In-Kind” in order for it to func-tion adequately, it’s important that the repair materials selected becompatible with the existing concrete substrate, matching as closelyas possible:

• Modulus of Elasticity (Y = σ/εσ/εσ/εσ/εσ/ε)• Thermal Expansion (ΔΙ/Ι = αΔΤΔΙ/Ι = αΔΤΔΙ/Ι = αΔΤΔΙ/Ι = αΔΤΔΙ/Ι = αΔΤ)• Low Material Drying Shrinkage (crack-free)• Repairing like-with-like!

Obviously, besides matching some of the engineering propertiesfor repair construction, cementitious repairs should incorporatecorrosion inhibitors as the concrete will be subjected to ample oxy-gen and moisture in service. Being chemically resistant to sulfatesis also important, especially in those Cooling Towers that dependon Sulfuric Acid for cooling water pH control. Generally, the sulfateresistance of a Portland Cement Concrete stems from low levels ofTricalcium Aluminate, C

3A, so as to not react with sulfate ions

which can initiate expansive reactions within the concrete mass4.Chemical resistance can also be improved by the reduction of thePortland Cement fraction within the repair concrete and replacingthat portion of the cement with mineral or pozzolan admixtures (e.g.flyash, microsilica, etc.) that also have cementitious properties5.Typically, a durable Cooling Tower Basin repair design involvesone or more of the following details:

• Non-Corrosive Fiber-Reinforced Plastic embedment prod-ucts (FRP)

• Dense, durable conventionally-reinforced cast-in-place con-crete

• Mechanical anchorages for composite bonding of repairs toexisting concrete substrates

• Implementation of corrosion mitigation systems (i.e., sacrifi-cial or impressed)

View of a Cooling Tower Basin Wall Repair, being preparedOn-Line – note splash wall placement above repair.

Deteriorated concrete removed, mechanical anchorsinstalled, embedded reinforcing steel cleaned and

augmented, just prior to formwork installation.

Crack Repair: Crack repair requires a basic knowledge as to whyreinforced concrete cracks. Modern concrete is the end-product ofan 80-year trend toward faster hydrating cements and ever-highercement contents. This trend has produced very strong but alsovery crack-prone concrete. Major reinforced concrete structuresexhibit significant distress because they are more restrained againstvolume change, undergo greater moisture and temperature changes,the concrete is stronger, has a high modulus, and little creep capac-ity to relieve the self-stress from thermal contraction, autogenousshrinkage, and drying shrinkage1. Understanding the root-causemechanisms associated with observed cracking will assure that therepair, when implemented, will be long lasting with two repair typesshown in Figure No. 4.

Standard crack repair technology typically “glues” disjointed con-crete members together, stabilizes individual segments of a onceunified concrete mass or stops the ingress of groundwater/egressof process fluids. Occasionally cracks form in Cooling Tower Basinstructures that result from movement during service. Often, de-signers place joints within structures that are designed to move,such as in the case of expansion joints. Both moving cracks andexpansion joints require special attention. Durable yet flexible con-struction materials are required to address in-service movement as“rigid” repair efforts will fail from forces developed via restraint.Typically, chemically-resistant, high-temperature tolerant mem-branes (e.g. Hypalon, etc.) are specified for repair of moving cracks,glued using high-strength paste adhesives or mechanically fas-tened to crack shoulders. Expansion joints will generally incorpo-rate neoprene gland assemblies or pneumatically inflated diaphragmglands installed into joint cavities, allowing the structure to move,yet maintaining containment integrity2.


Figure No. 4 – Two types of common crack repair employedat Cooling Tower Basins. Rout & Seal works best on

“positive-side” (i.e., water-side) applications and can beperformed only during Off-Line, drained Basin conditions.Pressure-Injecting Chemical Grouts into crack fissures

consists of either hydrophilic or hydrophobic urethane groutswhich expand in the presence of moisture and can beperformed while the Cooling Tower Basin is On-Line.

View of a Failed Cooling Tower Basin Wall Water-Stop – anoriginal construction defect as the water-stop was

misplaced and failed in service.

Structural Member Repair: Aggressive deterioration mechanismsassociated with embedded metal corrosion and/or sulfate-relatedchemical attack of the concrete mass can significantly affect thestructural integrity of reinforced concrete members within a struc-tural system (e.g. walls, base slab, column support pedestals, etc.).Reduction in both concrete and embedded reinforcing steel barcross-sections can create conditions of impending Structural Risk,in some cases requiring immediate action in the form of temporarysupport shoring or process bypass. At-Risk structural behaviorcan range from slow, barely noticeable, structural member deflec-tions to “failure-without-notice” of structural systems supportingCooling Tower fill components.

Should significant distress conditions be exposed during a regu-larly scheduled maintenance outage, an evaluative approach, asdiscussed earlier, should be initially employed:

• Locate the deterioration• Qualify the distress mechanisms and determine the “root-

cause”• Quantify the amount of repair to assess repair methodology

- determine whether to Repair or Replace-in-KindOnce a repair methodology has been selected, follow the ConcreteRepair Industry Best-Practices2:

• Demolish and remove unsound/deteriorated concrete mate-rials

• Prepare resultant sound/competent concrete substrate surfaces

• Assess and augment, if necessary, deteriorated embeddedsteel reinforcing systems3

• Implement corrosion control measures, if evidence indicatessignificant embedded metal corrosion activity – embed sac-rificial zinc anodes, metalize exterior repaired concrete sur-faces or install an impressed current system

• Select appropriate concrete repair materials that have con-sistent plastic and hardened characteristics and propertiesto ensure composite behavior between the existing concretesubstrate and new “cured” repair materials

• Install the selected repair materials using placement/appli-cation techniques consistent with the desired end repairproduct that achieves adequate bond and results in lowshrinkage cracking

Concrete Placement of a Cooling Tower Basin Wall Repairperformed On-Line – repair materials were mixed on-site

and then hydraulically pumped into formwork cavities. NoteFRP formwork ties and temporary splash wall.

In conclusion, reinforced concrete Cooling Tower Basin Structurescan be successfully repaired On-Line & Off-Line, providing a sig-nificant extension to their service-life. These types of repairs how-ever, can only be implemented once we understand:

• Owners Requirements;• Process items specific to the Facility;• Deterioration mechanisms in-place within the structure and;• Securing of Repair Professionals who offer an “engineered

approach” and have the background and experience to imple-ment the repair successfully.

References1. Burrows, R. W., The Visible and Invisible Cracking of

Concrete, American Concrete Institute Monograph No.11, 1998, pg. 1.

2. Concrete Repair Manual, 1999 Edition, Published jointlyby the International Concrete Repair Institute, Sterling ,VA and the American Concrete Institute, Farmington Hills,MI, 1999, 861 pgs.

3. Manual of Standard Practice, 27th Edition (MSP-2-01),Concrete Reinforcing Steel Institute, Schaumburg, IL, 2001,pgs. 4-4 & 4-5.

4. “Guide to Durable Concrete,” ACI Manual of Practice,Part 1, ACI 201.2R-92, American Concrete Institute, De-troit, MI, 1998.

5. Kosmatka, Steven H., Panarese, William C., Design andControl of Concrete Mixtures, Portland Cement Associa-tion, 13th Edition, Portland Cement Association, Skokie,IL, 1988, pgs. 68-70.


Robbie McelveenBill MartinRyan SmithBaldor Electric

AbstractImproved reliability of cooling tower fan drives isnow possible due to new advancements in motortechnology. This paper discusses the developmentof low speed, permanent magnet motors and howthey can be used in direct-drive applications to elimi-nate the gearbox, NEMA motor, driveshaft, and disccouplings from cooling tower designs. A case studyis presented where a tower was refurbished using adirect-drive motor designed to fit the exact footprint

quired to only 12.5% of the rated value [1]. However,when any air flow even slightly above that providedby half speed operation is required, a two speedmotor must be run at rated horsepower as there is noother speed available. Two speed motors do providesome energy savings, but still must be cycled on andoff to maintain the desired water temperature. Thiscycling involves many “across the line” starts draw-ing high amps and placing unnecessary strain on themechanical components of the system [2]. Whileproviding some flexibility in the tower control logic,two speed motors are not optimal when it comes tomaximizing energy savings during times of reducedheat load.

Recent Developments In MotorTechnology Allow Direct Drive Of LowSpeed Cooling Tower Fans

and height of the existing gearbox. Design considerations, perfor-mance data, maintenance, and efficiency comparisons will be dis-cussed.

I. IntroductionThe most common solution for driving the fan in current coolingtower designs utilizes an induction motor, driveshaft, disc cou-pling, and gearbox arrangement, as shown in Figure 1. Few changesto this design have been made in the last twenty years.

Figure 1: Typical Fan Drive Arrangement

The motor used is normally a standard NEMA induction motor. Forreduced energy consumption, two speed motors have been ap-plied for use when full fan speed is not required due to decreasedheat load. As the horsepower required to drive the fan varies as thecube of the fan speed, it is advantageous to reduce the fan speedwhen possible. When the heat load decreases enough, the drivemotor can be run at half speed. This lowers the horsepower re-

The use of variable frequency drives (VFDs) has become muchmore commonplace in recent years. Data from a noted coolingtower manufacturer indicates that VFDs are being installed in themajority of all new towers being constructed. Additionally, mosttowers being upgraded or refurbished are also being equipped withVFDs. These drives have the advantage of a soft mechanical start,no large starting current draw, and the ability to run the fan at anydesired speed from zero to the maximum design speed for the appli-cation 3. The energy savings realized by using a VFD are wellrecognized and documented, so no further discussion will be intro-duced here 4. Several factors that must be considered when apply-ing a VFD are any critical speeds of the mechanical system, thecooling ability of the induction motor at low speed, and the properlubrication of the gearbox at slow speeds. For practical purposes,the fan is generally not run at speeds below 30% of the nominaldesign speed.

Historically, the mechanical components of the fan drive system,specifically the right angle gearbox, have been the largest mainte-nance issue for cooling tower installations 5. Gearbox failures, oilleaks, oil contamination, failed drive shafts, misaligned drive shaftsand excessive vibration are all significant problems related to thistype of fan drive system 6, 7.

In this paper, recent developments in motor technology are pre-sented. It is demonstrated how these innovative designs can beused to improve the reliability and reduce maintenance associatedwith today’s cooling tower installations. The design and installa-tion of a 208 rpm, 50 horsepower PM motor for a retrofit applicationis discussed in detail. The possibility of improved efficiency andlower energy consumption with the proposed solution is discussed.

II. Improvements In MotorTechnologyIncreased efficiency and improved power density are being de-

Robbie Mcelveen


manded in the motor industry. To achieve these goals, along withlower noise and variable speed operating capability, other tech-nologies beyond simple induction motors should be considered.Permanent magnet (PM) motors have long been recognized as pro-viding higher efficiencies than comparable induction motors. How-ever, limitations in terms of motor control, as well as magnet mate-rial performance and cost, have severely restricted their use. Dueto dramatic improvements in magnetic and thermal properties ofPM materials over the past 20 years, synchronous PM motors nowrepresent viable alternatives. Figures 2 & 3 show typical efficien-cies and power factors for various motor types 8.

Figure 2 - Typical Partial Load Efficiencies of

75 HP, TEFC, 1800 RPM Motors

Figure 3 - Typical Partial Load Power Factors of

75HP, TEFC, 1800 RPM Motors

Another innovation which merits discussion is the laminated framemotor technology used in this design. Laminated frame motorsconsist of a stack of laminations permanently riveted under con-trolled pressure. The cast iron outer frame is eliminated, allowingmore room for active (torque producing) magnetic material. Figure 4below is a representation showing how the stator frame is constructed.

Figure 4 – Laminated Frame Construction

Another advantage of this construction is that the air used to coolthe motor is in direct contact with the electrical steel. There is nothermal resistance path as that which exists in a traditional cast ironframe with contact to the stator lams. The heat transfer mechanismin a cast iron frame motor is highly dependent upon the stator toframe fit. Laminated frame construction eliminates this issue.

In recent years, industry drivers have forced the development of anoptimized, finned, laminated motor design. To improve the coolingand increase power density, fins have been added to the exterior ofthe stator laminations. The addition of the optimized cooling finsincreases the surface area available for heat dissipation. The resultis improved heat transfer and a power increase of 20-25% is typicalfor a given lamination diameter and core length. Figure 5 shows theincreased surface area achieved by including these cooling fins.

Figure 5 – Finned vs. Non-finned lamination

It is this improved cooling method, along with the higher efficiencyand power factor achieved with the PM technology that allows forincreased power density in these motor designs. Power density isthe key for being able to match the height restriction of the existinggearbox. For comparison, a paper study was performed to deter-mine the approximate sizes and weights of various motor types foruse in this application. The results are shown in Table 1 below. Therating is 50 horsepower at 208 rpm. Each motor was designed forthe same temperature rise.


Table 1 – Motor Size Comparisons

III. Case StudyThe case study involves the retrofit of an existing cooling towerconstructed in 1986 at Clemson University in South Carolina. Thetower information is as follows:

Fan Diameter: 18’-0"

Flow 4,250 gallons per minute (GPM)

Rates: per cell - 8,500 GPM total

Motor Information: Frame – 326THP – 50/12.5Speed – 1765/885 rpm

Gearbox: Size – 155, Ratio – 8.5:1

As shown in the above data, this tower is comprised of two identi-cal cells. For this study, one cell was retrofitted with the new slowspeed PM motor and VFD while the other was left intact as origi-nally constructed. This allows for a direct comparison of the twofan drive solutions. Figure 6 below shows Cell #1 in the originalconfiguration, while Figure 7 shows the PM motor installed in placeof the gearbox in Cell #2.

Figure 6 – Original Installation

Figure 7 – PM Motor Installed in Place ofGearbox & Driveshaft

Prior to the installation, the current being drawn by the two originalinduction motors was measured with the fans running at full speed.An ammeter was used and the current was measured to be fortyseven (47) amps, rms on both induction motors. As the inductionmotors are identical, this is a good indication that both cells wereoperating under the same load conditions. After the PM motor andVFD installation was complete, the current was again re-checkedand found to be only forty one (41) amps for the PM motor. Theinduction motor on the original, identical, tower was still drawingforty seven (47) amps.

A power meter was used to measure the input power to both solu-tions. The fans were running at the same speed. Data was taken atboth the input and output of the drive to allow for a direct compari-son of the induction motor / gearbox combination to the PM motor.The results of the measurements are shown in Table 2 below.

Table 2 – Power Consumption Comparison,Original blade pitch, manufacturer data

From this data, it was determined that both cells were running atless than full load and that the load should be increased on eachcell. To this end, the pitch of the blades on each fan was increasedto 12°. This change of pitch caused the fans to draw more air, thusincreasing the load on each motor. Further, the increased air flowimproved the effectiveness of the overall tower performance. Again,power measurements were made and a third party testing servicewas engaged to verify the manufacturer’s results. The data is shownin Tables 3 & 4 below.


Table 3 – Power Consumption Comparison,12°blade pitch, manufacturer data

Table 4 – Power Consumption Comparison,12° blade pitch, testing service data [9]

For the final blade pitch, 4.5 kW less power consumption was ob-served on the cell with the PM motor installed. In order to docu-ment the savings realized at various speeds on this application,input power was recorded at intermediate speeds for the PM motorcell. Figure 8 below shows the actual measured input power for theinduction motor / gearbox solution and the PM motor solution atvarious speeds.

Figure 8 – Input Power vs. Speed, 12° blade pitch

As shown in Tables 2-4, the PM motor solution requires less inputpower for each load point (blade pitch). Figure 8 shows the totalinput power in kilowatts for each solution over a range of operatingspeeds from 50-100%. Again, the PM motor has an advantage overthe induction motor / gearbox solution. Using an average price of$.08/kWh, the annual cost savings for various applications andduty cycles are shown in Table 5. This table does not account forthe additional savings achieved by using a VFD and having theability to run at speeds between 50% and 100% of rated.

Table 5 – Annual Energy Savings Based onVarious Duty Cycles

IV. Electrical ConsiderationsPM Control AlgorithmIn addition to the PM motor design features already detailed, an-other challenge of this application was that the PM motor had to berun sensorless. There was no room to install a speed feedbackdevice, such as an encoder or resolver, and still meet the heightrestriction of the existing gearbox. In this harsh environment, afeedback device would be a liability as far as reliability is con-cerned. Therefore, a sensorless PM control scheme was developedto satisfy the requirements of this application. Several things hadto be considered when forming this algorithm. One challenge wasthe inertia of the fan. This was taken into account to prevent themotor from falling out of synchronism when starting and changingspeeds. Figure 9 is a portion of a typical start from rest. Note thesmooth acceleration and low starting current required. A typical480 volt induction motor started across the line would draw 347amps 10, compared to 12 amps for this PM design started on theVFD.

Figure 9 – Motor Starting Performance

Improved Process ControlAs mentioned earlier, the addition of the VFD allows the user tomore accurately and efficiently control the process. Figure 10 showshow the motor speed is changed automatically with control logic asthe heat demand on the system changes with time.


Figure 10 – Motor Speed Variation with Changing Heat Load

Braking and Condensation ControlThe use of a VFD also provides the opportunity to offer someadditional features that across the line systems do not. The drivemay be configured to apply a trickle current to the motor windingsto act as a brake during down time. This prevents the fan from freewheeling due to nominal winds or adjacent cooling tower turbu-lence. However, a mechanical locking mechanism should be usingduring any maintenance procedures. This trickle current also actsas an internal space heater by raising the winding temperature,preventing condensation when the motor is not running.

Insulation SystemInside the fan stack is an extremely humid environment. Therefore,the insulation system on the stator windings must be robust andhighly moisture resistant. To this end, an insulation system derivedfrom a system originally developed for use by the US Navy wasemployed. This system utilizes an epoxy compound applied via avacuum pressure impregnation (VPI) system. The VPI system iswidely recognized as a superior insulation system for harsh appli-cations such as this. This particular system has been successfullyemployed on “open” motors in tough applications such as oil plat-forms operating in the North Sea.

V. Mechanical ConsiderationsShaft SealDue to the harsh environment inherent with a cooling tower appli-cation, the motor’s drive end is protected by a metallic, non-con-tacting, non-wearing, permanent compound labyrinth shaft sealthat incorporates a vapor blocking ring prevent an ingress of mois-ture. This seal has been proven to exclude all types of bearingcontamination and meets the requirements of the IEEE-841 motorspecification for severe duty applications. This type of seal hasbeen successfully used in cooling tower gearboxes for many years 11.

MaintenanceAnother consideration is overall system maintenance. For motor /gearbox combination drives, the lubrication interval is determinedby the high speed gear set. The recommended lubrication intervalfor this type of gear is typically 2500 hours or six months, which-ever comes first. In addition, gear manufacturers recommend adaily visual inspection for oil leaks, unusual noises, or vibrations.As these units are installed in areas which are not readily acces-sible or frequented, this is an unreasonable expectation and burdenon maintenance personnel. When a gear is to be idle for more thana week, it should be run periodically to keep the internal compo-

nents lubricated because they are highly susceptible to attacks byrust and corrosion. When being stored for an extended period, it isrecommended that the gearboxes be completely filled with oil andthen drained to the proper level prior to resumed operation. Be-cause the high speed input has been eliminated with the slow speedPM motor design, the lubrication cycle can now be extended up totwo years. The PM motor need not be inspected daily for oil leaks,as the motor contains no oil. As mentioned previously, the VFDcan provide a trickle current to heat the stator windings to a tem-perature slightly above ambient to prevent moisture from forminginside the motor.

VibrationWith the elimination of the high speed input to the gearbox, thesystem dynamics from a vibration standpoint have been simplified.There are no longer any resonance issues with the driveshaft. Themaximum rotational excitation is now limited to the rotational speedof the fan. The number of bearings in the drive system has beenreduced from six to two for a single reduction gearbox and fromeight to two for a double reduction gearbox. This reduces thenumber of forcing frequencies present in the system.

Noise LevelMany cooling towers are in locations where airborne noise can bean issue, such as hospitals and universities. To this end, a thirdparty testing company was engaged to conduct comparative soundtests between the two cells. Data was taken at both high speed andlow speed for both cells. The induction motor cell was designatedas Cell #1 while the PM motor cell was designated as Cell #2. Soundlevel measurements were taken on Cell #1 while Cell #2 was turnedoff. There were twelve 30-second readings taken at high speed andtwelve 30-second readings taken at low speed around the perimeterof the tower and the fan motor. As there was no motor outside ofthe fan stack on Cell #2, only nine readings were taken on Cell #2with Cell #1 turned off. A single point measurement was takenwhere the old induction motor was mounted on Cell #2 in order tohave some reference to Cell #1. It was not possible to turn off thewater flow for either cell at any time so there was a significantamount of background noise, but as this condition was the samefor both cells, it should not affect the comparative data 9. AverageA-weighted sound pressure results are shown in Table 6 for bothhigh speed and low speed operation.

Table 6 – Sound Pressure Data

At high speed, the PM motor cell was 4.6 dBA lower than theinduction motor cell. For low speed operation, the PM motor cellwas 5.4 dBA lower. Although there may be some slight differencesin the background noise for each cell, these likely do not accountfor all of the noise level reduction realized with the PM motor solu-tion. The removal of the high speed induction motor from theoutside of the fan stack appears to have the biggest influence onthe noise level of the tower itself.

VI. ConclusionsCooling tower fan drives have changed very little over the past twodecades. Failures of the gearbox, driveshaft, or disc couplingshave been the biggest reliability issue facing tower manufacturersand end users. Increasing energy costs have placed a premium onpower consumption for all motors and applications.


Many of the problems associated with cooling tower maintenanceand reliability are solved with the PM motor design. The relativelyhigh speed (typically 1750 rpm) induction motor has been elimi-nated. The motor itself has not historically been a problem, but theassociated resonances and potential vibration concerns have beenan issue. The driveshaft and associated disc couplings have beenremoved, thus eliminating problems associated with misalignment,improper lubrication, natural frequencies, or delaminating of thedriveshaft itself 12. The right angle spiral-beveled gearbox has beenremoved. Difficult maintenance associated with changing the oil,proper oil fill levels, contamination of the oil, oil leaks, and gearboxfailures is no longer a concern.

New motor technology now provides an alternative solution, thedirect drive of cooling tower fans. PM motor technology combinedwith the finned, laminated frame design now allows the construc-tion of low speed, compact motors for use in place of the existinggearbox. Data obtained to date indicates this solution will eliminatethe problems associated with the right angle gearbox and driveshaft design. By eliminating the gearbox, which is a significantsource of loss in the system, improved system efficiencies can berealized.

VII. AcknowledgementsThe authors of this paper extend their thanks to Clemson Univer-sity and Tower Engineering, Inc. for their contributions and partici-pation in the project.

VIII. References1. Benjamin Cohen, “Variable Frequency Drives: Operation

and Application with Evaporative Cooling Equipment”,Cooling Technology Institute Paper No. TP07-22, 2007

2. William F. Immell, “Variable Speed Fan Drives for CoolingTowers”, Cooling Technology Institute Paper No. TP96-03. 1996

3. Rick Foree, “Cooling Towers and VFDs”, Cooling Tech-nology Institute Paper No. TP01-07, 2001

4. M.P. Cassidy and J.F. Stack, “Applying Adjustable SpeedAC Drives to Cooling Tower Fans,” PPIC, 1988

5. Jim Horne, “How to Address Your Cooling System Woes,”PTOnline, 2008

6. Dave Gallagher, “Condition Monitoring of Cooling TowerFans”, Reliability Direct

7. Philadelphia Gear, “The Cooling Tower Gear Drive Dilemma:Why Applying Commodity Products to an EngineeredSolution Can Cause Premature Failure”

8. Steve Evon, Robbie McElveen and Michael J. Melfi, “Per-manent Magnet Motors for Power Density and EnergySavings in Industrial Applications”, PPIC 2008

9. Dustin Warrington, “Clean Air Engineering Report:Clemson East Chiller Plant”, July 2008

10. NEMA MG 1-2006, Motors and Generators11. Inpro/Seal Company, “An Introduction to Bearing Isola-

tors”, March 200512. Robert Poling, “Natural Frequency Characteristics of Drive

Shafts”, Cooling Technology Institute Paper No. TP05-04, 2005


Dustin TroutmanCreative Pultrusions, Inc.Jess SeawellComposite Cooling Systems, Lp

AbstractPultruded profiles dominate the cooling towermarket as the material of choice for field-erectedcooling towers. Pultruded profiles offer manypositive attributes. However, unlike the steel,concrete and wood industry, the pultrusion in-dustry has not matured to the point of havingASCE-endorsed design codes and quality re-quirements.

Therefore, the quality control and quality as-

ance (QA) Department, priorto receiving the raw materi-als. In some instances, ac-tual viscosity, temperature,and weights are verifiedagainst the manufacturer’srequirements prior to releas-ing the raw materials to pro-duction. The frequency andtiming of the inspectionsshould be outlined in themanufacturer’s qualitymanual.

Inspection of Pultruded Cooling TowerComponents

First Article InspectionThe pultrusion die setup and production startup are the two mostimportant processes associated with pultrusion manufacturing. Inorder to assure that the pultruded profiles meet both customer andinternal quality requirements, a documented First Article Inspec-tion (FAI) process is followed. In general terms, the FAI involvesthe use of an FAI document that contains a series of check boxescorresponding to specific procedures, test results, and measure-ments.

For example, the following series of check boxes demonstrate thesteps necessary to be competed prior to a product being releasedto production status. Note that many of the check boxes corre-spond to additional documents required to be completed duringthe die setup and preproduction stage. For example, the PultrusionDie Set Up Procedure/Checklist includes many additional docu-mented checks that occur during the die set up and preproduction.One example is a documented audit verifying that the proper rein-forcements are in the part at the preproduction stage.

o Is the Routing Sheet & Production Orders on Priority Sheeton line?

o Pultrusion Die Set Up Procedure Checklist Complete?o Does the Resin Mix Sheet Match the Engineering Specifi-

cation Sheet?o Does the profile conform to the QA Visual Specifications

provided on line?o Part Conforms to the Dimensional Specifications provided

on line?o Part Conforms to Length Tolerances as Established by

Quality Assurance?o Does the profile pass UL 94V0? _____NAo Does the profile pass the Full Section Modulus of Elasticity

(MOE) Requirement?The series of checks ensures that the pultruded sections meet theminimum internal and agreed upon physical properties and that the

surance standards are up to the individual pultruder. This paperwill discuss the visual and structural quality assurance and qualitycontrol programs utilized by an FRP pultrusion manufacturer andtheir customer for custom designed FRP structural members. Thispaper covers QA procedures utilized, both at the production plantand at the construction site.

Keywords: Fiberglass Reinforced Polymer (FRP), Pultrusion, Qual-ity Control (QA)

IntroductionThe pultrusion industry has been aggressively pursuing a Load &Resistance Factor Design (LRFD) standard in order to address majorcomponent and structure design with pultruded sections. Althoughthe standard will contain a Code of Standard Practice for Fabrica-tion and Installation of Pultruded FRP Structures, it will not ad-dress the quality control requirements for the production ofpultruded structural profiles.

This paper aims to address critical quality control standards thatare essential prior, during and post manufacturing of structuralprofiles for cooling tower applications. This paper will describe, ingeneral terms, the quality control requirements established betweena pultruder and a field erected cooling tower manufacturer.

Inspection of Incoming Raw MaterialsIncoming raw materials including resin, roving, fiberglass mats,fillers, and catalyst are required to be purchased based on a Quali-fied Products List (QPL). The list includes all products that arepermitted to be utilized in production. The documented materialapproval process is required to be followed any time a new materialis to be added to the QPL. The process includes an evaluationbased on mechanical, physical, and process parameters as com-pared to the legacy materials utilized for production.

The incoming resin and glass reinforcements require a Certificationof Conformance (COC) report, to be delivered to the Quality Assur-

Jess Seawell Dustin Troutman


die was set up via a consistent practice and per the requirementsdictated by engineering.

The FAI sheet is required to be signed by both the production lineoperator and the Pultrusion Supervisor. This action releases theprofile to production status. The First Article Inspection ensuresthat the pultrusion meets the minimum standards at the time theprocess was started. The First Article Inspection documents arescanned and saved in the internal QA files for future reference.

In-Production InspectionOnce the initial startup has been commenced and the First ArticleInspection has been completed, the QA complaint profile is thenreleased to production status. At this time, the In-Process Inspec-tion (IPI) procedure is required to be followed.

Specifically, the IPI procedure details the necessary inspectionsand the frequency of inspections required to be documentedthroughout the production shift. The IPI inspections include thefollowing:

1.0 Raw material input checks including documented rovingcounts and mat audits

2.0 Production line speeds3.0 Die temperatures4.0 Resin audit prior to accepting delivery into pump station5.0 Visual inspections6.0 Dimensional inspections

The intent of the inspections is threefold. One, document that theparts are in compliance with the standards as set forth between themanufacturer and the customer. Two, document that the part isbeing manufactured per the engineering specifications. Three, cre-ate a production log so that the values can be statistically evalu-ated for continuous process improvement metrics.

The IPI documents are reviewed by a Quality Technician, for accu-racy and completeness, prior to being scanned and saved in elec-tronic format for future reference.

Visual InspectionsIt is crucial to establish the visual requirements necessary for theapplication and for the fitness for use and to convey these require-ments to the pultuder. ASTM D 4385 Standard Practice for Classi-fying Visual Defects in Thermosetting Reinforced Plastic PultrudedProducts is an excellent reference document and can be used toestablish documented visual standards. Of course, it is always agood idea to modify the document as required depending upon thefitness for use. The standard allows for three levels of visual clas-sification, with level one being very stringent and level three beingthe least stringent.

Standard visual tolerances should be built around the ASTM D4385 standard. This standard needs to be clearly communicated tothe plant floor. For Example, each production line contains a binderthat houses the visual standards, in power point format, with defi-nitions and color photos. An example is presented in IllustrationOne.

Illustration One

Visual inspections are also critical for evaluating pultruded sec-tions for strength reducing defects. One such inspection is per-formed visually on each cut section. The cut section should beevaluated for possible delaminations, cracks, dry fiber, and pin-holes to name a few.

Illustration Two depicts a delamination or separation of reinforce-ment plies. Delaminations are critical to the structural behavior ofthe profile and are cause for rejection. Delaminations are also de-tected via visual inspection of the interior and exterior of the pro-file. Delaminations can show up as elevated sections or large blis-ters on the surface of the profile.

Illustration Two

All end cuts and 100% of the exterior surfaces of the profiles arevisually inspected. The Interior of profiles are visually inspectedwith the aid of a light.

Dimensional InspectionsAs with the visual inspections, dimensional requirements are alsocrucial and should be established dependent upon the fitness foruse. Pultruders typically publish their documented dimensionalstandards on their web site and in product literature. The majorityof the industry has standardized on ASTM D3917 as the governingminimum dimensional standard. Dimensional standards should be


developed based upon ASTM D3917, your pultruder’s experience,and the part application.

In-process dimensional checks are required to be performed asoutlined in the internal pultrusion manufacturing procedure. Spe-cifically, QA toleranced drawings along with engineering and vi-sual specifications are required to be on the production floor at thepultrusion machine prior to and during production. The drawingsare incorporated into an IPI inspection document as detailed inIllustration Three. The documented inspections are to be completedbetween every third and fourth hour per twelve hour shift. Theoperator is responsible for the quality of every part and to docu-ment that inspections are being performed and that the pultrusionis being manufactured within the set QA parameters. IllustrationFour depicts a wall thickness check of a square tube with calibrateddial calipers that have been calibrated per the documented calibra-tion procedure. Random documented audits are performed by QApersonnel in order to verify that the line operators are performingtheir measurement and inspections per the documented procedures.

Illustration Three

Illustration Four

Strength of Materials InspectionMost pultruders convey their products mechanical and physicalproperties via a pultrusion design manual and product data sheets.Currently there are no set standards for deriving published values.Some pultruders publish values based on two or three standarddeviations subtracted from the mean value based on a set numberof coupon values. No formal standard exists that describes to theengineering community or to a pultruder how the data is to bedetermined and reported. For this reason, it is necessary for thecustomer and designer to understand how a pultruder derived theirpublished design values.

Both custom and standard pultruded shapes require minimum de-sign values to be published which have been agreed upon by boththe customer and pultruder. It is essential that a QA system be builtaround the design requirements of the finished product. For ex-ample, both coupon and full section testing should be performed todetermine the minimum design values. The data should be derivedfrom multiple production runs and should contain a sufficient num-ber of data points. The number of data points may range from 25 to50. Typically, two to three times the standard deviation is sub-tracted from the mean in order to generate the minimum publisheddesign values. The manufacturer should then monitor the mechani-cal property values via coupon and full section testing at someagreed upon interval. As an example, at approximately 5,000 feetintervals, samples are pulled from the pultrusion line, tested andcompared to the minimum agreed upon values. The test results canbe used for both QA purposes, as well as, continuous processdevelopment tools.

An example of some common QA tests can be viewed in Illustra-tions Five and Six. Illustration Five depicts a full scale beam testedto failure. Specifically, a pultruded beam is positioned in a full sec-tion test fixture, and loaded in a three point bend scenario. Loadversus deflection measurements are recorded until failure. The load/deflection values are utilized to verify that the minimum stiffnessvalues are being acgieved. The failure load is utilized to verify thatthe beam meets the minimum strength requirements. Illustration Sixdepicts a full section test being performed on a connection. Thefull section test verifies that the minimum hole bearing strength is


Whether your project requiresnew construction or retrofit, stan-dard products or custom solu-tions, Shepherd Tower Compo-nents are a perfect fit.

· PVC Coated Hanger Grids· Stainless Steel Hanger Grids· Gull Wing Splash Fill Slats· V-Bar Splash Fill Slats· Film Pack· Drift Reduction Units· Nozzles & Accessories

C. E. Shepherd Company, L.P.2221 Canada Dry Street

Houston, TX 77023Telephone: 713.924.4300

Fax: 713.928.2324www.ceshepherd.com

[email protected]

Shepherd Standard high quality products for cooling towers include:

Since 1957, our primary business has been innovation!We encourage inquiries for custom product solutions!


being achieved. In addition to the full section hole bearing test,coupon level testing is performed to verify the lengthwise com-pression strength and modulus of elasticity. Both values are criti-cal to the structural behavior to cooling tower structures.

Illustration Five

Illustration Six

ASTM D 7290-06 initiativeSome specifications, evolving from the LRFD initiative, are leaningtowards the adoption of the ASTM D 7290-06 Standard Practice forEvaluating Material Property Characteristic Values for PolymericComposites for Civil Engineering Structural Applications.ThisASTM-endorsed procedure describes the statistical method re-quired to provide, the characteristic value material property, repre-senting the 80 % lower confidence bound on the 5th-percentilevalue of a specified population. Characteristic values determined

using this standard practice can be used to calculate structuralmember resistance values in design codes for composite civil engi-neering structures and for establishing limits upon which qualifica-tion and acceptance criteria can be based. The adoption of thisspecification will go a long way in standardizing the method usedby pultruders to establish minimum design values. The D7290-06statistical method weighs heavily on the number of specimens testedand the coefficient of variation (COV) between test specimens.Therefore, the tighter the COV, the less the reduction factor appliedto the mean property value. Pultruders with good process control,thus less variation in material properties, will be rewarded by pub-lishing higher values than those pultruders with less process con-trol.

The adoption of the standard will encourage the pultrusion indus-try to strive for continuous process improvement. The end resultwill be the output of shapes that will be more efficient and costcompetitive against wood, concrete and steel.

Fabrication InspectionFabrication inspection is very similar to the pultrusion inspectionprocess noted above. The initial fabricated profile is put through aFirst Article Inspection as outlined in the written procedure. Thedocumented inspection is required to be signed by the Fabricatorand the Fabrication Supervisor releasing the part to productionstatus. Then, during the fabrication process, a minimum of threedocumented inspections are required per shift. The first and lastpart fabricated requires a documented inspection. The Fabricator isresponsible for the quality of each part fabricated and should per-form more inspections than are required to be documented. Thedocumented inspections are reviewed by a QA technician, for com-pleteness, prior to being saved in the internal database for futurereference. Note that random documented inspections are performedby QA technicians in order to provide oversight to the fabricationinspection process.

Construction Site Receiving InspectionThe Pultrusion manufacturer, in conjunction with the field-erectedcooling tower provider, has developed a Quality Assurance/In-spection training program for the construction field superinten-dents. The program involves arming the field superintendents withmanuals that describe possible visual and mechanical defects as-sociated with pultrusion. The superintendents know exactly whatto look for when receiving a shipment of pultruded profiles. Thepultrusions are visually inspected to ensure, one last time, that theproducts meet the standards prior to becoming a permanent mem-ber in a cooling tower structure.

In the event a substandard section is found, either due to manufac-turing defects or shipping damage, a well trained construction su-perintendent can make the call to either repair the member or re-place it. Your inspection training program/inspection manual shouldinclude definitive information regarding which types of defects arefiled repairable, as well as, a step by step repair procedure for cor-recting the defects. The manual should also include a repair autho-rization form, to be completed by the cooling tower provider, and acorrective action plan in order to document the issue and to ensurethat actions have been instated to keep similar issues from arisingagain.


ConclusionThe pultrusion manufacturing process, like many others, can bedescribed as a combination of engineering and art, derived fromyears of manufacturing experience. As time progresses and moreengineering and process controls are applied to everyday produc-tion, the art portion will continue to diminish and sometime in thefuture will account for less of the pultrusion manufacturing disci-pline. But, there is no substitute for experience in developing andmaintaining a quality FRP pultrusion product.

A quote that I picked up from the book THE TOYOTA PRODUCTDEVELOPMENT SYSTEM has stuck in my mind. “A companythat cannot standardize will struggle to learn from experience andwill rely on little more than undocumented hearsay and a widerange of opinions among its employees only to eventually dis-cover that it has been here before.” I relate this statement to thepultrusion industry as a whole. Although the industry uses manyASTM specifications, and I’m sure every pultruder has their ownQA/QC systems and procedures, the pultrusion industry has a

great need to adopt standards that will place everyone on a levelplaying field.

The LRFD initiative, along with the adoption of the ASTM D 7290-06, will allow our industry to mature and will allow for the designand construction of more efficient and economical cooling towerstructures.

The Cooling Technology Institute will be well advised to monitorand adopt some of the procedures and standards that will come asa result of the pultrusion industry’s lead in the Load & ResistanceFactor Design initiative.

The general description of the QA/QC program between this majorpultrusion company, and one of its major customers for structuralFRP components is intended to get pultruders, designers, and own-ers thinking about the importance of an agreed upon manufactur-ing and design standard, as well as, an agreed upon QA and In-spection Program. As we see more and more pultruded parts beingshipped in from new overseas suppliers, it is crucial that allpultruders be held to the same high quality standards.


Cooling Technology InstituteLicensed Testing Agencies

For nearly thirty years, the Cooling Technology Institute hasprovided a truly independent, third party, thermal performancetesting service to the cooling tower industry. In 1995, the CTIalso began providing an independent, third party, driftperformance testing service aswell. Both these services areadministered through the CTIMulti-Agency Tower Perfor-mance Test Program and providecomparisons of the actual operat-ing performance of a specifictower installation to the designperformance. By providing suchinformation on a specific towerinstallation, the CTI Multi-Agency Testing Program standsin contrast to the CTI CoolingTower Certification Programwhich certifies all models of aspecific manufacturer's line of cooling towers perform inaccordance with their published thermal ratings.

To be licensed as a CTI Cooling Tower Performance Test

Licensed CTI Thermal Testing AgenciesLicense Agency Name Contact Person TelephoneType* Address Website / Email Fax

A,B Clean Air Engineering Kenneth Hennon 800.208.61627936 Conner Rd www.cleanair.com 865.938.7569

Powell, TN 37849 [email protected]

A, B Cooling Tower Technologies Pty Ltd Ronald Rayner 61 2 9789 5900PO Box N157 [email protected] 61 2 9789 5922

Bexley North, NSW 2207AUSTRALIA

A,B Cooling Tower Test Associates, Inc. Thomas E. Weast 913.681.002715325 Melrose Dr. www.cttai.com 913.681.0039

Stanley, KS 66221-9720 [email protected]

A, B McHale & Associates, Inc Thomas Wheelock 865.588.26546430 Baum Drive www.mchale.org 425.557.8377

Knoxville, TN 37919 [email protected]

* Type A license is for the use of mercury in glass thermometers typically used for smaller towers. Type B license is for the use of remote data acquisition devices which can accommodate multiple measurement locations required by larger towers.

Licensed CTI Drift Testing AgenciesAgency Name Contact Person Telephone

Address Website / Email Fax

Clean Air Engineering Kenneth Hennon 800.208.61627936 Conner Rd www.cleanair.com 865.938.7569

Powell, TN 37849 [email protected]

McHale & Associates, Inc. Thomas Wheelock 865.588.26546430 Baum Drive www.mchale.org 425.557.8377

Knoxville, TN 37919 [email protected]

Agency, the agency must pass a rigorous screening process anddemonstrate a high level of technical expertise. Additionally, itmust have a sufficient number of test instruments, all meetingrigid requirements for accuracy and calibration.

Once licensed, the Test Agenciesfor both thermal and drift testingmust operate in full compliancewith the provisions of the CTILicense Agreements and TestingManuals which were developedby a panel of testing expertsspecifically for this program. In-cluded in these requirements arestrict guidelines regarding conflictof interest to insure CTI Tests areconducted in a fair, unbiasedmanner.

Cooling tower owners and manu-facturers are strongly encouraged

to utilize the services of the licensed CTI Cooling TowerPerformance Test Agencies. The currently licensed agencies arelisted below.


As stated in its opening paragraph, CTI Standard 201... “sets forth a pro-gram whereby the Cooling Technology Institute will certify that all modelsof a line of water cooling towers offered for sale by a specific Manufacturerwill perform thermally in accordance with the Manufacturer’s published rat-ings...” By the purchase of a “certified” model, the User has assurance thatthe tower will perform as specified, provided that its circulating water is nomore than acceptably contaminated-and that its air supply is ample andunobstructed. Either that model, or one of its close design family members,will have been thoroughly tested by the single CTI-licensed testing agencyfor Certification and found to perform as claimed by the Manufacturer.

CTI Certification under STD-201 is limited to thermal operating conditionswith entering wet bulb temperatures between 12.8°C and 32.2°C (55°F to90°F), a maximum process fluid temperature of 51.7°C (125°F), a coolingrange of 2.2°C (4°F) or greater, and a cooling approach of 2.8°C (5°F) or

greater. The manufacturer may set more restrictive limits if desired or publish less restrictive limits if the CTI limits are clearly defined andnoted in the publication.

Following is a list of cooling tower models currently certified under STD-201. They are part of product lines offered by Advance GRP(Advance) Cooling Towers, Pvt, Ltd.; Aggreko Cooling Tower Services; Amcot Cooling Tower Corporation; AONE E&C Corporation Ltd;Baltimore Aircoil Company, Inc.; Delta Cooling Towers, Inc.; Evapco, Inc.; Fabrica Mexicana De Torres, S.A.; HVAC/R International, Inc.;Imeco, div of York International; Ltd; KIMCO (Kyung In Machinery Company, Ltd.); King Sun Industry Co. Ltd.; Liang Chi IndustryCompany, Ltd.; Marley (SPX Cooling Technologies); Mesan Cooling Tower, Ltd; Nihon Spindle Manufacturing Company, Ltd.; Polacelb.v.; Protec Cooling Towers; RSD Cooling Towers; Ryowo (Holding) Company, Ltd; Ta Shin F.R.P. Company, Ltd.; The Cooling TowerCompany, L.C; The Trane Company; Tower Tech, Inc; and Zhejiang Jinling Refrigeration Engineering Company who are committed to themanufacture and installation of full-performance towers. In competition with each other, these manufacturers benefit from knowing thatthey each achieve their published performance capability. They are; therefore, free to distinguish themselves through design excellenceand concern for the User’s operational safety and convenience.

Those Manufacturers who have not yet chosen to certify their product lines are invited to do so at the earliest opportunity. You cancontact Virginia A. Manser, Cooling Technology Institute, PO Box 73383, Houston, TX 77273 for further information.

Cooling Towers Certified by CTI Under STD-201


Aggreko Cooling Tower Services ...... 36-37AHR Expo ............................................... 70Amarillo Gear Company ....................... IBCAmcot Cooling Tower ............................. 19American Cooling Tower, Inc. ................ 33AMSA, Inc. ............................................ 13Bailsco Blades & Casting, Inc. ............... 25Bedford Reinforced Plastics ................... 15Brentwood Industries ............................. 43ChemTreat, Inc. ....................................... 49CleanAir Engineering .............................. 31CTI Certified Towers .......................... 62-66CTI License Testing Agencies ............... 58CTI ToolKit ........................................ 68-69Composite Cooling Solutions, LP ........... 55Cooling Tower Resources ....................... 71DesJardins & Associates .......................... 6Dynamic Fabricators ................................. 7Fibergrate Composite Structures ............ 17Gaiennie Lumber Company ....................... 2Gas Turbine Users Symposium (GTUS) . 61GEA 2H Water Technologies .................. 67Glocon ................................................... 3, 9Howden Cooling Fans .............................. 5Hudson Products Corporation ............... 39IMI Sensors a PCP Piezotronics Div ...... 41Industrial Cooling Towers ................ IFC, 4McHale & Associates ............................. 23Midwest Towers, Inc. ............................. 21Moore Fans ............................................ 45Paharpur Cooling Towers Limited .......... 51Power Gen ............................................... 27Rain for Rent ........................................... 11Research Cottrell .................................... 56Rexnord Industries ................................. 35C.E. Shepherd Company, LP ................... 53Spraying Services, Inc. ........................... 57SPX Cooling Technologies ................. OBCStrongwell ............................................... 59Structural Preservation System .............. 29Swan Secure Products, Inc. .................... 47Tower Performance, Inc. ......................... 72

Index of Advertisers

CTI Journal, Vol. 30, No. 2 · 4 CTI Journal, Vol. 30, No. 2 View From The Tower Denny Shea Welcome...

Documents

Transcript of CTI Journal, Vol. 30, No. 2 · 4 CTI Journal, Vol. 30, No. 2 View From The Tower Denny Shea Welcome...