Item No. 24229 NACE International Publication 11106

This Technical Committee Report has been prepared by NACE International Task Group 152* on Cooling Water Systems: Monitoring and Control

Monitoring and Adjustment of Cooling Water Treatment Operating Parameters

© May 2006, NACE International

This NACE International technical committee report represents a consensus of those individual members who have reviewed this document, its scope, and provisions. Its acceptance does not in any respect preclude anyone from manufacturing, marketing, purchasing, or using products, processes, or procedures not included in this report. Nothing contained in this NACE report is to be construed as granting any right, by implication or otherwise, to manufacture, sell, or use in connection with any method, apparatus, or product covered by Letters Patent, or as indemnifying or protecting anyone against liability for infringement of Letters Patent. This report should in no way be interpreted as a restriction on the use of better procedures or materials not discussed herein. Neither is this report intended to apply in all cases relating to the subject. Unpredictable circumstances may negate the usefulness of this report in specific instances. NACE assumes no responsibility for the interpretation or use of this report by other parties. Users of this NACE report are responsible for reviewing appropriate health, safety, environmental, and regulatory documents and for determining their applicability in relation to this report prior to its use. This NACE report may not necessarily address all potential health and safety problems or environmental hazards associated with the use of materials, equipment, and/or operations detailed or referred to within this report. Users of this NACE report are also responsible for establishing appropriate health, safety, and environmental protection practices, in consultation with appropriate regulatory authorities if necessary, to achieve compliance with any existing applicable regulatory requirements prior to the use of this report. CAUTIONARY NOTICE: The user is cautioned to obtain the latest edition of this report. NACE reports are subject to periodic review, and may be revised or withdrawn at any time without prior notice. NACE reports are automatically withdrawn if more than 10 years old. Purchasers of NACE reports may receive current information on all NACE International publications by contacting the NACE FirstService Department, 1440 South Creek Drive, Houston, Texas 77084-4906 (telephone +1 281/228-6200).

Foreword

The efficient and safe operation of a cooling tower system typically involves a substantial amount of routine monitoring of chemical, physical, and microbiological phenomena. This technical committee report is intended for personnel directly responsible for daily operation and control of cooling tower systems, facility engineering and maintenance personnel, and water treatment company sales and technical staff personnel. The purpose of this report is to provide a concise compilation of what are considered common practices in this area. Monitoring and control of cooling systems generally occurs in three phases: • Initial system surveys, conducted after assuming

responsibility for the management or operation of a new or unfamiliar cooling system;

• Monitoring and adjustment of cooling tower operating parameters during a campaign of operation; and

• Inspections and measurements of the condition of a cooling tower system during off-line periods such as outages or turnarounds.

This report is specifically concerned with the second topic—monitoring and adjustment of cooling tower operation on a day-to-day basis during periods of routine operation. This technical committee report was prepared by Task Group (TG) 152 on Cooling Water Systems: Monitoring and Control. TG 152 is administered by Specific Technology Group (STG) 11 on Water Treatment, and sponsored by STG 46 on Building Systems and STG 62 on Corrosion Monitoring and Measurement—Science and Engineering Applications. It is issued by NACE International under the auspices of STG 11.

1

___________________________ *Chair Michael E. Rogers, Alberta Technology & Sciences, De Winton, Alberta, Canada and Donald A. Johnson, Nalco Energy Services, Naperville, Illinois.

NACE International NACE technical committee reports are intended to convey technical information or state-of-the-art knowledge regarding corrosion. In many cases, they discuss specific applications of corrosion mitigation technology, whether considered successful or not. Statements used to convey this information are factual and are provided to the reader as input and guidance for consideration when applying this technology in the future. However, these statements are not intended to be recommendations for general application of this technology, and must not be construed as such.

Introduction

Monitoring of cooling tower operation falls into two major categories. The first deals with the monitoring and control that is directed at maintaining cooling water chemistry within defined specifications. This includes both the components introduced by the supply of make-up water and those added to control scale, corrosion, and microbiological activity. The

second monitoring activity includes measurements that assess the severity of scaling, corrosion, or microbiological processes during a campaign of operation. The data obtained from the second category of measurements are used to refine the specifications that form the basis for the water treatment strategy.

Monitoring and Control of Water Chemistry

Concentration Factor and Blowdown Rate A cooling tower is a device that removes process heat from a stream of water through evaporation (see Figure 1). Water that is evaporated is replaced by fresh water (make-up) from some source. Typically, water used as make-up to cooling systems is taken “as-is” from the supply and contains dissolved mineral components. Because water leaving as a result of evaporation doesn’t remove any dissolved solids, these materials concentrate in the recirculating water. If the concentration of these impurities is not controlled, scale formation and corrosion problems often result. To

limit the concentration of these impurities, a portion of the cooling tower water is frequently discharged. This wastage is referred to as blowdown. Concentration factor is the term that is frequently used to quantify the degree of concentration that has occurred in a cooling tower. It is defined as the ratio of the concentration of an impurity in the cooling tower water divided by the concentration of the same impurity in the make-up water. For example, if the concentration of sodium ions in the cooling tower water is 500 mg/L and the concentration of these ions in the make-up water is 100 mg/L, then the concentration factor would be 5.

FIGURE 1: Streams in Cooling Tower Operation

Make-up (C = 1)

Blowdown (C = CH)

Evaporation C = 0

Hydraulic Concentration Factors The concentration factor is typically calculated (see Equations [1] and [2]) when the flow rates of both the make-up water and the blowdown are known. Leaks and water lost to airborne entrainment (drift) are considered part of the blowdown for purposes of this analysis.

BD

MU=CH (1)

or equivalently,

EMU

MUCH

−= (2)

2

NACE International

where: CH = hydraulic concentration factor (dimensionless)(1) MU = make-up flow rate BD = blowdown flow rate E = evaporation rate MU, E, and BD are always in the same units. If make-up and blowdown flow rates are not available, it is possible to estimate evaporation (see Equations [3] and [4]) from a cooling tower. In metric units using a typical evaporation efficiency of 80%: E = 0.132R × ∆T (3) where: R = recirculation rate (m3/min) ∆T = hot return water temperature (°C) - cold supply water temperature (°C) E = evaporation (m3/min) Or in English units: E = 0.0008R × ∆T (4) where: R = recirculation rate (gallons per minute [gpm]) ∆T = hot return water temperature (°F) - cold supply water temperature (°F) E = evaporation (gpm) Although the application of Equation (1) appears to be simple, it is often not very useful because the true blowdown flow rate is seldom known with sufficient accuracy. This is because the blowdown rate includes the cooling tower misting rate (drift) and all unintentional water losses such as pump seal leaks, splashing from the cooling tower, etc. However, in the case of cooling towers employing high-efficiency drift eliminators and little or no unintentional leakage, Equations (1) or (2) are often employed to estimate CH. Chemical Species Concentration Factors The concentration factor of an individual species (Cs) is typically determined by measuring the concentration of a species in both the make-up water and the cooling tower water (or blowdown), and then determining the ratio. See Equation (5):

M

B

S C

C=C (5)

where: CB = concentration of a species in the blowdown or tower water CM = concentration of the same species in the make-up water Care is typically used in selecting the species to measure. Problems can arise if anything alters the concentration of the selected species other than concentration due to evaporation. Examples of potential problems include the loss of calcium and carbonate caused by the precipitation of calcite (CaCO3), the loss of carbonate (CO3

-2) and/or bicarbonate (HCO3

-) by acid addition and degassing of carbon dioxide (CO2), and the addition of sulfate (SO4

-2) by the use of sulfuric acid (H2SO4) for pH control. For these reasons, the species that are most often used to determine the Cs are magnesium, chloride (when halogen is not being used for pH control, or chloride-producing biocides are not added), and sulfate (when H2SO4 is not being used for pH control). The conductivity of water is approximately proportional to the total dissolved solids concentration of the water. Conductivity is also one of the simplest and cheapest water assays. A method that is commonly used to estimate Cs is to determine the ratio of the conductivity of the blowdown (or tower water) divided by the conductivity of the make-up water. Scalant Mass Balance Calculations One method that is frequently used to estimate whether mineral deposits are forming or not is to first calculate the expected concentration (see Equation [6]) of a potential scale-forming species (such as calcium), then compare the expected value to the actual value. If the expected concentration is significantly greater than the actual concentration, mineral deposit formation is likely. The expected concentration for a species such as calcium is often determined by first determining a value for CH, then multiplying by the concentration of the species in the make-up water.

MHExpCCC ×= (6)

where: CExp = expected concentration of a potential scale-forming species CM = concentration of species in the make-up water CH is typically estimated by either the chemical concentration factor (Cs) of a nonprecipitating species such as Mg (Equation [5]) or, if the data are available, from the flow rates of the make-up and blowdown streams (Equation

3

___________________________ (1) Hydraulic concentration factor and cycles of concentration are synonymous.

NACE International

[1]). If a chemical concentration factor is used, it is usually not a species that is added to the system in any chemical treatments or removed from the system by any means other than blowdown. This method is only capable of identifying gross scaling conditions. Because the amount of scale precipitating is often only a small fraction of the total amount of scale-forming ions in the water, it is indeed possible for scaling to be occurring and not be detected analytically using this method. Control Strategies In most cases, cooling tower system operations are controlled by maintaining one or more chemical species in the circulating water between specific upper and lower limits. Species typically limited in cooling systems are calcium, silica, phosphate, alkalinity, and chloride. High levels of species such as calcium or alkalinity can create scaling conditions, and low levels of these species contribute to corrosiveness. Ions such as chloride and sulfate contribute to corrosiveness at high levels. In addition, treatment programs have defined control limits provided by the vendor. Control of ion concentrations in the circulating water is typically accomplished by controlling the blowdown rate. Once the limiting concentration of the controlling species has been identified (usually by the water service representative), it is relatively simple to determine the maximum allowable concentration factor (CH). Equations (1) or (2) are used to determine the approximate blowdown rate that is used to achieve the desired CH. The blowdown rate is either set manually or controlled automatically. A common practice is to use the control capability of commercially available conductivity analyzers to open or close a blowdown valve. pH and Alkalinity Monitoring and Control In some cases, cooling tower operation involves some form of pH control. The tower circulating water pH is typically a factor in determining the corrosiveness or scale-forming tendency of cooling tower water. Cooling Water Acid/Base Chemistry Most natural water supplies contain significant amounts of basic or alkaline species such as carbonate (CO3

-2), bicarbonate (HCO3

-), hydroxide (OH-), phosphate (PO4-3),

silicate (SiO2-2), or borate (BO4

-3). Of these, the most common species are the CO3

-2 and HCO3- species.

In most water sources, the acid/base chemistry is dominated by the carbonate species. The concentrations of CO3

-2 and HCO3- are determined in the field by simple

titration and are represented by a term known as alkalinity. Alkalinity is defined as the acid-neutralizing capacity of

water. In cooling water systems, the two important alkalinity measurements are: • “M” (for methyl orange) or total alkalinity: The amount

of acid typically used to reduce the pH of a water sample to 4.3, the methyl orange endpoint.

• “P” (for phenolphthalein) alkalinity: The amount of acid typically used to reduce the pH of a water sample to 8.3, the phenolphthalein endpoint.

These values are often reported in units of mg/L or ppm as calcium carbonate (CaCO3). In cooling water systems, the “M” or total alkalinity is the more useful measurement. It is defined by Equation (7): M alkalinity = [HCO3

-] + 2 × [CO3-2] (7)

The total alkalinity is a useful control parameter because it is conserved in cooling water systems, provided that no mineral acid or base is added to the system. This means that it follows the same concentration relationship (see Equation [5]) as other ionic species in cooling water. This is not true of the “P” alkalinity and certainly not true of the pH. Thus, a cooling tower using make-up water with 100 mg/L of total alkalinity (as CaCO3) and operating at 5 hydraulic cycles has recirculating water with 500 mg/L of total alkalinity. Cooling tower system acid/base chemistry is controlled by the following reactions, shown in Equations (8), (9), and (10): CO2 + H2O ↔ H2CO3 (8) H2CO3 ↔ HCO3

- + H+ (9) HCO3

- ↔ CO3-2 + H+ (10)

These are all equilibrium reactions that are controlled by the pH of the cooling water. In turn, the pH of these systems is, to a large extent, controlled by the amount of CO2 that is dissolved in the cooling water. CO2 can either transfer from air to cooling water or transfer from cooling water to air in a cooling tower. This transfer of CO2 influences the pH and “P” alkalinity, but not the “M” alkalinity. If (as is usually the case) CO2 is removed or stripped from the tower water, the pH increases while the “M” alkalinity remains constant. If CO2 is absorbed from the atmosphere to the tower (e.g., in the case of high-pH lime-softened water), the pH decreases, again with the “M” alkalinity remaining constant. Because this process of mass transfer is controlled by variables such as CO2 concentration in the air, water temperature, air temperature, and water pH, it is very difficult to predict. For that reason, an accurate relationship between pH and cooling water alkalinity is not readily available. Every tower has a slightly different pH vs. alkalinity profile. Some equations to estimate the pH vs. alkalinity relationship empirically have been published by Kunz1 and Caplan.2

4

NACE International

The Kunz equation (Equation [11]) is: H =1.6 × LOG (“M” Alkalinity) + 4.4 (11) The Caplan relationship (Equation [12]) is: pH = 1.718 × LOG (“M” Alkalinity) + 4.133 (12) The addition of strong mineral acids such as H2SO4, hydrochloric (HCl), nitric (HNO3), or others do modify the “M” alkalinity. These acids reduce the total alkalinity in proportion to their concentration and molecular weight. Cooling Tower Operation With and Without pH Control Cooling towers are either operated with or without pH control. The latter mode is sometimes described as “alkaline” operation. The two modes of operation use different approaches to limiting the operating pH of a cooling tower and implement different control strategies. In the pH-controlled mode of operation, acid (usually H2SO4 or HCl) is normally used to control either the pH or the alkalinity of cooling water. When mineral acid is added to cooling water, CO2 is formed from CO3

-2 and HCO3- ions

and stripped from the cooling water by air, effecting a reduction in both the pH and the alkalinity. Because the relationship between pH and alkalinity is not completely predictable, it is only possible to control one of these parameters accurately. Some water treatment programs are managed by controlling the pH of the cooling water while other types of programs are managed by controlling the alkalinity. In general, cooling tower scale/corrosion inhibitor packages designed for pH-controlled operation specify an operating range based on pH, while those designed for pH-free operation specify an acceptable operating range based on alkalinity. Other methods of specifying pH or alkalinity control ranges based on CaCO3

supersaturation are also sometimes employed. In this case, the cycled-up tower water alkalinity is used as a basis for determining conductivity set-points. The pH of the cooling water is commonly directly controlled by controlling the rate of acid addition using an on-line pH controller to adjust the acid feed. Control based directly on alkalinity is used less because instrumentation for the on-line monitoring of alkalinity is more expensive and complex. It is common to employ indirect means such as controlling the acid flow rate (or in other cases, acid addition time) based on off-line (grab sample) alkalinity measurements. In the non pH-controlled or alkaline mode of operation, no acid is added to the system. Instead, the alkalinity is allowed to concentrate along with the other species in the system. The pH reaches a consistent equilibrium value, depending on the cycled-up alkalinity and the other aforementioned parameters. In the alkaline mode, the pH and alkalinity are limited by blowdown cycle control along with the other ions in the tower. Control limits in alkaline mode operation are usually defined in terms of maximum alkalinity rather than maximum pH because alkalinity tracks the concentration factor of a tower in a linear fashion, while pH is difficult to predict precisely. On-Line pH Measurement and Control Figure 2 shows a generalized pH control system. Note that two pH probes are shown. The signal from one of the pH probes is used to control the acid feed while the signal from the other probe is tied into the alarm and shutdown circuit. If a low pH is detected by either probe, then an alarm is sounded, and the acid feed system is shut down until it is manually reactivated. The purpose of the redundant pH monitoring is to avoid a low pH excursion and severe corrosion problems that frequently result.

FIGURE 2: Schematic of typical pH control system

pH Probe(Controller)

Acid FeedSystem

pH Probe(Shut-Down)

AcidTank

Acid Feed System

pH Probe (Controller)

Acid Tank

pH Probe (Shut-Down)

Acid addition is typically controlled by a solenoid-activated pump. A similar system is sometimes used to control caustic addition when the pH and alkalinity need to be

increased. pH probes are regularly cleaned and calibrated in order to be reliable. Comparison of the readings of the two probes can indicate probe malfunction.

5

NACE International

Inhibitor Monitoring and Control Inhibitors, dispersants, and microbiocides are added to cooling water systems to control scaling, corrosion, particulate fouling, and microbiological activity. These inhibitors have typically defined working ranges of concentration within which they are effective and economical. The task of the cooling tower operator is to maintain and operate the inhibitor dosing system and conduct water tests to ensure that the inhibitors are maintained at the specified concentration in the tower water. The majority of inhibitor packages are sold in liquid form, and a metering pump for accurate dosing is typically used.

There are many water treatment inhibitor packages available commercially. The selection and configuration of an inhibitor program is a process that involves analysis of many factors and is beyond the scope of this report. An appropriate treatment chemical package is often selected in consultation with a qualified water treatment vendor. As part of this process, a set of control limits on various components of the package is typically specified. For reference purposes, Table 1 lists some of the commonly used inhibitor components. Biocides are treated differently and are discussed later in this report.

Table 1: Cooling Water Corrosion and Scale Inhibitors

Corrosion Inhibitors Scale Inhibitors and Dispersants Orthophospate and polyphosphate salts Organic phosphorous compounds Organic phosphorous compounds (e.g., phosphonates) Lignins and tannins Molybdate Polymers, copolymers, complex polymers Zinc (Zn) salts Polyphosphate salts Sodium nitrite Copper inhibitors (e.g., azoles) Proprietary organic compounds

With the inhibitor dosing parameters established and the cooling system in operation, it typically becomes the joint responsibility of plant personnel and the treatment vendor to monitor and control the tower operating parameters and chemical dosages within the established control ranges. Most of the corrosion and scale inhibitors used in modern cooling water systems function best within prescribed dosage ranges. System startup or upset conditions often call for different dosing strategies. However, continued operation above or below specified ranges can lead to a range of operating problems. Effects of Underdosing The protective films formed by common corrosion inhibitors for steel, including phosphates, phosphonates, molybdate, and zinc, are stable only in the presence of a sufficient concentration of the active inhibitor in the water. If the inhibitor level drops too low, the film tends to dissolve. Dosage variations that are quickly corrected do little harm, but repeated excursions or continuous low-level dosing lead to loss of corrosion protection. When this happens, some corrosion can occur and corrosion products form on the metal surface. These deposits, in turn, can interfere with reestablishing the protective film, even if the corrosion inhibitor is raised to a higher than normal level. Corrosion can continue beneath the new deposits, and this mechanism is often the beginning of future under-deposit attack corrosion problems. A similar problem appears when mineral scale inhibitors, such as phosphonates and polymers, are underdosed. These compounds increase the delayed precipitation of scaling minerals such as CaCO3 and calcium sulfate

(CaSO4). When scale inhibitors are underdosed or not fed, these minerals (calcium carbonate and sulfate) form as scales on heat transfer surfaces. Once formed, these scale deposits cannot be removed by normal dosages of scale inhibitors. Very high dosages of inhibitors over long periods of time or severe chemical cleaning (lower pH) procedures are typically used to remove mineral scale deposits. As with any deposit, mineral scales can become sites for under-deposit corrosion. Most mineral scales form on hot metal surfaces and, because of their low thermal conductivity, act as insulators and greatly reduce the efficiency of heat transfer. Effects of Overdosing As with underdosing, single-point overdoses that are quickly corrected have little effect. However, repeated or prolonged overdosing of corrosion and scale inhibitors can create a new set of problems. The most obvious problem is the unnecessarily increased cost of chemical treatment and increased environmental impact. In addition, undesirable chemical reactions can occur. At high concentrations, Zn ions can precipitate as zinc hydroxide to form a nonprotective deposit. Some phosphonates react with calcium ions in the water to form a nonprotective calcium phosphonate precipitate or complex. Higher than normal levels of inorganic phosphates greatly increase the danger of forming calcium phosphate scale. At high dosages, anionic polymeric dispersants can react with cationic biocides to form precipitates that remove both the biocide and the dispersant from solution.

6

NACE International

Mechanical Methods of Inhibitor Dosage Control The usual objective of inhibitor dosage control is to maintain the level of active corrosion and scale inhibitors in the circulating cooling water within the control parameters set by the water treatment vendor. Several different methods are available for accomplishing this control. However, all of these methods operate on the same principle, the replacement of inhibitor lost to chemical consumption due to film maintenance and physical removal by system blowdown. In some small cooling tower systems with moderate heat loads, cycles of concentration are simply controlled by conductivity, and chemical injection pumps are set to provide relatively high levels of inhibitors in the system. Service is usually provided entirely by the water treatment vendor. This method, although simple and relatively inexpensive, is common only for noncritical systems. More critical systems providing steady-state environments, cooling for data processing operations, and process control often employ more precise and reliable control of inhibitor dosing rate. Chemical pumps are controlled based on one or more of the following: • Blowdown flow; • Make-up water flow; • Analytical tests of the circulating water; and • Monitored performance parameters. The pump delivery rates are adjusted as needed to maintain the desired chemical levels in the water. Alternative methods for accomplishing this control are discussed below. Chemical testing is discussed in a later section. Feed Based on Blowdown This method, commonly called the “bleed and feed” method, typically involves a conductivity controller with outputs to drive one or more chemical pumps. The conductivity set point is defined by the desired cycles of concentration in the circulating water. When the conductivity rises above this point, the blowdown solenoid valve opens, and the chemical pumps operate. When the conductivity drops below the set point, the solenoid valve closes, and the chemical pumps stop. Chemical levels in the water are determined analytically by methods described later, and the pumping rates are adjusted to raise, lower, or maintain the desired levels. “Bleed and feed,” as this method is often called, is the simplest, least expensive, and most commonly used way to control chemical feed to cooling water systems. It is also the least reliable method. Anything that interferes with operation of the conductivity controller or the blowdown solenoid valve also interferes with proper chemical feed. When blowdown begins, make-up water typically does not flow until the water level in the basin drops sufficiently to

activate the level control device (either a simple float or an electronic control). Because the water level is controlled by evaporation as well as blowdown, make-up water flows without blowdown, causing chemical levels to vary. With careful testing and incremental adjustment of pumping rates, these variations can usually be kept within the established control parameters. In many cases, this is all that is used. Feed based on blowdown is applicable only in systems with consistent make-up water quality. Feed Based on Make-up A more precise way to control chemical levels in a cooling tower system is to feed based on make-up water flow. This method ensures that all make-up water entering the tower is treated, regardless of blowdown, and that no treatment is fed when make-up water is not flowing. Make-up control of chemical feed helps to eliminate the overfeed and underfeed problems that are inherent in the “bleed and feed” method based on conductivity control. Control of chemical feed based on make-up water flow is often accomplished using either a flow meter or a totalizing water meter. The simplest method is to include a sensing flow meter that sends an electrical signal whenever water is flowing. This signal is used by the tower controller to activate one or more chemical pumps. As in the “bleed and feed” method, chemical levels are measured analytically and the pump stroke is adjusted to maintain the chemical dosage within range. This method, although it is about as simple as the “bleed and feed” method, is not often used. The cost is higher, the flow meter needs maintenance, and results are not as favorable as those achieved with a totalizing water meter. A totalizing make-up water meter is usually set to send an electrical impulse to a controller every time a preset volume of water passes through the meter. This impulse is then used by the controller to trigger one or more chemical pumps that operate for a preset time. With a totalizing water meter, there are three methods available for adjusting the chemical feed rate: 1. The amount of make-up water flow used to trigger the chemical pumps; 2. The pump stroke—that is, the amount of chemical delivered per stroke; and 3. The time that the pump operates during each delivery sequence. The water meter is sized for the system, and the pulse frequency is not normally adjustable. Properly sizing the meter and incrementally adjusting the pump stroke and timing normally keeps inhibitor dosage levels well within control ranges. Barring sudden changes in operating load or timing, frequent adjustments in pump parameters are not typically made. As a further refinement of totalizing make-up water meter control, special chemical pumps that deliver an accurate

7

NACE International

preset volume of chemical with each pump cycle are available. This method eliminates the need for pump stroke and timer adjustments and typically results in a constant rate of chemical delivery based on make-up water flow. As with other methods, analytical tests of the circulating water are used to set the chemical delivery rate. This is the most precise method for feeding chemicals based on make-up water flow. It is readily adaptable to automated control, and it is also the most expensive method. Accurate feeding by this method depends on accurate control of concentration cycles. Feed Based on Performance Indicators Another method of on-line control utilizes fouling simulators, corrosion detectors, and performance parameters such as oxidation reduction potential (ORP) in combination with make-up flow(s) and other on-line nonperformance monitors such as pH and conductivity, plus product detection analyzers. These analyzers are configured to incorporate algorithms that adjust the feed of corrosion inhibitors, pH adjustment chemicals, deposit inhibitors, and microbiocides. Feed forward and feed back trims are also incorporated to achieve the desired cycles of concentration, as well as fouling and corrosion set points. A difficulty with the use of this method is that effects of dosage changes on corrosion and scaling are typically quite slow, introducing significant lag into the control. The algorithms contain effective compensation for this lag time, as well as a verification function of the nonperformance-based monitor’s data acquisition. Test devices used in connection with these systems typically measure the corrosion or fouling tendency of water on a simulated surface or probe. The assumption is that the simulator is configured and operated to reflect the behavior of the water in the production equipment. This assumption is not always valid. There are currently no accepted industry standards in this area. Chemical Methods of Inhibitor Dosage Control Monitoring and control of corrosion and fouling inhibitor levels in cooling water systems typically involve some analytical testing of the circulating water. The method of analysis has traditionally been based on manual laboratory techniques. However, much progress has been made in the development of automated, on-line chemical analysis techniques. The results of these tests are used—either directly or indirectly—to control the feed of chemicals to the water. It is therefore very valuable to know exactly what is being tested, how the results are expressed, and what the data mean in terms of levels of total product and active components in the water. These topics are discussed in this section of the report. Tracers and Active Ingredients Tracer chemicals, intended only for dosage rate measurement and control, are sometimes added to water treatment products. Ideally, tracer chemicals are soluble

over the range of cooling water conditions, and are inert and unreactive in the system. That is, they do not precipitate or in other ways react with other water treatment chemicals or components of the water. They do not take part in any corrosion- or scale-inhibiting reactions, and are not attacked by oxygen, chlorine, or other aggressive chemicals. Obviously, simple and reliable analytical methods are available for any tracer chemical. A chemical that meets these conditions is sometimes used to control the dosage rate of scale or corrosion inhibitor formulations. The water treatment vendor often establishes a control range for this chemical. Examples of such tracers are low levels of molybdate, vanadate, and fluorescent molecules used in on-line control. Molybdate, usually in the form of sodium molybdate, is also used as a corrosion inhibitor. There is often a distinction between the measurements of tracers and active ingredients. That is, the tracer measures the amount of total product fed into the water. It does not measure the remaining amount of any specific active component of the product. Active ingredients in a formulation can precipitate, will be absorbed on surfaces, chemically degraded, or otherwise consumed in the cooling system. To the extent that these processes occur, there may be differences between the “theoretical” active ingredient concentration determined by the tracer and the actual levels seen by chemical analysis. If the differences are excessive, it can indicate a performance problem. If tracer chemicals are used for routine monitoring and control of treatment chemical levels in cooling water, these analyses are supported by periodic analyses for active components. This activity is normally performed by water treatment vendor personnel during regular service visits. On-Site Water Analyses To be useful in practical field situations, analytical tests are typically fast, simple, and usable by personnel who are not trained chemists. These tests are reliable over a range of concentrations normally found in circulating cooling water, and in the presence of common dissolved and suspended impurities. These tests are used to control the concentrations of both chemical species in the water and active materials that are often added for corrosion and mineral scale control. Reagents and procedures for field water analysis are typically provided by water treatment vendors. They are also available from other sources such as those outlined by L.S. Clescerl, et. al.3 Units of Expression When interpreting the results of water analysis, it is important to remain aware of the units used to express concentrations. It is common in the water treatment industry to represent concentrations in units of milligrams/liter (mg/L) or parts per million (ppm). Mg/L is a

8

NACE International

unit of mass per unit of volume, while ppm is a unit of mass per unit of mass. These two units are equivalent only in dilute water solutions where the density of the solution is very close to 1, as in most cooling water systems. It has become common practice to represent all forms of hardness and alkalinity (Ca+2, Mg+2, HCO3

-, CO3-2) in units

of mg/L or ppm of CaCO3. This practice dates from the early days of water treatment, when a common unit of measure was needed to express these concentrations for water softening calculations. For example: The concentration of calcium in cooling tower water is usually represented as mg/L of Ca+2 or as mg/L of CaCO3. The equivalent weight of the Ca+2 ion is 20 (atomic weight [at. wt.] divided by charge 40 ÷ 2). The equivalent weight of CaCO3 is 50.05 (molecular weight divided by charge of components 100.1 ÷ 2). A water sample that contains 100 mg/L of calcium represented as Ca+2 is equivalently represented as having 250 mg/L of calcium represented as CaCO3. Similarly, ionic concentrations of other species of hardness and alkalinity are often presented in units of mg/L as CaCO3. For example, a solution containing 100 mg/L of Mg+2 with the units represented as (Mg+2) can also be said to contain 410 mg/L of Mg+2 represented as CaCO3. The purpose of this conversion calculation is to put the components of hardness and alkalinity on a common “apples-to-apples” basis, so that they are additive. Thus, a water containing 100 mg/L of Ca and 100 mg/L of Mg, each expressed as CaCO3, has an additive “total hardness” of 200 mg/L. If the Ca and Mg concentrations were expressed in mg/L as the ions, the concentrations would not be additive. In a similar way, the concentrations of all the ionic species in a water sample are expressed as equivalents per liter in order to test the accuracy and completeness of the analysis (the total concentrations of cations and anions in the water are equal.) Tables of water analysis conversion factors are readily available in water treatment handbooks. A similar principle applies to the expression of water treatment formulation concentrations. When formulation concentrations are reported, it is usually in terms of mg/L of the formulation. However, these numbers are based on the analysis of some component of the formulation. For example, if molybdate is used as a tracer and a formulation contains 1% molybdate as MoO4, an analyzed concentration of 1 mg/L of MoO4 would be reported as 100 mg/L of the formulation. Total vs. Soluble Assays When using analytical data to monitor the performance of corrosion and scale inhibitors, it is helpful to understand the water chemistry of the operating system and to control the form in which the active chemical exists in order to provide protection. A prime example of this type of analytical monitoring is the use of zinc ions (Zn+2) as a corrosion inhibitor for steel. The protective film formed by Zn+2 ions

on steel is a form of zinc hydroxide, combined with zinc carbonates and phosphates. Under cooling water conditions, with the water pH above 8.0, Zn+2 can precipitate in the bulk water as zinc hydroxide. Polymeric organic compounds added to the water as part of the product treatment can “stabilize” Zn+2—that is, inhibit the precipitation of zinc hydroxide. However, at cathodic sites on the steel surface where hydroxide ions are generated and adsorbed, the effective pH is substantially higher than in the bulk water, thus providing additional potential for the precipitation of zinc hydroxide. The objective of corrosion protection with Zn ions is to properly adjust the water chemistry and the concentrations of Zn and polymer. Zn ions remain mostly soluble in the bulk water, but precipitate when reaching the higher pH film at the metal surface, thus forming a protective film on the surface. The success or failure of this process is monitored analytically by measuring concentrations of total and soluble Zn in the water. As a practical matter, the total Zn value is often close to that calculated from the product dosage. At least half of the total is typically soluble Zn. If the soluble Zn level is too low, most of the Zn precipitates in the bulk water and corrosion protection is lost. If the soluble Zn level is too high, it is likely that the Zn is overstabilized and will not precipitate, even at the high-pH cathodic areas on the metal surface. Again, there is no corrosion protection. The problem is corrected by changing the system conditions and monitoring the results analytically. If too much Zn is precipitating, the pH is typically reduced, or the amount of polymer in the product is increased. Conversely, if too much Zn is soluble, the pH is often raised slightly, or the polymer dosage is decreased. Using Analytical Data to Monitor Performance Ongoing analytical records from an operating cooling water system are a valuable tool for monitoring and controlling the use of corrosion and scale inhibitors. To be useful, these records are usually not simply accumulated in log books. The data are typically entered into spreadsheets so that calculations are made easily and graphs plotted. Commercial spreadsheets are excellent for this purpose. Alternatively, most of the major water treatment vendors have proprietary software programs that plot graphs and also do statistical process control calculations. The records are reviewed regularly by the water treatment vendor and by plant people with the training and experience necessary to draw sound conclusions and make recommendations. Trend Graphs Trend graphs—that is, plots of chemical levels versus time—provide a clear visual understanding of the progress of a cooling water monitoring program. Graphs sometimes include control ranges established by the vendor, analyses for total chemical concentration as shown by tracer analyses, and specific analyses for active components. All data typically agree with expected values within reasonable experimental error, usually about 10 to 15%. The variability

9

NACE International

within the control ranges is often acceptable for the specific site. If high or low excursions exist beyond the control ranges, the plant records and service reports provide explanations for the problem, and the trend graphs show that the excursions were quickly corrected. Problems in these areas provide opportunities for joint plant/vendor projects to improve the monitoring and control program. Cycle Calculations A second way in which plant and vendor records are used for monitoring system performance is by calculating cycles of concentration. This is typically calculated from a spreadsheet format. As explained at the beginning of this section, system operating parameters, including expected cycles, are established when the chemical treatment program is set up. Ongoing cycle calculations then compare operating results with the original plan. Cycles are often calculated for soluble parameters such as conductivity, chloride, molybdate, and potentially precipitating species such as calcium, magnesium, total alkalinity, phosphate, and silica. Allowances are made for site-specific contaminants, but overall agreement is typically within 10 to 15%. Poor agreement and some low cycles indicate precipitation of mineral scales or formation of corrosion product deposits. Selection of Control Methodology All of the chemical feeding, control methods, and monitoring techniques described in this section provide a technical basis for selecting the most cost-effective methodology for use at specific facilities but may not necessarily be the most

effective corrosion control technique. Cooling tower systems range from small to very large. Heat loads vary from simple space cooling with a 3 to 6°C (5 to 11°F) temperature drop, to critical process control systems with a 17 to 22°C (31 to 40°F) temperature drop. Personnel availability ranges from adequate to very low. At the same time, this report shows that systems available for feeding, controlling, and monitoring chemical feed to cooling tower systems are rudimentary to sophisticated. It is incorrect to assume that simple cooling systems use only simple controls, and that large, complex systems use advanced technology. In practice, this is often the case. However, a relatively small cooling tower may provide heat removal for a critical data processing system or for a manufacturing unit that typically relies on precise temperature control for product quality. It is important, therefore, to survey the system and understand the equipment, heat load, operating requirements, and personnel availability before designing a chemical feed and control system. Automation is often most effective with sophisticated feed systems, including fixed-delivery chemical pumps or on-line control with automated analytical methods. However, automation is neither needed nor cost-effective in specific cases. Maintenance of an automated system typically uses a different level of skill than is used for a simple “bleed and feed” or make-up control system. The vendor and plant personnel—working together—typically consider all of these factors and match the chemical feed system, control technology, and monitoring methods to each specific cooling water installation.

Microbiological Monitoring and Control

Cooling tower water is an excellent environment for microbiological growth. If not properly controlled, microbiological activity can cause corrosion, fouling, accelerated decay of cooling tower lumber, and can also represent a public health hazard. Biocides are specific inhibitors that reduce the extent of microbiological growth. Biocides are fed separately from corrosion and scale inhibitors and have specific dosing and control strategies. A complete review of biocide application and selection is beyond the scope of this report. Basic monitoring and control strategies for biocides are discussed here.

Biocides are normally subdivided into two major categories: oxidizing and nonoxidizing. Each type commonly uses different monitoring and control techniques. The following sections discuss the monitoring and control of these biofouling control programs. Oxidizing Biocide Monitoring and Control Table 2 lists commonly used oxidizing biocides and control methods.

10

al

NACE Internation

Table 2: Oxidizing Biocide Monitoring and Control

Biocide

Most Common Method of Feed

Typical Dosage

Monitoring Methods

Chlorine Low-level continuous

0.2 to 0.4 mg/L free residual

N,N-diethyl-p-phenylenediamine (DPD) Iodine Amperometric Oxidation reduction potential (ORP)

Bromine Low-level continuous

0.1 to 0.3 mg/L free residual

DPD Iodine Amperometric ORP

Chlorine dioxide

Low-level continuous or intermittent

0.02 to 0.1 mg/L residual

DPD Iodine Amperometric ORP

Ozone Low-level continuous

0.01 to 0.05 mg/L

Indigo blue

residual

Oxidizing biocide residuals are normally measured and controlled both manually using grab sample measurements or continuously online using automated feedback monitoring and control. A variety of tests and monitoring options are available for both approaches. The following paragraphs provide details. Chemical Tests for Oxidizing Biocides N1N-dicthyl-p-phenylene diamine (DPD) is probably the most popular method for measuring oxidizing biocides. This method is often either colorimetric or titrimetric and has been used for measuring chlorine, bromine, and chlorine dioxide. In the colorimetric method, a known amount of DPD is added and the color intensity is proportional to the biocide residual present. This method is very accurate when used with a spectrophotometer to measure color intensity, but is also often used with a visual color comparison chart. The titrimetric method uses the DPD as an indicator and is titrated with ferrous ammonium sulfate to an end point. When both chlorine and bromine or chlorine and chlorine dioxide are present in a cooling water, a variation of the DPD method involving the addition of glycine is sometimes used to differentiate between chlorine and the other biocide. In this method, the glycine complexes the oxidizing chlorine and the subsequent measurement is only the bromine or chlorine dioxide fraction of the oxidizing residual. The iodometric titrimetric method of measuring oxidizing biocide involves the conversion of iodide (I-) to iodine (I0) by the oxidizing biocide present in the test system. Starch indicator is used to detect the presence of I0, and the I0 is then titrated (reduced to I-) with sodium thiosulfate until the color disappears.

The indigo blue method is frequently used for measuring ozone. This method is based on the reaction of ozone with indigo dye to cause fading of the blue color. The amount of fading is proportional to the amount of ozone in the sample. Amperometric titration is an electronic titration that measures an abrupt decrease in current flow across an electrode as a reducing agent is added. Equipment is often costly and this method involves operator training. Amperometric titration is considered to be the most accurate chlorine, bromine, and chlorine dioxide measurement method. On-Line Control of Oxidizing Biocides In addition to grab sample tests, oxidizing biocides are sometimes monitored and controlled using on-line feedback monitoring and control. In an automated system, a monitoring device measures either the oxidizing biocide residual or some other parameter that correlates with the residual, and this information is fed back to the feeding equipment. Typically, the monitoring device signals when the measured parameter is too low, and this turns on the feed equipment. The feed equipment then runs for a set time or until the monitoring device determines the residual has reached an appropriate level. The three types of monitoring devices typically used for on-line monitoring and control are: • On-line DPD monitoring; • ORP; and • Amperometric electrode. The on-line DPD monitoring device is an automated version of the DPD colorimetric test. The monitor takes a sample of

11

NACE International

water, adds DPD and other necessary reagents, and the water sample passes it through a spectrophotometer that measures residual based on color intensity. The resulting halogen residual reading is then converted to a signal to control the feed equipment based on instructions programmed in by the operator. ORP is often used to monitor and control oxidizing biocide residuals. An ORP system measures the potential difference between two electrodes. The method is based on the potential change that occurs when oxidizing species are present in a water. The ORP increases with oxidizing biocide concentration. The correlation between ORP level and a given biocide residual is site- and species-specific and varies greatly depending on pH, temperature, and other site parameters. Because of this, the set point for the ORP controller is determined through on-site operating experience. ORP is also used to detect process leaks, as discussed later. An amperometric electrode, either closed or membrane covered, measures oxidant residual proportional to a current between working and counter electrodes. In the open cell, the electrode uses a constant sample flow. In the membrane-covered cell, the current is measured in an electrolyte and changes in flow rate do not affect the measurement. Specific ion electrodes to measure specific chlorine, bromine, or chlorine dioxide species and to handle specific contaminants in systems are sold.

Nonoxidizing Microbiocides In selecting and applying a nonoxidizing biocide, consideration is often given to a variety of factors, including compatibility of the biocide with the cooling water pH and with other components of the chemical treatment program, the type of microorganisms in need of control, and potential system contaminants. Some nonoxidizing biocides react very rapidly and are normally used in systems with short half-lives. Others are successful with longer contact times and are typically utilized in systems with relatively long half-lives. Method of feed and ease of biocide testing are also usual factors for consideration. Nonoxidizing biocides are quite reactive chemicals that are deactivated by some water species encountered in cooling water systems. Table 3 is a summary of typical monitoring and control considerations for the nonoxidizing biocides approved by the U.S. Environmental Protection Agency (EPA)(2) for use in cooling water systems. It also lists the applicable pH range for each biocide and the water contaminants known to deactivate each biocide. Methods for measuring the concentration of nonoxidizing biocides are biocide-specific and are also listed in Table 3. Many of the methods are laboratory-based, limiting applicability for routine in-system monitoring.

12

___________________________

(2) Environmental Protection Agency (EPA), Ariel Rios Building, 1200 Pennsylvania Avenue NW, Washington, DC 20460.

NACE International

Table 3: Nonoxidizing Microbiocide Monitoring and Control

Biocide Typical Feed

Method

Typical Dosage

Monitoring Method

Effective pH Range

Incompatibilities

Methylchloro and methyl isothiazolones (MCMI)

Slug 0.5 to 3.0 mg/L

High-pressure liquid chromatography (HPLC)

5 to 9.5 Reducing agents, sulfides, mercaptans, thiols, bisulfite, secondary amines

Dodecyl-guanadine hydrochloride (DGH)

Slug 5 to 20 mg/L

Colorimetric field test

6.0 to 9.5 Anionics

2-Bromo-2-nitropropane-1, 3-diol (BNPD)

Slug 20 to 100 mg/L

HPLC 5 to 9 Sulfides

Quaternary ammonium compounds (quats)

Slug or continuous low level

5 to 20 mg/L

Colorimetric field test

6.5 to 9.5 Anionics, salinity

Tetrakishydroxymethyl phosphonium sulfate (THPS)

Slug 25 to 50 mg/L

Colorimetric field test

7 to 10 Sulfites, halogens, oxidizing compounds

2-(Decylthio) ethanamine hydrochloride (DTEA)

Slug 10 to 30 mg/L

HPLC 7.5 to 10 Oxidizers

Methylene bis thiocyanate (MBT)

Slug 10 to 20 mg/L

HPLC 6 to 7.5 Sulfides

Glutaraldehyde Slug 75 to 150 mg/L

Colorimetric field test

7 to 9 Ammonia, primary and secondary amines

Dibromonitrilopropion-amide (DBNPA)

Slug or continuous low level

5 to 10 mg/L

Iodometric titration or HPLC

< 8 Reducing agents, H2S, mercaptobenzo-thiazole

Dithiocarbamates Slug 5 to 20 mg/L

Titration with metals

7 to 9 Oxidizers

Feeding Nonoxidizing Biocides As indicated in Table 3, nonoxidizing biocides are often fed on a slug or continuous basis. Slug feed is the most common and generally most cost-effective approach for feeding nonoxidizing biocides. The strategy of slug feed is often based on dosing in a sufficient quantity of biocide to exceed the minimum inhibitory concentration of the biocide. The biocide then remains in the cooling water until it dissipates from the system as a result of hydraulic turnover

or washout. The frequency of slug feed is determined by many factors. If used as the sole biocide approach, frequency varies from once to several times per week. If used in conjunction with an oxidizing biocide program, nonoxidizing biocide is typically added only once or twice per month. Equipment for feeding nonoxidizing biocides is based on safety and reliability. Most feed systems employ a chemical feed pump that is either timer or manually activated.

Process Contamination Monitoring and Control

The Significance of Process Contamination Another significant aspect of the monitoring of cooling tower water quality is the detection and correction of leakage of process fluids or other contaminants. This monitoring is performed because process leakage represents an equipment failure that is typically corrected, but also because process fluid contamination can lead to increased corrosion, scaling, or especially microbiological problems.

The number of possible process contaminants is as varied as the number of process unit operations. Typical contaminant types are: • Ammonia, amines, urea, or other nitrogen-containing

species; • Heavy or light hydrocarbons; • Sulfides, mercaptans, or other “sour” materials;

13

NACE International

• Alcohols, sugars, glycols, or other alcohol derivatives;

or • Acids or bases. The specific problems caused by leaks vary with the contaminant type. Contaminants that react with or destroy halogen residuals or provide microbiological nutrients can cause uncontrolled microbiological growth and all the associated problems. Acids or bases can cause dramatic swings in cooling water pH and the corrosion or scaling problems associated with these excursions. Heavy oils can cause fouling problems. Detecting Process Contamination The specific quantitative detection of process fluids is possible, but the methodology is quite specific to the particular fluid. If process contamination is a recurring problem, specific manual or automatic analytical assays are often employed. An on-line analyzer is often used, either when government regulations request it or when damage is rapid. For instance, if the process side of the heat exchangers contain strong acids or bases, or these materials are being added to the cooling tower for pH control, an on-line pH controller is appropriate. For rapid detection of volatile hydrocarbon leaks, the appropriate analyzer is often installed on the return water. Heavier water-soluble hydrocarbons are usually detected by an on-line total organic carbon (TOC) analyzer or an on-line gas chromatograph (GC) designed for a specific component. On-line process contamination analyzers often include alarm and data management capability. Samplings for any method of analysis are often representative of the whole tower return. In addition to their use for biocide control, ORP measurements are used to detect a variety of contaminants. When on-line analyzers are not available, manual methods are employed. The test method varies with the potential contaminant, but a plant may have sophisticated methods for testing for process components in process streams. These methods are generally adapted to testing cooling water. For hydrocarbons that remain at least partially soluble in water, direct injection of a water sample into a GC with an appropriate detector or a mass spectrometer is often performed. For hydrocarbons that are insoluble in water, the return water is typically extracted with a solvent such as hexane or fluorohydrocarbon, using a method such as EPA 90704 or 16645 for oil and grease. Significant oil content suggests a leak. When there are several potential leak sources, the extracted oil is usually analyzed by GC or true boiling point (TBP) analysis to identify the leaking stream. For volatile hydrocarbons or industrial gasses, the sample is typically analyzed by the purge and trap GC method. A

common modification of this method is to do head-space analysis. A quart (liter) bottle is rapidly filled 90% with return water and quickly stoppered with a septum. A vapor sample is withdrawn from the headspace and GC analysis is performed. This often helps to identify the source of the leak. In other instances, there are common detection methods for other types of leaks such as ammonia or hydrogen sulfide. When government regulations apply, frequency of testing is often dictated. Aside from that, the cost and frequency of testing is weighed against the consequences of not detecting a leak. There are some simple indications of process contamination that are often employed. They include, but are not limited to: • Loss of halogen residual; • Changes or spikes in water pH; • Foaming; • Sheen, scum, or other visible changes in water

appearance; • Odors; • Changes in monitored corrosion or deposition rates;

and • Appearance of turbidity. Controlling Process Contamination Although the impact of process fluid contamination varies with the type and intensity of the leakage, a general practice is that if process fluid contamination is detected, the source is found and the leak fixed. This typically means a system or unit shutdown for repairs. This depends on the type of leak and the characteristics of the system. For example, if the system has a film-fill tower and the leak consists of insoluble or soluble-biodegradable hydrocarbons, a leak probably cannot be tolerated for more than a few days without plugging or damaging the tower fill. Other leaks, such as volatile hydrocarbons, often do not represent a water problem but can cause fire hazards and air pollution concerns. Measures that are typically employed on a short-term basis to control severe leaks or on a longer-term basis for minor leaks include: • Reduction in operating cycles; • Increased halogen addition rates; • Use of nonoxidizing biocides; • Filtration, skimming, or other contaminant removal

techniques; • Use of oil dispersants or other supplemental treatments

to reduce fouling; and • Mechanical antifouling measures such as air rumbling

or flow reversal.

14

NACE International

Monitoring Corrosion Control

The monitoring methods described in the previous section are primarily aimed at maintaining the quality of cooling tower water, including the dosing and application of corrosion, scale, and microbiological inhibitors. In this and the following sections, the focus is on typical methods of measuring the success of these measures. The most relevant assessment of waterside damage to a cooling system is normally made by direct inspection. However, access to the internals of a system for inspection is generally limited during operation. Most of the other corrosion, scaling, or fouling measurements consist of indirect measurements made by some type of simulation device or test probe. Interruption of operation is not typically needed to gather data for this type of measurement. This section focuses on the use of monitoring methods to obtain useful information about system corrosion control and about the performance of water treatment programs during normal production operations. The next section provides similar information about monitoring mineral scaling, general deposition, and microbiological fouling. Direct vs. Indirect Measurements Performance monitoring methods have tradeoffs between the goals of relevance and timeliness. The most relevant data are provided by equipment inspections and condition assessments. These measurements are considered direct because they reflect the actual condition of production equipment. Because the internals of production equipment are typically accessible only at plant or unit outages, and these outages occur at very infrequent intervals, direct inspection data are not timely. This lack of timeliness makes direct inspection data impractical for daily adjustments of operating parameters. Conversely, the most timely data are provided by on-line instrumentation measuring the behavior of test devices. The assumption supporting the relevance of these data is that the devices behave identically to the production equipment. This is nearly always a shaky assumption because the units exposed to cooling water represent a variety of conditions and respond differently to changes in operating parameters. Consequently, the relevance of these indirect measurements is lower than that of direct inspection. Corrosion Coupons Corrosion coupons and other corrosion specimens such as spool pieces are indirect corrosion measurements. The primary purpose of coupons and other test probes is to measure the corrosiveness of water. Corrosion coupons

are small test specimens of relevant materials that are exposed to the circulating tower water. After a period of exposure, they are removed and evaluated with respect to the amount of corrosion and deposition they have experienced. Because coupons represent an indirect method of performance measurement, their value is determined by how well they mimic the conditions of the cooling system. Sources Corrosion coupons are normally supplied by water treatment vendors as a component of a service package. They are also available from suppliers of metal specimens. Coupons can be ordered “ready to use,” preweighed with appropriate surface preparation and engraved identification numbers. Installation In cooling tower systems, corrosion coupons are typically installed in a coupon rack, an assembly of piping containing a representative stream of cooling system water flowing at a controlled rate. Coupons are typically mounted on a nonmetallic holder in electrical isolation from other components of the system. Another method used is the direct insertion of a coupon into a flowing stream within a pipeline. Processing and Evaluating Corrosion Coupons After a period of exposure, coupons are typically removed from service and sent to a metallurgical laboratory for processing and evaluation. Processing procedures are described in other publications. At a minimum, the coupons are typically dried, weighed, cleaned, and weighed again, producing weight gain and loss numbers that are reflective of average deposit and corrosion rates. In addition, types of corrosion and deposit are evaluated by inspection, microscopic evaluations of pitting penetration rates are provided, and a chemical analysis of deposits is completed. This information is typically summarized in an analytical report. Factors That Influence Corrosion Coupon Data Coupon results are an indirect measurement. Some of the factors that influence corrosion and deposition in cooling systems and that are replicated as much as possible in a coupon installation are: • Metallurgy: Corrosion coupons are available in a wide

variety of materials. Selection of coupon materials often mimics the variety of metallurgies encountered in the system. Typical materials used are carbon steel,

15

NACE International

copper, cupronickel, Admiralty brass, stainless steel (SS) (UNS(3) S30400 [304 grade] and UNS S31600 [316 grade]), and aluminum (Al). Although galvanized steel is often used in packaged cooling towers, galvanized coupons are not often used because of the difficulty in obtaining meaningful corrosion rates. Galvanized coupons can provide a visual indication of conditions in the system.

• Temperature: Metal corrosion is strongly influenced by water and surface temperature. Coupons exposed to water from the hottest part of a cooling circuit typically give higher corrosion rates than those from cooler sections. If only one coupon rack is installed, it is typically on the water returning to the tower from the plant, because that is normally the location of the highest temperature. If problems are encountered in specific parts of the plant, it is sometimes worthwhile to install a coupon rack at that location.

• Exposure Time: Corrosion rates are highest on fresh

metal specimens and decline over the period of exposure. Thus, a fresh coupon inserted into the system corrodes at a faster rate than similar metal surfaces that have had longer periods of exposure. If the corrosion rate of the coupon is expected to represent the rate on similar materials in the system, the period of exposure of the coupon is typically as long as possible. However, long exposure times reduce the usefulness of the data because fewer data points are obtained and the usefulness of the data for making system adjustments declines over time. Data from exposure times of less than 30 days are generally considered meaningless. Exposure times of 60 to 90 days are considered more representative.

• Water Flow Rate: The corrosion rate of metal surfaces is influenced by the velocity of flow across the surface. Because of this, the flow rate of water is often matched through the coupon rack to the operating equipment as closely as possible. Flow control through coupon racks is typically achieved through orifice flow controllers. Inaccurate or variable flow rates are a frequent cause of nonrepresentative corrosion rates from coupons.

Advantages and Limitations of Corrosion Coupons Coupons have the advantage of being a simple and well-understood method of corrosion monitoring. If the exposure conditions of the coupon are carefully matched to those of the system components, they normally give representative results. With proper processing, they typically provide information on localized corrosion penetration rates. Coupons are lacking in timeliness with respect to on-line measurements. In the 60 to 90 days typically used to obtain

a valid data point from a coupon, the system is sometimes exposed to a wide variety of conditions. The exposure period represents a composite result of conditions encountered over that time period. Coupons do not replicate either the surface temperature or the hydrodynamics of a heat exchanger tube. Coupons measure the corrosiveness of the water to the coupon metal, under the specific test conditions. Extrapolation from these data to what is happening in plant equipment requires interpretation and site-specific experience. Additional References Corrosion coupons have been used for many years for cooling water corrosiveness monitoring. The intent of this section is to provide an overview of issues in their use. NACE Standard RP01896 briefly describes the use of corrosion coupons in cooling water systems. ASTM(4) Standards G 17 and G 468 are also valuable references. The Cooling Technology Institute(5) (CTI) has recently published a standard9 for the use of corrosion coupons in cooling systems. Electrochemical Corrosion Measurements Electrochemical corrosion measurements are performed for similar purposes as coupon measurements. A test specimen (electrode) is typically exposed to cooling tower water and experiences corrosion that is hoped to be representative of that in the cooling system. However, electrochemical measurements differ from coupon measurements in that they produce data on-line during the period of exposure. They are similar to coupons in the sense that they only measure the corrosiveness of the water stream, and they may experience corrosion rates that differ from the operating equipment. A more detailed reference is NACE Publication 3T199.10 Linear Polarization Resistance (LPR) Electrodes LPR electrodes operate on the principle of electrochemical linear polarization. To make a measurement, a small potential (10 mV) is imposed between two identical electrodes (specimens). The steady-state current resulting from this imposed potential is measured and is roughly proportional to the corrosion rate. The current is translated by the instrument into an “instantaneous” corrosion rate reading. LPR measurements are often taken on a continuous or intermittent basis. Test probes of representative metallurgy are installed in cooling water streams at points representative of the monitoring strategy. Typically, probes are located in the cooling water near identified problem

16

___________________________ (3) Metals and Alloys in the Unified Numbering System (latest revision), a joint publication of ASTM International (ASTM) and the Society of Automotive Engineers Inc. (SAE), 400 Commonwealth Drive, Warrendale, PA 15096. (4) ASTM International (ASTM), 100 Barr Harbor Drive, West Conshohocken, PA 19428-2959. (5) Cooling Technology Institute (CTI), 2611 FM 1960 West, Suite H-200, Houston, TX 77068-3730.

NACE International

spots to expose the probe to conditions that are similar to those in production equipment. The principal advantage of LPR probes vs. coupons or inspection technique is the timeliness of the data. Once the probe has been exposed to water for 36 to 48 hours, it can provide data on day-to-day changes in operating conditions, including short-term upsets. Inspection data are only available infrequently (based on system or unit turnarounds). Coupons typically need 1 to 3 months of exposure to produce a single data point. LPR probes can provide data across time intervals of as little as a few minutes. LPR probes suffer the same limitation as other indirect corrosion measurements. They inherently measure the corrosiveness of the water in a simulated situation. The relevance of the data obtained from an LPR electrode depends on how closely its exposure conditions reflect the conditions in the target plant equipment. Also, the corrosion rate readings on commercial LPR instruments are based on assumed constants that may or may not be correct in a given application. Along with this general limitation, LPR probes cannot measure corrosion accurately in low-conductivity (<50 µS/cm) water. This limitation is a function of the corrosion rate vs. the conductivity of the water. If the corrosion rate is high, then the minimum conductivity typically used is increased. Usually, recirculating water in cooling tower systems has sufficient conductivity for LPR probes to function effectively. Data from LPR probes are typically representative of the rate of general corrosion. Although some instruments can provide a pitting index, this measurement is not generally well correlated to actual pitting rates. More advanced devices for detection of pitting are described in the section on Localized Corrosion Monitoring. Electrical Resistance (ER) Electrodes11 ER probes are seldom used in cooling water systems. They are an indirect simulation measurement. Consequently, they are subject to many of the same requirements as corrosion coupons in terms of time of exposure, flow conditions, temperature, and other simulation parameters. However, the measurement principle is different. This difference introduces some additional considerations in their application. ER probes measure changes in the electrical resistance of the probe. These changes are caused by reduction in the physical cross-section of the corroding element as a result of loss of metal to corrosion. Because of this measurement principle, ER is most useful in systems in which the conductivity of the corrosive environment is low, making LPR impractical. Also because of this measurement

principle, the ER is acutely sensitive to localized corrosion and pitting. In an aqueous system, depending on the design of the probe itself, a single pit can significantly distort an ER measurement in a short period. The signal returned by the sensor is not a corrosion rate, but rather the electrical resistance of the probe. The corrosion rate is determined from the slope of a plot of the reading (probe resistance) vs. time. ER is a successful monitoring method for systems in which the corrosion is general in nature. It is typically not a useful tool for systems in which the corrosion is localized in nature. This is the situation in most cooling water systems. For this reason, ER is not generally used in cooling water systems. Localized Corrosion Monitoring On-line monitoring of localized corrosion is an emerging and potentially valuable technology. There are currently no accepted industry standards in this area. It is known by most experienced water treatment practitioners that equipment failures are seldom caused by general corrosion damage. If that were the case, typical low carbon steel heat exchanger tube bundles constructed of 2.1-mm (0.083-in.) tubes, corroding at 51 mm/y (2 mpy) would last 40 years. A survey12 conducted by the Chemical Society of Japan(6) concluded that the typical lifetime of these bundles is in the 5- to 20-year range. This survey also concluded that pitting was the predominant form of corrosion in 80% of the bundles surveyed. Because it only takes one penetrating pit to produce a leaking heat exchanger, accurate and timely measurement of pitting rates assumes high relevance to plant reliability and lifetime. Techniques for the on-line measurement of localized corrosion are relatively new and do not have the validation of more established general corrosion measurements. The equipment is more elaborate and expensive, and methodology for data interpretation is not as well established. Methods currently in use or under study for specific measurement of localized corrosion include the following: Physical Measurements It is possible to measure pit penetration depths on coupons or directly on heat exchanger tubes by microscopic examination or by ultrasonic test methods. These are direct methods that are typically performed only at inspection opportunities or when coupons are processed. ASTM G 468 provides methods for pit depth measurements. Physical measurement methods provide the most relevant measurements of pit penetration rates. However, similar to other direct measurements, they suffer from poor

17

___________________________ (6) Division of Colloid and Interface Science, The Chemical Society of Japan (CSJ DCSC), Surugadai 1-5, kanda, Chiyoda-ku, Tokyo 101, Japan.

NACE International

timeliness. The rate of pitting corrosion is greatly intensified during system upsets. Physical measurements with their poor time resolution are not useful in identifying and characterizing the impact of upsets. Electrochemical Imbalance Some commercial LPR devices include a capability for measurement of an imbalance. In this mode of operation, the device measures the difference in direct current (DC) between the electrodes, when the polarity is reversed. In principle, if there is more localized corrosion on one of the electrodes than on the other, a differential current is produced. This technique has been available since the introduction of commercial LPR electrodes but has not gained wide acceptance. Because the probes are exposed to exactly the same environment, it is a matter of chance whether one of the probes develops more pitting corrosion than the other. If both electrodes are “equivalently” pitted, then no differential current is produced. Electrochemical Noise (ECN) Development of measurement techniques and devices based on ECN is currently an active area of research. ECN measurements operate on the principle that when corrosion pits initiate, grow, or passivate, they produce fluctuations in the corrosion potential (potential noise) or in the electrochemical current (current noise). Statistical analysis of the magnitude and frequency of these fluctuations can provide information about the extent and severity of localized corrosion. ECN techniques have been widely used in various application areas. However, their use in cooling system monitoring has been limited to some experimental projects. There is some controversy on the interpretation of noise data. Occluded Cells Occluded cell devices have been used for the measurement of localized corrosion. These devices function by creating a

synthetic crevice (occluded cell). Measurements of the differential current between a metal specimen in the occluded cell and one exposed to unrestricted cooling water conditions are translated into an estimate of localized corrosion in the crevice area. Differential Flow Cells Another approach to monitoring of localized corrosion involves the creation of an environment conducive to localized corrosion by exposing metal specimens to different flow conditions. Because corrosion processes are flow-dependent, this results in the formation of a heavy layer of corrosion product on the low-flow specimen. Once this condition is created, the differential current between the fouled surface and a relatively clean high-flow surface of identical composition is measured. These data are compared with other electrochemical information and typically yield quantitative pit penetration rates. In the differential flow cells, the localized corrosion rate is typically calculated in accordance with Equation (13). CR = K × ZRA + K × I-Corr (13) where: ZRA = the galvanic coupled current density flowing between the slow flow electrode (or the anode) and the fast flow electrode (or the cathode) and it is measured by a zero resistance ammeter; I-Corr = the corrosion current density on the slow flow electrode measured when it is temporarily de-coupled with the fast flow electrode; and K = the coefficient used to convert corrosion current density to corrosion penetration rate. The K×ZRA term in the equation represents the contribution of cathodic reaction current (e.g., oxygen reduction) on cathode (i.e., fast flow area) toward the corrosion of the anode (slow flow area). The K×I-Corr term in the equation represents the contribution of cathodic reactions (e.g., oxygen reduction and hydrogen evolution) on the anode toward its own corrosion.

Monitoring Scale, Deposit, and Biofilm Control

This section describes methods for monitoring mineral scale formation, general deposition, and biofilm. Test Heat Exchangers Test heat exchangers are simulation devices that, if properly designed and operated, allow measurement of fouling and corrosion under conditions that approximate the hydrodynamic and thermal conditions in plant heat exchangers. Because of this consideration, test heat exchangers are probably the most reliable and useful methods for monitoring corrosion and fouling. Unfortunately, the test heat exchanger methods are more

expensive to implement and typically use more instrumentation, operational attention, and data processing than other monitoring techniques. For these reasons, test heat exchangers are used relatively infrequently. This topic is thoroughly covered in NACE Standard RP0189.6 Operating Parameters Test heat exchangers are simulations, as previously defined in the Monitoring Corrosion Control section of this report. Consequently, in order to obtain representative data, the operating conditions are typically thoughtfully chosen. One approach is to choose a target heat exchanger and set the

18

NACE International

operating conditions to mimic that exchanger as closely as possible. Another approach is to choose operating conditions that approximate the most severe conditions (low velocity, high metal temperature) under which the treatment system is expected to perform satisfactorily. This approach allows the user to impose a “safety factor” by setting conditions more severe than those in the plant exchanger by a known amount (generally 1 to 3°C [2 to 5°F]). Some of the parameters include: • Bulk water temperature; • Heat exchanger surface temperature; • Water velocity (to simulate the shear stress of the water

film in the actual heat exchanger); and • Heat exchanger metallurgy. Test Heat Exchanger Measurement of Corrosion—Gravimetric/Visual Methods These devices are similar to a corrosion coupon in that they typically need an extended period of exposure, followed by an off-line destructive measurement. With a multi-tube unit, tubes are typically removed after different periods of exposure. Measurement of the deepest corrosion penetration on each tube then allows plotting the penetration vs. time, determining the corrosion rate from the slope of the curve. For a detailed description of this type of device refer to NACE Standard TM0286.13 Measurement of Fouling—Resistance to Heat Transfer Devices relying on this principle are typically electrically heated annular-tube heat exchangers. They rely on thermocouples or thermistors to calculate the temperature difference between a heated surface and the bulk water. Provided the water flow rate and heat transfer rates remain constant, an increase in the temperature difference (∆T) between the surface and the bulk water is an indication of increased fouling. Other electrically heated annular devices accurately measure and control flow and power applied to the heat exchange surface. That, combined with calibrated instruments, allows the development of a fouling factor when compared to clean operating conditions. Accurate modeling makes the fouling factor relevant to an operating heat exchanger. By collecting sufficient data around the test heat exchanger, an average heat-transfer coefficient is easily calculated. Comparison of this value to the clean value allows direct determination of the fouling factor. The heat transfer coefficient under clean conditions is typically determined from the heat exchanger design specifications, or it is sometimes empirically measured by taking data immediately after a clean test exchanger starts up. It is seldom possible to maintain steady-state conditions in any heat exchanger other than one that is electrically heated.

Pressure Drop and Friction Factor Measurement Fouling deposits also increase resistance to flow, thus decreasing energy efficiency. This effect is usually measured by means of a carefully controlled flow rate put through a test section. The resulting pressure drop is monitored. By calculation, the fouling is normally stated in conventional engineering terms useful for analysis of heat exchanger performance and design. Plant Heat Exchanger Efficiency Measurements One direct measurement of treatment success that is frequently obtained during an operating campaign is the efficiency of heat exchangers. The measurement is direct because it is a measurement of the performance of a piece of production equipment. The measurement is typically made during a production campaign because the data are often obtained without disturbing the operation of the heat exchanger equipment. Typical steps used to determine the operating efficiency of a plant heat exchanger appear in Appendix A. From these data, conclusions about heat exchanger condition and efficiency are typically drawn. For example, it might be found that: • The exchanger is operating at a process load much

higher than the original design, which could explain apparently inadequate heat transfer rates.

• The actual water flow rate is much lower than design, which could explain high water-side fouling or inadequate effectiveness of anti-corrosion and anti-fouling additives.

An experienced heat exchanger designer is typically brought in to assist in this analysis. Microbiological Monitoring Microbiological monitoring in cooling systems is conducted either by monitoring microorganism counts in the bulk water (planktonic biomass) or by measuring biological deposits (biofilms or sessile biomass) using a biological deposit monitor. Generally, the goal of any microbiological monitoring program is to ensure that sessile biomass within the cooling system is prevented or minimized. Planktonic counts are used to infer biological surface cleanliness whereas biological deposit monitors are used to monitor sessile growth directly. The following sections discuss the most commonly used methods for monitoring planktonic and sessile microorganisms. Planktonic Biomass Monitors Monitoring bulk water microorganism counts is the traditional and most widely used approach for cooling water biocontrol monitoring. Determination of planktonic counts is generally performed using the standard plate count (SPC) method. Other assay methods, including the dip-stick

19

NACE International

method, are also frequently used and are summarized in Table 4. Results of culture assay tests are usually expressed as colony-forming units (CFU) per mL. A major limitation of culture tests is that results are generally not available for 48 to 72 hours. To reduce response time,

metabolic quick tests such as adenoisine triphosphate (ATP) monitoring are normally used. These tests give a rapid indication of biomass activity, with results available in as little as 5 to 10 minutes.

Table 4: Cooling Water Microorganism Enumeration Methods

Method Method Basis Advantages/Disadvantages

Standard plate count, total plate count, agar tube count

Organism growth on culture media

Accurate, widely used, selective growth media are often used for differential analysis, uses dilutions, cumbersome for field use, 48 to 72 hours incubation typical for results.

Dip-slides Organism growth on culture media

Convenient for field use, widely used, 48 to 72 hours incubation typical for results, not as accurate as plate or agar tubes.

Metabolic quick tests(A) Detection of specific cellular components, enzymes, or metabolic end products

Results available in as little as 5 to 10 minutes, many tests species-specific, often utilize expensive equipment, lower limit of detection is typically high. Often does not correlate directly with plate counts.

(A) Examples include ATP photometry, respirometry, fluorescent antibody tests, and redox indicator tests.

Results from planktonic counts are used as an indicator of biofouling control in a cooling system and for establishing microbiological trends. Planktonic counts are also useful for evaluating biocide effectiveness and for quantifying and differentiating between different types of troublesome microorganisms. Experience with a particular cooling system indicates what level of bulk water counts is maintained and not exceeded to help ensure acceptable biofouling control for that system. Planktonic counts are an indicator, but often not a direct measure of biofouling. Planktonic counts are typically correlated with measurements of sessile biomass as discussed below. Sessile Biomass (Biofilm) Monitors The direct approach to microbiological monitoring in cooling water systems is to measure sessile (deposited) biomass using a biofouling monitor. Several different types of

biofouling monitoring devices are used. They are based on monitoring either changes in pressure drop or heat transfer resistance. Biofilm is sometimes measured directly using a surface colonization coupon, or indirectly by measuring electrochemical activity related to microbiological growth. Biofilm monitors typically operate on a slip stream from the cooling system and can provide real-time assessment of system biofouling conditions. Biocontrol treatment program performance is then typically monitored and adjusted to maintain clean surface conditions. In combination with an appropriate planktonic monitoring program, complete cooling system microbiological status is evident and biocontrol treatment can be fine tuned for optimum performance. Table 5 provides a summary of biofilm monitoring techniques.

20

NACE International

Table 5: Cooling System Biofouling Monitors

Type Measurement Advantages/Disadvantages Differential Pressure6

Changes in fluid frictional resistance

Sensitive to onset of significant biofouling, simple and easy to use, real-time fouling assessment, can provide electronic signal for data acquisition. No information on types of fouling organisms.

Heat Transfer6

Changes in heat transfer resistance

Based on cooling system parameter, real-time assessment, can provide electronic signal. Because heated surface scale formation is possible, no information on types of fouling organisms. This type of monitor is not specific to biofouling.

Direct Sampling Removable coupons for direct measurement of biofilm deposits

Test section with removable colonization coupons that are often used for deposit analysis, microorganism enumeration, etc.; coupons are typically used for information on microbial corrosion.

Electrochemical6

Increased applied and generated currents on electrodes provide a measure of biofilm activity

On-line, real-time measurements of biofilm activity on sensor electrodes. Signal is typically used in control systems or for biocide optimization.

Biofilm Pressure-Drop Monitors The pressure-drop tube is simply a straight length of stainless steel pipe equipped with two pressure taps to allow the determination of frictional resistance to flow by measuring pressure drop. A manometer, differential pressure gauge, or pressure transducer is used to measure the pressure drop. The pipe is installed in parallel with the cooling water system being monitored. Return water to the cooling tower is usually used to optimize temperature conditions for biofilm development. Flow in a pipe is always accompanied by frictional resistance, and consequently a loss in static pressure in the direction of flow. Frictional resistance, at a known flow rate, is typically calculated based on a pressure-drop measurement. Biological deposits tend to increase fluid frictional resistance dramatically because of the nonuniform and nonrigid nature of biofilms, making measurement of pressure drop a very sensitive approach for monitoring the extent of biofilm formation. Biofilm Heat Transfer Monitors Heat transfer monitors are a second approach to monitor biological activity in cooling systems. Biofilm heat transfer monitors measure the resistance to heat transfer resulting from the insulating effects of a biofilm. As discussed earlier in this report, increases in heat transfer resistance are often monitored and expressed in terms of a fouling factor. The primary limitation in using a heat transfer system for monitoring sessile biomass is the possibility of scale formation and the agglomeration of other foulants on the heated surface.

Biofilm Sampling Monitors Corrosion coupons or other types of removable test surfaces are sometimes used as test specimens for monitoring biofouling. Installation of test coupons or nipples on a bypass loop allows for convenient physical inspection and sampling. The presence of biofilm or slime on the test surface provides a visual qualitative indication that fouling conditions exist, or samples are removed for quantitative biofilm assessment. Various side-stream biofilm sampling devices are available in the market. They all contain removable colonization coupons that are often used for biofilm sampling and analysis. Biofilm Electrochemical Monitors This approach utilizes an electrochemical biofilm activity monitor. The monitor consists of a pair of metallic electrodes (typically stainless steel or titanium) that are polarized relative to each other by an integrated control/data acquisition and analysis system. The polarization causes a current to flow between the electrodes. The current is affected by the growth of biofilm on the electrodes. As biofilm forms, the applied current increases, providing an indication of the presence of the biofilm on the electrodes. The amount of deviation in applied current from a baseline, obtained with no biofilm present, provides a measure of the activity of the biofilm. The current between the electrodes when the external polarization is off is also monitored. This current, the “generated current,” is essentially zero when there is no biofilm present. However, as biofilm develops on the electrodes, slight differences in the microbial populations of the biofilm caused by the different electrode polarities cause small currents to be generated between the electrodes. This generated current also produces a deviation from the baseline condition and provides a second measure of

21

NACE International biofilm formation and activity. This technique can be considered a form of electrochemical noise analysis (ENA).

Deriving Control Actions from Monitoring Data

Monitoring produces data, which are then used to control or modify a process. There is typically a defined connection between data acquired and control or corrective actions taken. These connections are typically subdivided into two categories. 1. Data that determine whether the operating parameters are within specifications. 2. Data that indicate the extent to which waterside problems are being controlled. Data of the first category are analyzed most frequently. The water treatment vendor sets up specifications for water quality and inhibitor parameters based on knowledge of the product and the system characteristics. Along with this, the vendor defines corrective actions to take if the parameters are out of specification. A simple example would be to increase or decrease the blowdown if the system conductivity is too high or too low. It is then generally the

responsibility of the user to execute this plan by gathering the data and taking control actions as appropriate. This sequence of actions constitutes the innermost loop of measurement, analysis, and control. Data from the second category are acquired concurrently. These data, however, are used to appraise the results of the water management strategy, to adjust the control specifications, or to modify the type of treatment used. This analysis and response represents the outermost loop of measurement, analysis, and response. Trend graphs of analytical data, discussed earlier in this report, are a good example of this control strategy. Like most other things, the performance delivered is a tradeoff with cost. Cost-conscious users are unwilling to pay for excessive control measures, but also do not want to suffer the consequences of inadequate measures.

References

1. R.G. Kunz, A.F. Yen, T.C. Hess, “Basic Cooling Water Calculations: General Algorithm for Cooling Water Chemistry,” in AIChE(7) Symposium Series, 76 (197) (New York, NY: AIChE, 1977), pp. 175-83. 2. G. Caplan, G. Zamett, “Cooling Water Computer Calculations—Do They Compare?” CORROSION/90, paper no. 100 (Houston, TX: NACE, 1990). 3. L.S. Clescerl, A.E. Greenberg, A.D. Eaton, eds., Standard Methods for Examination of Water & Wastewater, 20th ed. (Washington, DC: APHA,(8) 1999). 4. EPA Method 9070 (latest revision), “Oil and Grease in Water by Gravimetry” (Washington, DC: EPA). 5. EPA Method 1664 (latest revision), “Guidelines Establishing Test Procedures for the Analysis of Oil and Grease and Non-Polar Material” (Washington, DC: EPA). 6. NACE Standard RP0189 (latest revision), “On-Line Monitoring of Cooling Waters” (Houston, TX: NACE). 7. ASTM G 1 (latest revision), “Standard Practice for Preparing, Cleaning, and Evaluating Corrosion Test Specimens” (West Conshohocken, PA: ASTM).

8. ASTM G 46 (latest revision), “Standard Guide for Examination and Evaluation of Pitting Corrosion” (West Conshohocken, PA: ASTM). 9. CTI Code STD-149 (latest revision), “Corrosion Testing Procedures” (Houston, TX: CTI). 10. NACE Publication 3T199 (latest revision), “Techniques for Monitoring Corrosion and Related Parameters in Field Applications” (Houston, TX: NACE). 11. ASTM G 96 (latest revision), “Standard Guide for On-Line Monitoring of Corrosion in Plant Equipment (Electrical and Electrochemical Methods)” (West Conshohocken, PA: ASTM). 12. “Field Usage Data on Soft-Steel Heat Exchangers in Cooling Water Environments,” Japanese Society of Chemical Engineers, Chemical Equipment Materials Committee, Corrosion Subcommittee, 1990. 13. NACE Standard TM0286 (latest revision), “Cooling Water Test Unit Incorporating Heat Transfer Surfaces” (Houston, TX: NACE).

22

___________________________ (7) American Institute of Chemical Engineers (AIChE), 3 Park Avenue, New York, NY 10016-5991. (8) American Public Health Association (APHA), 800 I Street NW, Washington, DC 20001.

NACE International

Bibliography

Ashland Chemical Co., Drew Industrial Division: Principles

of Industrial Water Treatment. 11th ed. Boonton, NJ: Ashland Chemical Co., 1994

ASTM D 2688 (latest revision). “Corrosivity of Water in the

Absence of Heat Transfer (Weight Loss Methods).” West Conshohocken, PA: ASTM.

ASTM G 4 (latest revision). “Standard Guide for

Conducting Corrosion Tests in Field Applications.” West Conshohocken, PA: ASTM.

Fontana, M.G. Corrosion Engineering. 3rd ed. New York,

NY: McGraw Hill, 1986. pp. 344-345. Freedman, A.J., A.S Krisher, and D. Steinmeyer.

Guidelines for Troubleshooting Water Cooled Heat Exchangers. Materials Technology Institute,(9) 2004.

GE Betz: Betz handbook of Industrial Water Conditioning.

9th ed. Trevose, PA: GE Betz, Inc., 1991. Licina, G.J. “Monitoring Biofilms on Metallic Surfaces in

Real Time.” CORROSION/2001, paper no. 442. Houston, TX: NACE, 2001.

McCoy, J.W. The Chemical Treatment of Cooling Water.

New York, NY: Chemical Publishing Company, 1983. NACE RP0300/ISO(10) 16784-1 (latest revision). “Corrosion

of metals and alloys—Corrosion and fouling in industrial cooling water systems—Part 1: Guidelines for conducting pilot-scale evaluation of corrosion and fouling control additives for open recirculating cooling water systems.” Houston, TX: NACE.

NACE Standard RP0497 (latest revision). “Field Corrosion Evaluation Using Metallic Test Specimens.” Houston, TX: NACE.

Nalco Co.: The Nalco Water Handbook. 2nd ed. New York:

McGraw-Hill Book Company, 1988. Perry, R.H., and D.W. Green, eds. Chemical Engineers’

Handbook. 7th ed. New York, NY: McGraw Hill, 1997. Snoeyink, V.L., and D. Jenkins. Water Chemistry. New

York, NY: Wiley Textbooks, 1980. Sommerscales, F.C., and J.G. Knudsen. Fouling of Heat

Transfer Equipment. Washington, DC: Hemisphere Publishing Co., 1981.

Stumm, W.E., and J.J. Morgan. Aquatic Chemistry. 3rd ed.

New York, NY: John Wiley and Sons, 1996. “Water Chemistry and Treatments.” In CTI Manual, Chapter

6. Houston, TX: CTI, 1990. White, G.C., ed. Handbook of Chlorination and Alternative

Disinfectants. 4th ed. New York, NY: John Wiley and Sons, 1998.

Yang, Bo. “Real-Time Localized Corrosion Monitoring in

Industrial Cooling Water System.” Corrosion 56, 7 (2000): pp. 743-756.

Yang, Bo. “Minimizing Localized Corrosion via New

Chemical Treatments and Performance Based Treatment Optimization and Control.” CORROSION/99, paper no. 307. Houston, TX: NACE, 1999.

Appendix A: General Procedure for Plant Heat Exchanger Efficiency Measurement

1. The heat transfer design specifications and mechanical construction details of the exchanger are typically collected. The operator usually determines that the heat exchanger is used in the application for which it was originally designed and that the heat duty is within 5% of the original design. If not, the original heat transfer design specifications are updated for the new application or the heat exchanger is cleaned, and several sets of data taken immediately after startup. 2. Field data on the inlet and outlet temperatures are collected, as well as the pressures of both hot- and cold-side fluids and the flow rates of both streams. Note that in many (probably most) plants, there is not sufficient instrumentation to determine such information easily.

Often, additional instruments are installed and calculations—such as for heat balance—to determine the water flow rate are performed. 3. The information from the field data are compared to what was expected based on the design. Conclusions that relate to field deviations from the design basis are drawn. Heat transfer calculations are performed to determine what the values of the coefficient would be if the exchanger were clean and operating at the observed conditions (Uc). From this value and the observed overall heat transfer coefficient calculated as described above (Ua), a fouling factor is usually calculated. The fouling factor is the difference between the reciprocals of Uc and Ua, as shown in Equation (A1):

23

___________________________ (9) Materials Technology Institute (MTI), 1215 Fern Ridge Parkway, Suite 206, St. Louis, MO 63141-4408. (10) International Organization for Standardization (ISO), 1 rue de Varembe, Case Postale 56, CH-1121 Geneve 20, Switzerland.

NACE International

fouling factor = ca UU

11− (A1)

This value is a direct measure of the total resistance to heat transfer of the fouling in the exchanger. Note that the value

calculated in this manner is the sum of the process side fouling and waterside fouling. It is normally considered a reasonable estimate of the waterside fouling only for those cases in which the process side fluid is known to be nonfouling.

24