Engineering of Power

Plant and Industrial

Cooling Water Systems
 Engineering of
Power Plant and
Industrial Cooling
Water Systems

Charles F. Bowman
Seth N. Bowman
First edition published 2022
by CRC Press
6000 Broken Sound Parkway NW, Suite 300, Boca Raton, FL 33487-2742

and by CRC Press

2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN

© 2022 Charles F. Bowman and Seth N. Bowman

CRC Press is an imprint of Taylor & Francis Group, LLC

Reasonable efforts have been made to publish reliable data and information, but the authors and
publisher cannot assume responsibility for the validity of all materials or the consequences of their use.
The authors and publishers have attempted to trace the copyright holders of all material reproduced in
this publication and apologize to copyright holders if permission to publish in this form has not been
obtained. If any copyright material has not been acknowledged, please write and let us know so we may
rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced,
transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or
hereafter invented, including photocopying, microfilming, and recording, or in any information storage
or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, access www.copyright.com
or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-
750-8400. For works that are not available on CCC, please contact [email protected].

Trademark notice: Product or corporate names may be trademarks or registered trademarks and are used
only for identification and explanation without intent to infringe.

ISBN: 978-0-367-77528-5 (hbk)

ISBN: 978-1-032-00039-8 (pbk)
ISBN: 978-1-003-17243-7 (ebk)

Typeset in Times
by Straive, India
 To Nancy
Without whose help this book could not have been possible
Contents
Preface xi
Authors xiii
List of Abbreviations and Acronyms xv

Chapter 1 Introduction.......................................................................................... 1
1.1 Background������������������������������������������������������������������������������ 1
1.2 Types of Cooling Water Systems��������������������������������������������� 1
1.3 Types of Cooling Water System Pumps����������������������������������� 1
1.4 Intake Pumping Stations���������������������������������������������������������� 2
1.5 Cooling Water System Piping and Valves�������������������������������� 3
1.6 Cooling Water System Flow Deficiencies������������������������������� 3
1.7 Heat Exchanger Fouling���������������������������������������������������������� 5
1.8 Engineering Design Errors������������������������������������������������������ 5
1.9 NRC Oversight������������������������������������������������������������������������ 6
1.10 Authors’ Perspective���������������������������������������������������������������� 7
References������������������������������������������������������������������������������������������� 7

Chapter 2 Properties of Water.............................................................................. 9

2.1 Temperature................................................................................ 9
2.2 Pressure....................................................................................... 9
2.3 Density...................................................................................... 10
2.4 Viscosity.................................................................................... 11
2.5 Reyolds Number....................................................................... 11
2.6 Prandtl Number......................................................................... 12
2.7 Specific Gravity........................................................................13
Reference........................................................................................... 13

Chapter 3 Intake Structures................................................................................ 15

3.1 Intake Structure Design............................................................. 15
3.2 Intake Structure Design Deficiencies........................................ 18
3.3 Requirements of §316(b) of the Clean Water Act..................... 19
3.4 Potential Improvements in the Engineering of Intake
Pumping Stations��������������������������������������������������������������������� 22
References.......................................................................................... 25

Chapter 4 Pumps................................................................................................ 27
4.1 Types of Pumps......................................................................... 27
4.2 Types of Pump Impellers.......................................................... 27

vii
viiiContents

4.3 Specific Speed......................................................................... 29
4.4 Cavitation and Pump Suction Specific Speed......................... 30
4.5 System Head Curve................................................................ 31
4.6 Pump Head Capacity Curve.................................................... 31
4.7 Pump Size, Efficiency, and Speed.......................................... 33
4.8 Net Positive Suction Head...................................................... 34
4.9 Pump Minimum Flow Rate..................................................... 35
4.10 Air Venting.............................................................................. 35
4.11 Pump Line Shaft Seals............................................................ 36
4.12 Pump Motors.......................................................................... 36
4.13 Pump Margin and In-Service Testing..................................... 37
4.14 Potential Improvements in the Engineering of Cooling
Water Pumps�������������������������������������������������������������������������� 37
References.......................................................................................... 39

Chapter 5 Piping and Valves............................................................................... 41

5.1 Piping Applications................................................................... 41
5.2 Piping Materials........................................................................ 41
5.3 Pipe Linings, Coatings, and Wraps........................................... 41
5.4 Buried Piping............................................................................ 41
5.4.1 Concrete Pressure Pipes................................................ 42
5.4.2 Pre-Lined Ductile Iron Pipes........................................ 43
5.4.3 Plastic Pipes.................................................................. 44
5.4.4 Stainless Steel Pipes..................................................... 45
5.4.5 In Situ Lined Pipes........................................................ 45
5.5 Hung Piping.............................................................................. 50
5.6 Nuclear Safety Considerations................................................. 52
5.6.1 Seismic Qualification.................................................... 52
5.6.2 Codes and Standards..................................................... 55
5.6.3 NRC Approval of CML................................................ 56
5.7 Valves and Fittings.................................................................... 56
5.7.1 Valves............................................................................ 57
5.7.2 Fittings.......................................................................... 58
5.8 Piping Design Wall Thickness.................................................. 58
References.......................................................................................... 59

Chapter 6 Corrosion and Fouling....................................................................... 61

6.1 Corrosion of Cooling Water Piping.......................................... 61
6.1.1 Background................................................................... 61
6.1.2 Tennessee Valley Authority Investigation..................... 61
6.1.3 Industry Response to MIC............................................ 68
6.1.4 MIC Tubercles.............................................................. 76
Contents ix

6.1.5 Potential Improvements to Combat Corrosion in

Cooling Water Systems������������������������������������������������ 77
6.2 Fouling In Cooling Water Piping.............................................. 79
6.2.1 Background................................................................... 79
6.2.2 Asiatic Clams................................................................ 80
6.2.3 Zebra Mussels............................................................... 80
6.3 Chemical and Heat Treatment.................................................. 81
6.3.1 Chemical Treatment...................................................... 81
6.3.2 Coatings for Zebra Mussel Control.............................. 82
6.3.3 Heat Treatments for Zebra Mussel Control.................. 82
6.4 TVA Qualification Testing of Chemical Treatment.................. 82
6.4.1 Corrosion Inhibitor Testing........................................... 82
6.4.2 Biocide Testing............................................................. 84
6.5 TVA Experience with Chemical Treatment.............................. 87
6.5.1 Sequoyah Nuclear Plant................................................ 87
6.5.2 Watts Bar Nuclear Plant................................................ 88
6.5.3 Browns Ferry Nuclear Plant......................................... 89
6.6 Conclusions and Recommendations......................................... 90
References.......................................................................................... 91

Chapter 7 Pipe Flow........................................................................................... 95

7.1 Equations for Head Loss by Pipe Friction................................ 95
7.1.1 Darcy-Weisbach Equation............................................ 95
7.1.2 Hazen-Williams Equation............................................. 95
7.1.3 Selection of Roughness Coefficient.............................. 96
7.2 Pipe Form Losses...................................................................... 97
7.2.1 Form Loss Equation...................................................... 97
7.2.2 Equivalent Lengths....................................................... 97
7.3 The Bowman-Bain Equations for Friction Loss....................... 99
7.4 Flow and Pressure Drop Tests of Large Diameter
Piping������������������������������������������������������������������������������������� 106
7.5 Multiflow Piping Analysis Software....................................... 109
7.6 Piping Network Analysis Using Multiflow............................. 110
References........................................................................................ 112

Chapter 8 Cooling Water System Engineering Pitfalls.................................... 113

8.1 Hydraulic Pressure Gradient................................................... 113
8.2 Air Liberation.......................................................................... 115
8.3 Flashing................................................................................... 119
8.4 Cavitation................................................................................ 120
8.5 Water Hammer........................................................................ 121
8.5.1 Concepts of Water Hammer........................................ 121
8.5.2 Control and Analysis of Water Hammer..................... 124
References........................................................................................ 125
xContents

Chapter 9 Heat Exchangers.............................................................................. 127

9.1 Shell-and-Tube Heat Exchangers........................................... 127
9.2 Air-to-Water Heat Exchangers................................................ 131
9.3 Plate Heat Exchangers............................................................ 133
9.4 Heat Exchanger Fouling......................................................... 136
References........................................................................................ 137

Chapter 10 Heat Rejection................................................................................. 139

10.1 Regulatory Requirements..................................................... 139
10.2 Cooling Towers..................................................................... 139
10.2.1 Cross-Flow Mechanical Draft Cooling
Towers�������������������������������������������������������������������� 139
10.2.2 Cross-Flow Natural Draft Cooling Tower.............. 140
10.2.3 Splash-Type Cooling Tower Fill Material.............. 141
10.2.4 Counter-Flow Mechanical Draft Cooling
Tower��������������������������������������������������������������������� 142
10.2.5 Counter-Flow Natural Draft Cooling Tower........... 143
10.2.6 Thin Film-Type Cooling Tower Fill Material......... 143
10.3 Spray Ponds.......................................................................... 145
10.3.1 Advantages of Spray ponds..................................... 145
10.3.2 Conventional Flat Bed Spray Ponds....................... 145
10.3.3 Oriented Spray Cooling System............................. 146
10.3.4 Analysis of Spray Ponds......................................... 147
10.3.5 Spray Pond Test Results......................................... 147
References........................................................................................ 148

Nomenclature��������������������������������������������������������������������������������������������������������� 149
Index���������������������������������������������������������������������������������������������������������������������� 153
Preface
Although this book draws heavily on lessons learned from the authors’ experience in
the nuclear power industry, they are equally applicable to many other types of power
plants as well as to industrial plants. From 1945 to 1975, when most of the existing
electric power plants were designed and constructed, engineering of the plant cooling
water system was considered a low priority item. These systems made little or no
contribution to the actual production of electricity and their failures rarely resulted in
a plant outage. Established engineering handbooks such as the Cameron Hydraulic
Data Book and the Hydraulic Institute Standards were employed in a cook book
fashion to arrive at more-or-less standard designs. All of that started to change as the
current fleet of nuclear power plants began operating in the 1960s and 70s.
The Tennessee Valley Authority (TVA) entered the nuclear power industry in
1966 by beginning construction of Unit 1 of the Browns Ferry Nuclear Plant, an
essentially turnkey 1,100 MWe boiling water reactor plant designed by General
Electric. The TVA fleet was expanded with the start of construction of Unit 1 at
Sequoyah (1970), followed by Watts Bar Unit 1 (1973). In September 1973, the TVA
Division of Engineering Design was reorganized with design projects dedicated to
each nuclear project and engineering branches to support the projects. As part of that
reorganization, Charles F. (Chuck) Bowman, a graduate of the University of
Tennessee (BSME in 1964 and MSME in 1966), was selected to supervise a section
in the Mechanical Engineering Branch that was responsible for engineering the cool-
ing water systems for all of the TVA nuclear plants. Chuck is indebted to the out-
standing engineers who worked for him during that period and developed much of
the technology described herein. In 1986, he was promoted to be the corporate spe-
cialist for thermal performance and cooling water systems, supporting all seven of
the TVA operating nuclear units.
In 1976, when Chuck reviewed the results of preoperational flow tests conducted
on Browns Ferry Unit 1, he realized that these systems had major problems. The tests
showed that critical heat exchangers were not receiving the required cooling water
flow. What followed was years of learning lessons of how to deal with issues such as
microbiologically induced corrosion and Asiatic clams. Based on his experience and
that of others, a better way of engineering these systems for both nuclear and fossil
power plants as well as industrial plants is possible in the future.
In 1994, Chuck retired from the TVA to form Chuck Bowman Associates, Inc.,
serving the electric power industry and specializing in cooling water systems and the
analysis and testing of pumping systems, heat exchangers, cooling towers, spray
ponds, etc., as well as the thermal aspects of turbine cycle systems. The purpose of
this book is to impart to the next generation knowledge and experience received dur-
ing more than 50 years of working in the field of cooling water systems.
The term cooling water systems (sometimes referred to as service water systems)
refers principally to the systems required to cool the many heat exchangers used in
all power and industrial plants. Although many of the lessons learned also apply to
the large condenser circulating water system that condenses the steam coming from
the main condenser, that system is not the primary emphasis of this book.
xi
Authors
Charles F. (Chuck) Bowman, P.E., is the president of Chuck Bowman Associates,
Inc. (CBA), an engineering consulting firm serving the electric power industry since
1994. Chuck received his BS and MS degrees in Mechanical Engineering from the
University of Tennessee, and he is a registered professional engineer in Tennessee.
CBA specializes in engineering analysis of electric power generating cycles and
related fields including the design and analysis of heat exchangers, cooling towers,
spray ponds, cooling water systems, etc. Before forming CBA, Chuck was with the
Tennessee Valley Authority (TVA) for 28 years where he supervised the engineering
of the cooling water systems for all of the TVA nuclear plants. Prior to his retirement
from TVA, he was the senior engineering specialist for thermal performance and
cooling water systems in TVA’s Corporate Engineering Office. Chuck has served as
a consultant to the Electric Power Research Institute (EPRI). He is the author of
EPRI Report No. 1007248, Alternative to Thermal Performance Testing and/or Tube-
Side Inspections of Air-to-Water Heat Exchangers; EPRI Software Manual No.
3002005344, Turbine Cycle Equipment Evaluation Workbook, and was a paid
reviewer of EPRI Report No. 1021065, Heat Exchanger Performance Analysis.
Chuck serves on the American Nuclear Society (ANS) Working Group ANS 2.21,
Criteria for Assessing Atmospheric Effects on the Ultimate Heat Sink, and has served
on American Society of Mechanical Engineers (ASME) committees that authored
ASME PTC 23.1 – 1983, Code on Spray Cooling Systems and ASME PTC 12.5 -
2000, Single Phase Heat Exchangers and is a contributing editor to Marks’ Handbook,
12th Edition. His most recent publication is Thermal Engineering of Nuclear Power
Stations: Balance-of Plant Systems, published by CRC Press.

Seth N. Bowman received his BS and MS degrees in Chemical Engineering from the
University of Tennessee. He currently serves as the senior manager of the corrective
action program at the Y-12 National Security Complex, managed and operated by
Consolidated Nuclear Security, LLC. In prior roles, Seth has been responsible for the
assessment function for the Engineering Division at Y-12 and also served as a shift
technical adviser and shift manager for Building 9212. He is co-author of Thermal
Engineering of Nuclear Power Stations: Balance-of Plant Systems, published by
CRC Press.

xiii
List of Abbreviations and
Acronyms
ACRS Advisory Committee on Reactor Safeguards
ASME American Society of Mechanical Engineers
AWHX air-to-water heat exchanger
AWWA American Water Works Association
BCDMH bromochlordimethylhydantoin
BFNP Browns Ferry Nuclear Plant
BTA best technology available
C Colbert
CCW condenser circulating water
CFU colony-forming units
CGS Columbia Generating Station
CML cement-mortar lining
CP&L Carolina Power & Light
CPP concrete pressure pipe
CS carbon steel
CU Cumberland
CWA Clean Water Act
EECW emergency equipment cooling water
EPA Environmental Protection Agency
ERCW essential raw cooling water
FBSP flat bed spray pond
FRP fiberglass-reinforced plastic
G Gallatin
HDPE high density polyethylene
HX heat exchanger
HXs heat exchangers
IAM Institute for Applied Microbiology
IN Information Notice
INPO Institute of Nuclear Power Operations
IPS intake pumping station
JS John Sevier
K Kingston
MC main condenser
MIC microbiologically induced corrosion
MPY mils per year
NDCT natural draft cooling tower
NDE nondestructive examination
NPDES National Pollution Discharge Elimination System
NRC Nuclear Regulatory Commission
OSCS oriented spray cooling system

xv
xvi List of Abbreviations and Acronyms

PDI pre-lined ductile iron

PHX plate heat exchanger
PVC polyvinyl chloride
QA quality assurance
RCW raw cooling water
RT radiographic testing
SER Safety Evaluation Report
SI standard international
SM&E Singleton Materials & Engineering Laboratory
SNP Sequoyah Nuclear Plant
SRB sulfur-reducing bacteria
SS stainless steel
STHX shell-and-tube heat exchanger
SWAP Service Water Assistance Program
TEMA Tubular Exchange Manufacturer’s Association
TRC total residual chloride
TVA Tennessee Valley Authority
UT ultrasonic testing
WB Watts Bar
WBNP Watts Bar Nuclear Plant
WBT wet-bulb temperature
WC Widows Creek
 1 Introduction

1.1 BACKGROUND
The overall operating and safety record of the existing nuclear industry is outstand-
ing, but as one will see in this introductory chapter, the record of the engineering of
the cooling water systems has been less than sterling, as indicated by the public
records maintained by the Nuclear Regulatory Commission (NRC). The purpose of
this chapter is not to point fingers or rehash old problems but to learn from them so
that the engineering of future industrial and electric power plants, both fossil and
nuclear, will not repeat the same mistakes.

1.2 TYPES OF COOLING WATER SYSTEMS

The main condenser circulating water (CCW) system is not technically a “cooling”
water system, since that system serves to condense saturated steam exiting from the
low pressure turbine and that condensation process is isothermal, although the CCW
is heated in the process. For the vast majority of electric power plants, the CCW
system is either an open once-through or open recirculating system, in which case the
waste heat from the main condenser (MC) is rejected to the environment via a cool-
ing tower, principally by evaporation, and the lost water must be replaced by some
“makeup” source. In some power plants, the cooling water system operates in paral-
lel to the MC, drawing its cooling water from the CCW system and returning it
thereto. In other plants, the cooling water may be drawn directly from some river,
lake, or ocean and passes straight through the plant heat exchangers (HXs) and is
discharged back to its source or may serve as makeup to a recirculating CCW system.
In some plants, the cooling water system may be a closed recirculating system,
rejecting its heat to the environment via cooling towers, cooling pond, or spray pond.

1.3 TYPES OF COOLING WATER SYSTEM PUMPS

Cooling water pumps are either of the vertical wet pit mixed flow type with a motor
mounted above the pump column and a line shaft extending down to the pump bowl(s)
submerged in the water source or they are horizontal centrifugal pumps located in a
dry well. Horizontal centrifugal pumps are normally located inside a building, taking
suction from the CCW system or as a booster pump in the cooling water system.
Vertical mixed flow type pumps are normally located in an intake pumping station
(IPS), the structure located on the river, lake, or ocean that is the source of cooling
water. (See Figure 3.1.) Most CCW pumps are of the vertical turbine type variety, but
in some power plants with cooling towers, the CCW pumps may be horizontal cen-
trifugal pumps located in a dry well near and below the cooling tower basin.
Cooling water pump impeller, line shaft, and coupling failures are a continuing
problem area. In 1994 the NRC issued Information Notice (IN) 94-451 to inform
1
2 Engineering of Power Plant and Industrial Cooling Water Systems

licensees of a problem of high vibration levels on one of the cooling water pumps at
the Grand Gulf Nuclear Station. Disassembly of the pump showed extensive corrosion
of the carbon steel (CS) bolts and lock washers used in the pump shaft coupling
assembly. A survey of cooling water pump failures conducted by Operating
Experience Digest between 1998 and 2005 reported more than one failure per year,
with several nuclear plants experiencing more than one failure. The most frequent
cause was reported to be corrosion of shaft and/or bolting material. Pump performance
degradation occurs due to normal wear, requiring clearance adjustments or overhaul
of the pump impeller/bowl assembly.

1.4 INTAKE PUMPING STATIONS

Although the design of an IPS is a function of the geography of the power plant site,
almost all designs incorporate devices to restrict the inflow of unwanted material.
The first barrier is normally a trash rack with a rake for periodically removing
unwanted materials. Beyond the trash rack might be a stop log to permit the unwater-
ing of the sump. The next barrier would be a traveling water screen with a screen
wash pump that takes suction downstream of the traveling screen. Finally, there may
be an assortment of pumps including the CCW pumps, the cooling water pumps,
high pressure fire protection pumps, cooling tower makeup pumps, and miscella-
neous service water pumps. The sump for each pump should be designed to promote
smooth flow of water to the individual pump suction. Normally, the motors driving
these vertical wet pit pumps are located on the deck above.
Overwhelming of the features engineered to permit the free flow of water to the
pumps and protect them from the intrusion of foreign objects such as debris, fish and
other aquatic animals, algae, marsh grass, seaweed, leaves, ice, and cooling tower fill
material has been a major impediment to safe and reliable operation of service water
systems. In 1989 the NRC initiated Generic Letter 89-13,2 a review of cooling water
system events/failures that indicated that approximately 10 percent of these were due
to foreign material intrusion. In 1990, NRC issued IN 92-493 that reported on debris
accumulation on the traveling water screens at the following nuclear stations:

• Fitzpatrick Nuclear Plant, debris caused shear pins on both screens to fail and
causing the screen to bow inward and allowing some of the debris to pass
around the screens
• Millstone Nuclear Plant Unit 1, debris caused three of the five screens to collapse
• Arkansas Nuclear One, debris caused the screens to fail causing the screen to
bow inward and allowing the debris to pass around the screens, clogged strain-
ers on the discharge of the cooling water pumps on both operating loops.

More recently, nuclear plants have reported a large ingress of marsh grass at Hope
Creek and Oyster Creek Generating Stations, a massive algae problem at Pickering
Nuclear Plant, and frazil ice blocking the intakes of other plants. (Frazil ice is a
collection of loose, randomly oriented ice crystals formed in supercooled turbulent
water.) Clearly, the current design of intake structures makes them vulnerable to site-
specific intake blockage scenarios.
Introduction 3

1.5 COOLING WATER SYSTEM PIPING AND VALVES

With the exception of CCW piping, the vast majority of cooling water system piping
is CS. CCW piping, which is huge by comparison with cooling water system piping,
is typically concrete pressure pipe (CPP) and/or square cast-in-place concrete except
above ground within the IPS and turbine building where CS is used. Due to dismal
experience with using CS piping in cooling water systems, CS piping is being
replaced piecemeal with materials such as stainless steel (SS), copper, polyvinyl
chloride (PVC), fiberglass-reinforced plastic (FRP), and high-density polyethylene
(HDPE). In situ cement mortar lining of large buried CS piping has proven to be
effective but is not widely in use. In some instances, epoxy coatings have been
applied with mixed results. The surface must be prepared to bright metal, and any
imperfection results in preferential galvanic pitting corrosion attack. CS valves,
being thicker and of more robust material, generally provide better service in cooling
water systems than CS piping when unobstructed.
The problem of cooling water system piping leaking was recognized by the NRC
with issuance of Bulletin 80-244 as early as 1980 when significant multiple piping
leaks at the Indian Point Nuclear Power Station Unit 2 resulted in the discovery of
several inches of water on the floor inside the containment of the reactor. In 1985, the
NRC issued IN 85-305 to report that pinhole leaks were discovered in SS cooling
water piping at Unit 2 of the H. B. Robinson Nuclear Plant. In 1994, leaks were
reported in the cooling water piping serving the emergency diesel generators at the
Connecticut Yankee and Beaver Valley Nuclear Power Plants. In IN 94-79,6 the NRC
reported that these leaks were due to microbiologically induced corrosion (MIC). In
2006, a leak was detected in a portion of the cooling water system piping buried at
Oyster Creek Generating Station. As a result of these and other cooling water system
piping failures, the NRC issued Generic Letter 90-057 as guidance for performing
temporary repairs of the piping.
In 1983, the NRC reported in IE IN 83-468 that seven of eight motor-operated but-
terfly valves in the cooling water system at the Surry Power Station failed to open
during a test. The failures were attributed to marine growth, among other reasons. In
1994, The NRC reported that 15 of 31 cooling water system check valves were found
to be potentially inoperable due to the accumulation of silt and corrosion products at
the Fitzpatrick Nuclear Plant.9 In 1991, the NRC issued report IN 94-6110 stating that
the stem assembly on a gate valve in the cooling water system at the Farley Nuclear
Plant failed due to corrosion.
In 1985, the NRC issued IN 85-2411 to report the discovery of delamination and
peeling of the epoxy lining of the Palo Verde Nuclear Generating Station’s cooling
water system. In 1997, the NRC issued Licensee Event Report 97-03712 on pieces of
the PVC liner material found in the strainer basket serving the emergency diesel
generator HXs at the Millstone Nuclear Power Station, Unit 2.

1.6 COOLING WATER SYSTEM FLOW DEFICIENCIES

In 1981, the NRC issued Bulletin 81-0313 to report that the Resident Inspector dis-
covered that there was inadequate cooling water flow to the reactor containment
4 Engineering of Power Plant and Industrial Cooling Water Systems

cooling units at Unit 1 of the Arkansas Nuclear One, due to plugging of the heat
exchanger (HX) by Asiatic clams (Corbicula). Also in 1981, the NRC issued
IN 81-2114 to report on a growth of some form of sea mollusk that was discovered on
the cooling water piping serving the component cooling HX in Unit 1 at the San
Onofre Nuclear Generating Station. The mollusks impaired movement of the but-
terfly valves used to isolate the HX, reduced the cross-sectional diameter of the cool-
ing water piping, and blocked the HX tubes. As a result, the cooling water flow rate
to the HX was reduced. In that same year, the NRC issued IN 81-2114 to report that
the baffle plate in the U-tube type residual heat removal HX at Unit 1 of the Brunswick
Nuclear Plant was damaged, allowing the cooling water to bypass the tube bundle. A
flow test indicated that the HX was unable to remove the decay heat because oyster
shells had blocked the inlet side of the HX tubes. Similar events occurred with the
reactor building closed cooling water HX at the Pilgrim Nuclear Power Station. In
1986, the NRC issued IN 86-9615 to report that the HXs cooling the gearbox on the
charging pumps that are part of the emergency reactor core cooling system at Unit 1
of the Farley Nuclear Plant failed to adequately cool the gearboxes, causing them to
overheat due to silt in the cooling water system. A similar event occurred at Unit 2 of
the Salem Nuclear Power Plant.
In 1988, the NRC issued IN 88-3716 to report that when Unit 2 of the Catawba
Nuclear Plant was at 20% power after the first refueling outage, the feedwater
regulating valve failed open, causing a main turbine to trip off line due to a high-high
level in one of the steam generators. When both auxiliary feedwater pumps started
automatically, low suction pressure resulted in an automatic swap over to the cooling
water system, resulting in raw cooling water (RCW) from Lake Wylie being pumped
into the steam generators. When the water level in the steam generators dropped to
their low-low level set point, the reactor tripped automatically. After normal feedwater
flow was established, the auxiliary feedwater control valves were found to be clogged
with Asiatic clams. A similar event occurred at the Davis-Besse Nuclear Power
Station in 1985. Catawba has also experienced serious corrosion and debris buildup
in their cooling water system, challenging the ability of the system to provide the
required flow during an accident.
In 1989 the NRC issued IN 89-7617 concerning a presentation at an EPRI confer-
ence by representatives of the Ontario Ministry of the Environment, Detroit Edison
Company, and the U.S. Fish and Wildlife Service on zebra mussels. Bio-fouling due
to zebra mussels was observed at several power plants, water treatment plants, and
industrial facilities along Lake Erie where they threatened operation of the condens-
ers and cooling water and fire protection systems.
Also in 1989, the NRC issued IN 90-3918 to report on inadequate cooling water
flow to safety-related HXs at several nuclear stations. At the Farley Nuclear Plant,
flow was inadequate to certain safety-related HXs and required operator action. At
the Peach Bottom Power Plant, Unit 2, flow was found to be inadequate due to the
accumulation of silt and corrosion products in the cooling water piping. At Fitzpatrick
Power Plant, silt was found in check valves in the cooling water system lines to the
seal water coolers for residual heat removal pumps that could have prevented the
system from performing its safety function.
Introduction 5

1.7 HEAT EXCHANGER FOULING

In 1986, the McGuire Nuclear Station reported to the NRC that over the years a com-
bination of organic and inorganic compounds have fouled a number of HXs, includ-
ing the containment spray, component cooling water, and control room chillers.15 In
1996 in Licensing Event Report 96-001-0119 the staff at Unit 1 of the Calvert Cliffs
Nuclear Power Plant reported to the NRC that they had determined that their safety-
related HXs served by the cooling water system may not have been capable of meet-
ing their intended safety function during periods of high Chesapeake Bay water
temperatures, due to microfouling on the tube side of the HX. In 2007, the NRC
issued IN 2007-2820 to report that when the Palo Verde Nuclear Generating Station
noted elevated temperatures in the intake air for one of the emergency diesel genera-
tors, a lotion-like substance was found to be fouling the cooling water side of the
diesel generator HXs.

1.8 ENGINEERING DESIGN ERRORS

In 1990, the NRC issued IN 90-2621 to report that after conducting extensive flow and
pressure drop tests on the cooling water side of room coolers at the Clinton Power
Station, the staff found that the relationship between the flow and pressure drop pro-
vided by the manufacturer for these room coolers was wrong and that the actual
pressure drop was greater than advertised. The same manufacturer had supplied
room coolers to approximately 50 nuclear power plants.
A survey of cooling water pump failures conducted by Operating Experience
Digest concluded that the pump failures were due to the specification of incorrect
bolting and shaft material, improper heat treatment, and the use of dissimilar metals
resulting in galvanic corrosion. Some pumps were designed with the resonant
frequency of the line shaft being too close to the operating speed of the pump.
In 1996, the NRC issued Generic Letter 96-0622 when the staff at the Diablo
Canyon Power Plant determined that the component cooling water serving the reactor
containment air coolers located high inside the reactor building’s containment could
flash to steam in the cooling coils during an accident. The discovery of this engineering
design error caused the Westinghouse Electric Corporation to issue a nuclear safety
advisory letter. The same problem was identified at the Connecticut Yankee Nuclear
Power Plant where the containment air coolers are served by the cooling water
system. In that same year, a similar condition was discovered at the Point Beach
Nuclear Plant and at Salem Nuclear Power Plant. When the cooling water flashes to
steam, the flow rate is reduced due to the higher pressure drop through the HX and
heat removal capability is reduced. In the cases of Diablo Canyon and Salem, cooling
water flow ceased completely following the accident for a period of time while
pumps were restarted. At Salem, valves were required to operate to isolate non-
safety-related HXs. Since the peak containment pressure occurs within seconds
following an accident such as a pipe break, the cooling water would not be restored
in time to limit the peak containment pressure. Further, an analysis of the cooling
water system flows at Connecticut Yankee concluded that restoring the cooling water
6 Engineering of Power Plant and Industrial Cooling Water Systems

flow would result in hydrodynamic loads due to water hammer that would exceed the
piping and piping support structural limits.

1.9 NRC OVERSIGHT
In 1981, the NRC issued Bulletin 81-0313 to require nuclear plant licensees to deter-
mine whether or not Asiatic clams or other mussels were present in their cooling
water systems and if present to confirm that the individual components were receiv-
ing adequate cooling water flow. The response to that bulletin indicated that bivalves
were present in approximately 45% of the nuclear plant sites in the United States.
Serious fouling was reported at several sites as indicated above. Up until that time,
some nuclear stations had not ever verified that the HXs were receiving adequate
cooling water flow.
In 1988, the NRC published NUREG 1275. Vol. 3,23 a comprehensive review and
evaluation of cooling water system failures and degradations reported in nuclear
plants from 1980 to 1987. The report identified 29 events involving significant degra-
dation or failures of the cooling water system, an average of 3.6 events a year.22
In 1989, the NRC issued Generic Letter 89-132 requiring nuclear plant licensees
to do the following:

I. For open-cycle service water systems, implement and maintain an ongoing

program of surveillance and control techniques to significantly reduce the inci-
dence of flow blockage problems as a result of bio-fouling.
II. Conduct a test program to verify the heat transfer capability of all safety-
related HXs cooled by service water. The total test program was to consist of
an initial test program and a periodic retest program of HXs connected to or
cooled by one or more open-cycle systems.
III. Ensure by establishing a routine inspection and maintenance program for
open-cycle service water system piping and components that corrosion, ero-
sion, protective coating failure, silting, and bio-fouling cannot degrade the per-
formance of the safety-related systems supplied by service water.
IV. Confirm that the service water system would perform its intended function in
accordance with the licensing basis for the plant.
V. Confirm that maintenance practices, operating and emergency procedures, and
training involving the service water system are adequate to ensure that safety-
related equipment cooled by the service water system would function as
intended and that operators of this equipment will perform effectively.

It should be noted that Generic Letter 89-13 did not require the nuclear plants to
conduct flow balance tests of their cooling water systems. For example, in 1990,
when Clinton Power Station measured the actual flows to their room coolers, they
found that the flows were from 10% to 80% less than the design flows.21 Flows to
other HXs ranged from 2% to 42% less than the design flows, while other components
served by the cooling water system were receiving excess flow up to 213% of the
design value.21
Introduction 7

In 1992, the NRC issued Temporary Instruction 2515/11824 and initiated a pro-
gram of conducting operational performance inspections of the cooling water sys-
tems for nuclear plants the United States. The inspections included a review of the
mechanical systems engineering design and configuration control and nuclear plant
operations, maintenance, and testing and quality assurance (QA)/corrective actions.
Although the findings of these inspections are too numerous to enumerate here, they
did instill in the nuclear industry an awareness of the importance of cooling water
systems to nuclear safety. Additional inspections are ongoing.
In 2007, the NRC issued IN 2007-06 25 to report on potential common cause vul-
nerabilities in essential cooling water systems. Two of the nuclear stations were out-
side the United States. In the first such case, a manhole pipe broke; in the second a
cooling water pipe ruptured, flooding the piping gallery with sea water. In the United
States, a 3 gallons per hour through-wall leak in a 30-inch (76.2 cm) pipe was found
at Unit 2 of the South Texas Project due to cavitation damage downstream of a heav-
ily throttled butterfly valve, and a through-wall leak was found in the endbell of one
of the emergency diesel generator HXs due to MIC.

1.10 AUTHORS’ PERSPECTIVE
Over the years, workable solutions to many of the deficiencies in the engineering of
cooling water systems have been developed by the authors and others, but in many
cases, these solutions have not been widely disseminated. It is the hope of the authors
that this book will aid in imparting to the next generation the knowledge and experi-
ence gained during more than 50 years of working in the field of cooling water
systems.

REFERENCES
1. IN 94-45. Potential Common-Mode Failure Mechanism for Large Vertical Pumps,
Nuclear Regulatory Commission, Washington, D.C., 1994.
2. Generic Letter 89-13. Service Water System Problems Affecting Safety-Related
Equipment, Nuclear Regulatory Commission, Washington, DC, 1989.
3. IN 92-49. Recent Loss or Severe Degradation of Service Water Systems, Nuclear
Regulatory Commission, Washington, DC, 1992.
4. Bulletin 80-24. Prevention of Damage Due to Water Leakage Inside Containment
(October 17, 1980 Indian Point 2 Event), Nuclear Regulatory Commission, Washington,
DC, 1989.
5. IN 85-30. Microbiologically Induced Corrosion of Containment Service Water System,
Nuclear Regulatory Commission, Washington, DC, 1985.
6. IN 94-79. Microbiologically Influenced Corrosion of Emergency Diesel Generator
Service Water Piping, Nuclear Regulatory Commission, Washington, DC, 1994.
7. Generic Letter 90-05. Guidance for Performing Temporary Non-Code Repair of ASME
Code Class 1, 2, and 3 Piping, Nuclear Regulatory Commission, Washington, DC,
1990.
8. IE IN 83-46. Common-Mode Valve Failures Degrade Surry’s Recirculation Spray
Subsystem, Nuclear Regulatory Commission, Washington, DC, 1983.
8 Engineering of Power Plant and Industrial Cooling Water Systems

9. Third EPRI Balance-of-Plant Heat Exchanger NDE Workshop: A Regulatory

Perspective, Davis, J. A., et al, Electric Power Research Institute, Charlotte, NC, 1994.
10. IN 94-61. Corrosion of William Powell Gate Valve Disc Holders, Nuclear Regulatory
Commission, Washington, DC, 1994.
11. IN 85-24. Failure of Protective Coatings in Pipes and Heat Exchangers, Nuclear
Regulatory Commission, Washington, DC, 1985.
12. Licensee Event Report 97-037. Service Water Piping Liner Material Peeled Off and
Lodged in the Strainer for the Emergency Diesel Generator Heat Exchangers, Nuclear
Regulatory Commission, Washington, DC, 1997.
13. Bulletin 81-03. Flow Blockage of Cooling Water to Safety System Components by
Corbicula Sp. (Asiatic Clam) and Mytilus Sp. (Mussel), Nuclear Regulatory Commission,
Washington, DC, 1981.
14. IN 81-21. Potential Loss of Direct Access to Ultimate Heat Sink, Nuclear Regulatory
Commission, Washington, DC, 1981.
15. IN 86-96. Heat Exchanger Fouling Can Cause Inadequate Operability of Service Water
Systems, Nuclear Regulatory Commission, Washington, DC, 1986.
16. IN 88-37. Flow Blockage of Cooling Water to Safety System Components, Nuclear
Regulatory Commission, Washington, DC, 1988.
17. IN 89-76. Bio-fouling Agent: Zebra Mussel, Nuclear Regulatory Commission,
Washington, DC, 1989.
18. IN 90-39. Recent Problems with Service Water Systems, Nuclear Regulatory Commission,
Washington, DC, 1990.
19. Licensing Event Report 96-001-01. SRW Heat Exchanger Micro-fouling Higher Than
Assumed in Design Basis, Nuclear Regulatory Commission, Washington, DC, 1996.
20. IN 2007-28. Potential Common Cause Vulnerabilities in Essential Service Water Systems
due to Inadequate Chemistry Control, Nuclear Regulatory Commission, Washington,
DC, 2007.
21. IN 90-26. Inadequate Flow of Essential Service Water to Room Coolers and Heat
Exchangers for Engineered Safety-Feature Systems, Nuclear Regulatory Commission,
Washington, DC, 1990.
22. Generic Letter 96-06. Assurance of Equipment Operability and Containment Integrity
during Design-Basis Accident Conditions, Nuclear Regulatory Commission,
Washington, DC, 1996.
23. NUREG 1275. Vol. 3. Operating Experience Feedback Report – Service Water System
Failures and Degradations, Nuclear Regulatory Commission, Washington, DC, 1989.
24. Temporary Instruction 2515/118. NRC Inspection Manual: Service Water System
Operational Performance Inspection, Nuclear Regulatory Commission, Washington,
DC, 1992.
25. IN 2007-06. Potential Common Cause Vulnerabilities in Essential Service Water
Systems, Nuclear Regulatory Commission, Washington, DC, 2007.
 2 Properties of Water

2.1 TEMPERATURE
There are four temperature scales in common use, two in English units and two in
standard international (SI) units, The English units are degree Fahrenheit (°F) and
degree Rankine (°R). The SI units are degree Celsius (°C) and degree Kelvin (°K).
Figure 2.1 shows the temperatures in English units.
Figure 2.2 shows the temperatures in SI units. Virtually all engineering of cooling
water systems is performed in either Fahrenheit or Celsius units. The formula for
converting from SI units to English units is shown in Equation 2.1

F  1.8  C  32 (2.1)

2.2 PRESSURE
As seen in Figure 2.3, pressure may be measured in three different ways: (1) gauge
pressure (i.e. pressure above atmospheric pressure), (2) vacuum pressure (i.e. pres-
sure below atmospheric pressure), and (3) absolute pressure (i.e. pressure above zero
pressure or a perfect vacuum).
Perhaps one of the most common mistakes made in any power plant is confusing
gauge pressure with absolute pressure. Most gauges inside a power plant measure
absolute pressure. Since the turbine cycle is isolated from the atmosphere, gauge
pressure is largely irrelevant. Also, the properties of steam and condensate in a
turbine cycle are related to absolute pressure. However, there may be the odd gauge

Boiling at atmospheric pressure

212 ºF

Freezing at atmospheric pressure

32 ºF
0 ºF

–459.67 ºF

0 ºR (absolute zero)

FIGURE 2.1 Temperatures measurements in English units.

9
10 Engineering of Power Plant and Industrial Cooling Water Systems

Boiling at atmospheric pressure

100 ºC

Freezing at atmospheric pressure

-273.15 ºC

0 ºC (absolute zero)

FIGURE 2.2 Temperatures measurements in SI units.

Pressure

Gauge pressure

Absolute pressure Atmospheric pressure

Vacuum pressure

Zero pressure

FIGURE 2.3 Pressure measurements.

that reads in gauge pressure. There also may be an occasional instrument that mea-
sures vacuum. Be warned! In dealing with cooling systems, almost all instruments
read gauge pressure. The units lbf/in2 (English) and kPa (SI) refer to gauge pressure
herein. As will be seen later, much of the hydraulic analysis of cooling water systems
is performed in units of “head” in feet of water. For most cooling water systems,
1.0 lbf/in2 is equal to 2.31 ft of head.

2.3 DENSITY
The properties of water, whether in the form of ice, liquid, or steam, depend on the
state of the water and are independent of the path that was taken to achieve the state.
For example, the properties (e.g. temperature, pressure, etc.) of a volume of liquid
water are independent of whether it was condensed in a cloud or melted from ice.
Properties may be either extensive or intensive. For example, a bucket of water
Properties of Water 11

having a mass (the amount of matter contained in a substance), volume, pressure, and
temperature might be divided equally into two buckets. The water in each bucket
might still have the same pressure and temperature, but the mass and volume in each
bucket would be one half of that contained in the original bucket. The mass and vol-
ume are said to be extensive properties, while the pressure and temperature are said
to be intensive properties, because they are the same for each small element of water
in the bucket. However, as shown in Equation 2.2, if one divides the mass by the
volume, one gets density (an intensive property).

m
  (2.2)
V

where

ρ = density
m = mass
V = volume.

2.4 VISCOSITY
The viscosity of water flowing in a pipe is a measure of its resistance to flow. There
are two measures of viscosity: dynamic (or absolute) and kinematic. Water is said to
be a “Newtonian” fluid, meaning that its dynamic viscosity is a function of tempera-
ture alone. Dynamic viscosity is used to calculate the Reynolds number and Prandtl
number of water. Kinematic viscosity is simply the dynamic viscosity divided by the
density of the water. Dynamic viscosity is an indication of the pressure drop required
to pass the water through a pipe, and kinematic viscosity is an indication of how fast
the water is moving when that pressure is applied.

2.5 REYOLDS NUMBER
The Reynolds number of flow in a pipe is a non-dimensional property characterizing
the flow as laminar or turbulent. The Reynolds number depends on the liquid veloc-
ity, the viscosity, and the pipe diameter and is calculated as follows:

Vp di
Re  (2.3)


where
Vp = water velocity in a pipe or tube
di = inside diameter of a pipe or tube
μ = dynamic viscosity of the water.

Figure 2.4 shows the Moody diagram relating the Reynolds number to the friction
factor.1 If the Reynolds number is less that 2000, the flow is said to be laminar. If the
number is greater than 4000, it is said to be turbulent in the pipe or tube. Between
12 Engineering of Power Plant and Industrial Cooling Water Systems

FIGURE 2.4 Moody diagram.

these values, transitional flow exists. If one knew the relative roughness of the pipe,
one could calculate the pressure drop. Unfortunately, for CS cooling water piping,
determining the relative roughness is a function of the tuberculation that frequently
occurs in the pipe due to MIC, etc. In that case, the only major benefit derived from
the Moody diagram is to verify that the flow is turbulent. Although in piping systems
the flow is almost always turbulent, in some HX applications the flow may be lami-
nar, affecting both the pressure drop through the HX and the rate of heat transfer.

2.6 PRANDTL NUMBER
The Prandtl number is a key parameter in determining heat transfer in a HX. The
Prandtl number is defined as the ratio of the momentum diffusivity to the thermal
diffusivity and depends on the dynamic viscosity, specific heat, and thermal conduc-
tivity of the water and is calculated as follows:

 cp
Pr  (2.4)
k

where

Pr = Prandtl number
cp = specific heat
k = thermal conductivity of water.
Properties of Water 13

2.7 SPECIFIC GRAVITY
The specific gravity of water is the ratio of the density of water at a given temperature
to that of water at a reference temperature. In English units, the reference tempera-
ture is 60°F (15.6°C), and the density of water is 62.36 lb/ft3 (1,000 kg/m3).

REFERENCE
1. Moody, L. F., Friction Factors for Pipe Flow. ASME Transactions, Vol. 66, pp. 671–684,
1944.
 3 Intake Structures

3.1 INTAKE STRUCTURE DESIGN

The intake structure is a very important component of a cooling water system. Good
engineering practice in the design of vertical turbine mixed flow or axial flow pumps
in rectangular wet-pit intake structures is well established as a function of the pump
bell diameter, D, by the Hydraulic Institute Standards.1 Figure 3.1 shows a typical
rectangular wet-pit intake structure. The pump bay width, W, and depth, H, should be
sufficient to limit the maximum pump approach velocity to 1.0 to 1.5 ft/sec (0.30 to
0.46 m/s).1,2 The minimum submergence, S, required to avoid vortices is a function
of the dimensionless Froude number defined as follows1:

VS
FD = (3.1)
gD

where
FD = Froude number
VS = velocity at suction inlet
D = pump bell diameter
g = gravitational constant

The minimum submergence, S, is as follows:

Smin  D  1  2.3 FD 

where the units must be consistent to yield a dimensionless Froude number and S is
in the same units as D.
Although among pump types and manufacturers there may be some variation in
the bell velocity (i.e. flow rate divided by the pump bell area), this is of secondary
importance, since basing the dimensions on the pump bell diameter would ensure
geometric similarity of flow patterns.1
The intake should supply an evenly distributed flow of water to the suction bell, as
an uneven distribution of flow may result in vortices that may introduce air into the
pump, resulting in reduced pump capacity.1 The amount of submergence for success-
ful operation depends upon the design of the approach to the pump intake and the
size of the pump.1 Dimension “H” is based on “minimum normal water level”, taking
into consideration normal operating friction losses through intake channel, trash
rack, and the traveling screen.1
If the bottom elevation of the sump must be below the source of the cooling water
to meet submergence requirements, the slope of the floor of the intake structure
should not be greater than 10 degrees.1

15
16 Engineering of Power Plant and Industrial Cooling Water Systems

Trash Stop Traveling Screen Wash Cooling Water

Racks Log Screen Pump Pump Motor

W/2

W
A

Trash
Sluice
A
Plan View

Screen Wash Pump

Orifice Plug & Air Release Valve
Pump Motor Motor Butterfly
Check
Traveling Valve
Valve
Screen Motor
High Low
Water Water
Level Level
B

A
Y

H S
B

D
C

Elevation View

FIGURE 3.1 Intake pumping station.

The recommended values dimensions A, B, C, H, S, and Y as a function of “D”

are shown in Table 3.1 from Reference 1.
It is noted that Figure 1 in Reference 2 recommends B = 0.8D and C = 0.5D.
Of course the IPS frequently must be designed long before the cooling water
pump is selected. In that case, Figure 81 of Reference 3 may be consulted, where the
sump dimensions are a function of the pump flow on a log-log scale.3 Table 3.2 indi-
cates some examples of approximate dimensions from Reference 3.
Intake Structures 17

TABLE 3.1
Sump Dimensions as a Function of Pump Bell Diameter
Dimension Number of Diameters
A 5D
B 0.75D
C 0.3D to 0.5D
H S+C
S See Equation (3.1)
Y 4D

TABLE 3.2
Sump Dimensions vs. Pump Flow
Dimension Pump Flow Sump Dimension Pump Flow Sump
(gal/min) (in) (gal/min) Dimension (in)
A 3,000 87 10,000 165
B 3,000 14 10,000 26
C 3,000 13 10,000 19
H 3,000 41 10,000 58
W* 3,000 34 10,000 61
Y 3,000 60 10,000 108

* In Figure 81 of Reference 3, the sump width is noted as S, not W.

Assuming a pump bell velocity of 5.0 ft/sec ( 1.524 m/s), the sump dimensions in
Table 3.2 are in good agreement with those calculated from Table 3.1 with the excep-
tion of the value for C, which is considerably higher than that calculated from
Table 3.1 for 3,000 and 10,000 gal.min pumps (208% and 166%, respectively).
Of course many if not most IPS are considerably more complex than that shown
in Figure 3.1. In many cases, a single IPS houses both the CCW pumps and the cool-
ing water pump and may also house high pressure fire protection pumps and service
water pumps. Figure 3.2 illustrates such an IPS.
Preferably water should not flow past one pump to reach the next pump.3 If such
an arrangement is required, dividing walls should be located around each pump to
provide conditions similar to those in Figure 3.1, and/or turning vanes may be
required under the pump to deflect the water upward.3
18 Engineering of Power Plant and Industrial Cooling Water Systems

FIGURE 3.2 Intake pumping station with CCW pumps and cooling water pumps.

3.2 INTAKE STRUCTURE DESIGN DEFICIENCIES

As documented in Section 1.3, the failure of IPS to effectively protect cooling water
systems from the intrusion of all manner of debris can be a serious problem. In 2007,
the Institute of Nuclear Power Operations (INPO) issued a topical report on intake
structure blockage vulnerabilities indicating that these events were continuing to
happen. Many events result in power reduction or unit shutdown. Therefore, notwith-
standing the impact on nuclear safety, a business case may be made for enhancing the
design of the IPS to minimize the likelihood and the consequences of a blockage.
In many instances due to personnel turnover, power plant operators are unaware
of prior events such as fish runs, seaweed, etc., that have challenged the cooling water
system in the past. Operators are forced to rely on periodic walk-downs to detect
such serious problems. When these events occur unexpectedly, the trash racks and
traveling water screens can be quickly overwhelmed, leading to structural failure
during environmental events such as storms. In very cold areas, frazil ice can quickly
form on the screens. In other areas, oil spills can quickly coat the HXs served by the
cooling water system with a film that degrades their heat transfer capability.
Intake Structures 19

The screening material in many existing traveling water screens are of CS that has
been in service for decades. The ability of these screens to withstand the forces
resulting from a significant challenge is questionable.

3.3 REQUIREMENTS OF §316(B) OF THE CLEAN WATER ACT

Section 316(b) of the Clean Water Act (CWA) requires power and industrial plants
that draw water from the waters of the United States to implement the "best technol-
ogy available” (BTA) to reduce injury or death of fish and other aquatic life that may
be killed by being impinged on or entrained in the IPS and subsequently killed by
being exposed to the temperature and pressure changes associated with the CCW
system and/or the cooling water system as they pass through the plant. The
Environmental Protection Agency (EPA) estimates that the Act will affect approxi-
mately 670 U.S. power plants.4
The final EPA Rule promulgated on May 19, 2014, required new plants at existing
facilities to install closed-cycle recirculating systems or to reduce actual intake flow
to a level commensurate with that attained by a closed-cycle recirculating system.
Another alternative would be to require plants to demonstrate that they have installed
and will operate and maintain technological or other control measures that reduce the
level of adverse environmental impact to a level comparable to that achieved through
flow reductions commensurate with the use of a closed-cycle recirculating system.5
Existing plants with intake flows of over 2 million gallons of water per day that
use 25% of the water for cooling are subject to the rule, and they must implement one
of the following alternatives which may be deemed to be the BTA:

1. Install a closed-cycle recirculating heat rejection system;

2. Install an IPS with a maximum through-screen design intake velocity of 0.5 ft/
sec (0.15 m/s);
3. Operate an IPS with a maximum through-screen intake velocity of 0.5 ft/sec
(0.15 m/s);
4. Install an intake at least 800 ft (244 m) offshore with a velocity cap;
5. Install traveling water screens that the EPA determines meet the BTA;
6. Implement any combination of measures that the EPA determines meets the
BTA;
7. Achieve impingement mortality reduction to no more than 24%.5

These requirements are implemented through the system of required National

Pollution Discharge Elimination System (NPDES) permits.5
Alternatives 1, 2, 3, and 4 were pre-approved by the EPA as meeting BTA. The
other alternatives require proof that they are the BTA. The first and most expensive
alternative would require the utility to convert the CCW system from an open system
drawing water from a lake, river, or ocean to a closed system in which cooling towers
would be added to the plant, thus reducing the intake requirement to only that
required to replace the water evaporated from the system and the blowdown required
to control the concentration of dissolved solids in the CCW system. The fifth option,
20 Engineering of Power Plant and Industrial Cooling Water Systems

commonly proposed, is to modify or replace the existing traveling water screens with
screens designed to minimize fish impingement onto the screens. However, the Rule
specified that if the NPDES permit director determines that the modified screens are
insufficient, additional measures may be required. This option may pose greater
challenges for the reliability of the cooling water system. In a report dated November
2001, the EPA reviewed the efficacy of alternatives to conventional traveling screens
for cooling water intake structures including the following:

• modified traveling screens and fish handling and return systems

• cylindrical wedgewire screens
• fine mesh screens
• fish net barriers
• velocity caps
• aquatic micro-filtration barriers
• louver systems
• angled and modular inclined screens
• porous dikes and leaky dams
• behavioral systems.6

The EPA reported that conventional traveling screens used by approximately 60% of
power plants have 3/8-inch (0.9525 cm) mesh wire to prevent clogging of HX tubes
with screen wash at a typical pressure between 80 and 120 lbf/in (552 and 827 kPa).6
These screens are normally rotated and washed only intermittently, so fish that are
impinged on the screens by high intake velocity for extended periods of time die
there or are killed by the high-pressure wash when the screens are operating.5
Ristroph screens are conventional traveling screens that have been modified with
baskets or buckets on the front of the screens as shown in Figure 3.3 to hold fish in
the water until the screen rotates to a point where the fish are spilled onto a trough
where they are sluiced along with any trash back into the source water.6 Of course,
the screens should be continuously operating when large numbers of fish are present.

FIGURE 3.3 Ristroph traveling screen.

Intake Structures 21

Operating experience at a number of power plants has shown at least a 70-80%

reduction in impingement can be achieved over conventional traveling screens.5
Cylindrical wedgewire screens like the one shown in Figure 3.4 have been shown
to eliminate impingement at Eddystone Generation Station.6 In this design, ambient
flow is in the axial direction of the screen. As the aquatic organisms are carried past
the screen, they are restricted from entering by their inability to fit through the screen
spacing. Their momentum carries them downstream as their ability to actively
maneuver away from the screen’s hydraulic zone of influence aids in their passage.7
However, this technology has not been widely utilized in other power plants.6
One obvious solution to eliminate entrainment of fish eggs, larvae, and juvenile
fish would be to mount fine mesh screens on conventional traveling water screens.
This method was implemented at the Big Bend and Brunswick Power Plants,
suggesting that fine mesh screens can reduce entrainment by 80% or more but require
intensive maintenance when they are in use.6 Additional full-scale performance data
on fine mesh screens were not available at the time of publication of Reference 6.6
Fish net barriers are wide-mesh nets placed in the source water attached to a
floating boom in front of the entrance to intake structures where relatively low
entrance velocities exist. The mesh is sized to prevent the indigenous fish from
passing through the net. Fish net barriers have been used at numerous facilities where
they have proven effective at reducing impingement by over 80%.6 However, experi-
ence shows that high debris flows can cause significant damage to the nets, necessi-
tating frequent maintenance.
When a velocity cap is placed over a vertical inlet at offshore intakes, it converts
the vertical flow to horizontal flow which fish can more easily avoid, thus reducing
impingement.6 When the intake is extended to an area of deeper, cooler, water with
less biological density, the entrainment of aquatic organisms may also be reduced.7
(Note that installation of an intake at least 800 ft (244 m) offshore with a velocity cap
constitute BTA.)
Other than those technologies discussed above, none of the other technologies
described in Reference 6 had been implemented full-scale in power plants when the
document was published.
In 2014, Bowman proposed an alternative for reducing both impingement and
entrainment not previously considered by the EPA or others.8 The CCW systems of
many power plants were originally designed to take advantage of the abundant and

FIGURE 3.4 Cylindrical wedgewire screen.

22 Engineering of Power Plant and Industrial Cooling Water Systems

free CCW that was available from rivers, lakes, and oceans. The economic optimum
design was a high-flow, low head, system with a single-pressure MC designed with
minimal surface areas. Therefore, the potential may exist to reduce the required CCW
flow to existing power plants by redesigning and modifying the existing CCW system
to minimize intake flow based on current technology. The result could be a new
and improved MC and other turbine cycle equipment and perhaps new CCW pumps
and/or turbine rotors, resulting in the same or better plant performance. The following
is taken from Reference 9 where “NDCT” stands for “Natural Draft Cooling Tower”.9
“Alternative C – Refurbish the existing two NDCT and convert the existing MC
into a multi-pressure condenser.

Alternative C would also require the refurbishment of the existing two NDCTs.
However, in lieu of an expensive third NDCT, the existing MC would be recon-
figured such that the CCW flow through the three shells of the MC would be in
series rather than in parallel. The existing 10 foot square pressure equalizing
ducts between the three shells would be blocked, making it a multi-pressure
condenser. The CCW flow through the MC would be reduced by one-third and
the velocity through the 0.75-inch (1.91 cm) titanium tubes would be increased
from 5.5 to 10.9 ft/sec. To accommodate the higher pressure drop through the
MC tubes, the CCW Pumps would be replaced with higher head pumps and
motors. A new transformer would be provided to serve the larger motors. To
accommodate the higher head pumps, 10 foot diameter cement mortar-lined CS
piping would be installed inside of the existing 13.5’ × 13.5’ concrete conduits
between the IPS and the MC. By reducing the CCW flow rate by one-third, the
water loading on the existing NDCT would be reduced. Even though the tem-
perature rise through the MC would be increased and the hot water temperature
would be greater, the cold water temperature coming from the NDCT would
be less. For a NDCT, the hot water temperature has only a small effect on the
cold water temperature, as a higher hot water temperature increases the air flow
through the tower. The overall reduction in electrical output of this alternative is
41 MWe per unit per year or 3.5%.

A reduction on the CCW flow by one-third constitutes a significant reduction on both

impingement and entrainment and in some instances may reduce the intake velocity
to below 0.5 ft/sec (0.15 m/s), which is BTA or in other instances taken with other
measures described herein may also result in BTA. For example, a reduction of CCW
flow by one-third may reduce the cost of adding cooling towers by a similar amount.

3.4 POTENTIAL IMPROVEMENTS IN THE ENGINEERING OF

INTAKE PUMPING STATIONS
Since one of the significant challenges to the free flow of cooling water is the intru-
sion of aquatic animals, some of the technologies designed to protect aquatic life also
enhance cooling water system reliability. Offshore intakes with a velocity cap not
only reduce the intake of aquatic life but also that of algae and marsh grass. Floating
booms used to support fish nets also protect against oil slicks and security challenges.
Intake Structures 23

Skimmer walls across the intake channel that restrict intake flow to the bottom of the
channel can achieve similar results at some locations. Continuous operation of a
traveling water screen as required with fish baskets or buckets protects the cooling
water system from unexpected events that can quickly overcome the intake.
Figure 3.5 illustrates several modifications to the wet-pit intake structure shown in
Figure 3.1. The following enhancements are proposed:

Trash Stop Traveling Screen Wash Cooling Water Strainer

Racks Log Screen Pump Pump Motor

W/2

W
A

Trash Warm water recirc.

Sluice
A
Plan View

Screen Wash Pump

Orifice Strainer
Plug & Air Release Valve
Pump Motor Motor
Check Strainer
Traveling Valve Butterfly
Screen Motor Valve
High Low
Water Water
Level Level

A
Y

H S
B

D
C
Elevation View

FIGURE 3.5 Enhanced intake pumping station.

24 Engineering of Power Plant and Industrial Cooling Water Systems

FIGURE 3.6 Cooling water strainer. (Courtesy S. P. Kenney Engineers, Inc.)

• In areas where frazil ice is a hazard, design the IPS to accept recirculated warm
water from the condenser discharge;
• Provide redundant screen wash pumps so that the cooling water system is not
vulnerable to challenges when one pump or motor is out of service and so that
the second screen wash pump may be started to increase screen wash flow if
required during an intrusion;
• Provide automatic backwashing cooling water strainers at the discharge of the
cooling water pumps to strain out all but larvae, veligers (larval with ciliated
flaps for swimming) and fish eggs, etc.;
• Increase cooling water pump submergence such that the pumps would con-
tinue to operate satisfactorily even when the structural design maximum dif-
ferential pressure across the traveling screens is reached;
Intake Structures 25

• Provide an automatic start on high differential pressure across the screen if

traveling water screens are not continuously operated (recommended);
• Employ screens made of only SS mesh;
• Provide cameras to permit the operator to observe the condition of the traveling
screen and trash rack;
• Use anti-foulant paint on the interior walls of the wetted surface of the IPS if
located in an area where zebra mussels are prevalent.

As discussed in Chapters 6 and 7, straining the cooling water at the discharge of

the cooling water pump is critical for controlling biofouling and MIC. There are two
types of large automatic backwashing strainers: those that utilize wedgewire and
those that utilize individual baskets that screw into a drum like the one shown in
Figure 3.6 above which are recommended.
The pressure drop across a 24″ (61.0 cm) S. P. Kinney Model A strainer with wire
baskets rated at 16,000 gal/min (1,000 l/s) capacity was measured during the preopera-
tional testing of the safety-related cooling water system at the Sequoyah Nuclear Plant
(SNP). This test indicated a pressure drop across the strainer of 5.0 lbf/in (34.5 kPa)
which was several times more than that advertised by the strainer manufacturer.

REFERENCES
1. American National Standard for Pump Intake Design, ANSI/HI 9.8-1998, Hydraulic
Institute, Parsippany, NJ, 1998.
2. Greutink, Herman. Recommendations for Vertical Pump Intakes, Pumps and Systems
Magazine, Birmingham, AL, November, 1994.
3. Hydr aulic Institute Standards for Centrifugal, Rotary, and Reciprocating Pumps, 14th
ed., Hydraulic Institute, Parsippany, NJ, 1983.
4. Russell, Ray. EPA’s 316(b) Rule: Are You Ready? Power Engineering, September 2013.
5. 40 CFR Part 122 and 125, National Pollution Discharge Elimination – Final Regulations
to Establish Requirements for Cooling Water Intake Structures at Existing Facilities and
Amend Requirements at Phase I Facilities, Environmental Protection Agency,
Washington, DC, May 9, 2014.
6. Technical Development Document for the Final Regulations Addressing Cooling Water
Intake Structures for New Facilities, EPA-821-R-01-036, Chapter 5, Environmental
Protection Agency, Washington, DC, November, 2001.
7. Clubb, Richard. When Budgeting for §316(b) Compliance, Consider All Options. Power
Engineering Magazine, Tulsa, OK, March 2013.
8. Bowman, C. F. The Third Option for Meeting 316(b) Requirements, Power 2014-32113,
Proceedings of the ASME 2014 Power Conference, Baltimore, MD, July 2014.
9. Sequoyah Nuclear Plant §316(b) – §122.21(r)(10) – (13) Information, Appendix A to
Transmittal from TVA to Tennessee Department of Environment and Conservation, June
29, 2018.
 4 Pumps

4.1 TYPES OF PUMPS
Cooling water and CCW system pumps may be classified as either vertical wet pit
pumps as described in Chapter 3 or dry-pit pumps as discussed in Section 1.2. In
power plants with open once-through CCW systems, the CCW pumps are typically
of the vertical wet pit type as shown in Figure 3.2. In some power plants with cooling
towers, the CCW pumps may be arranged as dry-pit pumps with the shaft and motor
mounted either horizontally or vertically as shown in Figure 13 in Reference 1.
Cooling water system pumps that are supplied from a lake, river, or ocean, etc., as
shown in Figures 3.1 and 3.5 are of the vertical wet pit type. However, in some power
plants, the cooling water pumps take suction from the CCW pumps and are horizon-
tal pumps located inside the turbine building.

4.2 TYPES OF PUMP IMPELLERS

Pumps may be classified as having radial flow, mixed flow, or axial flow impellers.
Pumps having radial flow or mixed flow impellers are referred to as centrifugal
pumps, because they employ centrifugal force to sling the water in a direction per-
pendicular or almost perpendicular to the pump shaft, whereas pumps with axial flow
impellers (referred to as propeller pumps) move the flow along the axis of the pump.1
In power plants with open once-through CCW systems, the CCW pumps are typi-
cally low-head axial flow pumps. Since a siphon is established through the MC, the
CCW is only required to overcome the friction head losses through the system. In
power plants with closed CCW systems with cooling towers, the CCW pumps are
typically mixed flow pumps, because the CCW pump must overcome the static head
required to return the CCW to the cooling tower. Vertical wet pit cooling water pumps
are typically mixed flow pumps, whereas dry-pit pumps are typically radial-flow
centrifugal pumps.
A vertical wet pit pump impeller is housed in the pump bowl at the bottom of the
line shaft that transmits the torque required to drive the pump from the motor above.
Although not common in the case of cooling water pumps, there may be more than
one pumping stage with multiple bowls and impellers stacked on top of each other
with diffuser vanes in between to direct the flow from one stage to the next. A suction
bell is located below the first stage bowl to direct the cooling water smoothly into the
impeller. Replaceable bronze wear rings positioned between the impeller and the
bowl minimize wear of both the impeller and the bowl casing and enhance efficiency.
Figures 4.1 and 4.2 illustrate the two kinds of mixed flow impellers: semi-open
(Figure 4.1) and closed (Figure 4.2). Experts do not agree as to which design is more
efficient. Both have their advantages and disadvantages. With a semi-open impeller,
the impeller rides very close to the wear ring below, and this “endplay” or the amount
of vertical clearance between the impeller and the wear ring requires proper
27
28 Engineering of Power Plant and Industrial Cooling Water Systems

FIGURE 4.1 Semi-open impeller.

FIGURE 4.2 Enclosed impeller.

adjustment. Initially, the impeller rests on this seat, and the clearance is adjusted to
allow for the line shaft to stretch while maintaining a close clearance during opera-
tion. The impeller clearance is adjusted by means of an adjusting nut at the top of the
motor. However, if the line shaft and the pump column are of different materials, this
clearance can easily come out of adjustment due to cooling water temperature
changes. By contrast, the wear ring encloses a closed impeller in the vertical direc-
tion which is free to move up and down slightly due to changes in the cooling water
temperature as the line shaft expands or contracts relative to the pump casing.
However, whereas the clearance between a semi-open impeller may be adjusted to
compensate for wear, the same is not true with a closed impeller.
Figure 4.3 shows a cooling water pump with a semi-open impeller on the left and
a CCW pump with a propeller on the right.
Pumps 29

FIGURE 4.3 Vertical wet pit cooling water and CCW pumps. (Courtesy of Johnston
Pump Co.)

4.3 SPECIFIC SPEED
The specific speed of an impeller, n, is defined as the revolutions per minute at which
a geometrically similar impeller would run if it were of such a size as to discharge
one gallon per minute against one foot of head. Specific speed is a parameter used to
select the type of pump that is best suited to a particular application and to check
suction limitations of the pump.2,3 The shape of the pump impeller producing maxi-
mum efficiency is a function of the specific speed as defined by Equation (4.1).1

1
NQ 2
n = 3
(4.1)
H 4

where
N = rototional speed, rev/min
Q = flow rate, gal/min
Η = total head, ft.
30 Engineering of Power Plant and Industrial Cooling Water Systems

FIGURE 4.4 Pump impeller profile as a function of specific speed.

Figure 4.4 shows the various pump impeller profiles as a function of specific speed.
The following shows the type of impeller that would yield the best efficiency for
the indicated range of specific speeds in English units1:

500–3,500 radial flow

3,500–7,500 mixed flow
7,500–12,000 axial flow.

Typical cooling water pumps with flow rates of from 1,000 to 10,000 gal/min
would achieve maximum pump efficiency at a specific speed of approximately
3,000.1,2 A centrifugal pump impeller may be replaced with a smaller or larger impel-
ler in the same bowl size, within limits.4

4.4 CAVITATION AND PUMP SUCTION SPECIFIC SPEED

Cavitation occurs in pumps when the absolute pressure at the pump inlet decreases
below the vapor pressure of the fluid at which time vapor bubbles form at the impel-
ler inlet.3 When the bubbles pass through the impeller and enter the higher pressure
region of the pump, they collapse. When that collapse occurs on the surface of the
impeller, the liquid rapidly moves in to fill the space left by the bubble, impacting a
small area of the impeller with a very high localized pressure surge that can pit and
erode the impeller.3
Suction specific speed, s, is an index number that describes the characteristics of
the suction of a vertical wet pit pump and is defined in Equation (4.2).2

1 1
NQ 2 NQ 2
s  3
 3
(4.2)
 NPSHA  4
 NPSHR  4

where
NPSHA = net positive suction head available, ft.
NPSHR = net positive suction head required, ft.
Pumps 31

The available value of “s” must equal or exceed that required value in order to
prevent cavitation. A high value of “s” at the operating point of maximum efficiency
indicates a good suction design.
For a vertical wet pit pump, NPSHA may be calculated as in Equation (4.3)

NPSHA  Pb  S  Pv (4.3)

where
Pb = barometric pressure
S = submergence
Pv = vapor pressure at the cooling water temperature.
NPSHR is dictated by the pump design and is specified by the pump manufac-
turer. (See Figure 4.5 below.) A reasonable suction specific speed is 8,500 rev/min.2

4.5 SYSTEM HEAD CURVE

The dashed line in Figure 4.5 illustrates the system head curve for a cooling water
system. In this example, the static head of the system is indicated by the Y intercept
at zero flow rate. As flow increases, the required head to force the flow through the
system also increases by a ratio of the flow to the 1.8 to 2.0 power. Since most cool-
ing water systems serve more than one HX, the pressure drop through the various
parallel loops in the system is balanced to the extent possible to achieve the desired
flow to each component by judicious selection of pipe sizes. (See Chapter 7.)
However, the required pump head to achieve the desired total flow rate through the
system is normally based on the limiting flow path, and the flow to other components
is set by throttling control valves as required to achieve the desired flow balance.
(See Chapter 8.)

4.6 PUMP HEAD CAPACITY CURVE

The pump head-capacity curve in Figure 4.5 shows how much flow a particular cool-
ing water pump can deliver as a function of the required system head. The same
pump may meet the flow requirements for a wide variety of system head curves, but
as the flow demand increases, the available head decreases. The point at which the
pump head curve intersects the Y axis (i.e. zero flow) is known as “shut-off head”.
Although a continuously rising characteristic pump curve as it approaches shut-off
head is desirable for controllability, limiting the pump head at shut-off may be desir-
able to limit the system design pressure. At the other extreme, the maximum pump
flow is known as the pump “run-out”, which is limited, since pump cavitation may
occur when the pump operating head is much lower than the rated head and the
NPSHR increases exponentially.
Figure 4.6 shows the pump head-capacity curve for the CCW pump shown in
Figure 3.2. The absence of a continuously rising head curve is not uncommon among
CCW pumps. When this is the case, the point of normal operation should be well to
the right of the hump in the curve for stable operation.
32 Engineering of Power Plant and Industrial Cooling Water Systems
FIGURE 4.5 System and pump curves. (Courtesy Sulzer Pump Co.)
Pumps 33

50

Pump Head (ft) 40

30

20

10

0
0 50000 100000 150000 200000
CCW Flow Rate (gal/min)

FIGURE 4.6 CCW pump head-capacity curve.

4.7 PUMP SIZE, EFFICIENCY, AND SPEED

The impeller of a pump is configured to achieve maximum efficiency at a single point
along the pump head-capacity curve where there is minimum radial load on the
impeller and minimum cavitation.3 As one may see from Figure 4.5, pump selection
is based on the point of normal pump operating being at this point of maximum effi-
ciency. Normally, the efficiencies of centrifugal pumps vary between 50% and 85%
with larger pumps being more efficient than smaller ones as illustrated by Figure 4.7.
The capacity of a centrifugal pump is proportional to its rotational speed, which nor-
mally can vary from between 800 and 3,000 rev/min, depending upon the design and
rated capacity of the pump. Cooling water pumps normally operate at a constant
speed during operation. Pumps that operate at slower speeds are more efficient than
those that operate at higher speeds. The specific speed that is available according to
Equation (4.2) is a valuable criterion in determining the allowable maximum speed
of the pump.2
The brake horsepower, BHP, required to drive a pump in English units is defined
in Equation (4.4).5

Q H s.g.
BHP  (4.4)
3960  p

where
s.g. = specific gravity of the cooling water
ηp = pump efficiency.
34 Engineering of Power Plant and Industrial Cooling Water Systems

240

200
69%
Total Head [ft]

7.5" Dia.
160
8.375" Dia.
12 BHP
14 BHP
120 16 BHP
50% Eff.
55% Eff.
60% Eff.
80 63% Eff.
65% Eff
67% Eff.

40
0 40 80 120 160 200 240 280 320 360 400
U. S. Gallons per Minute

FIGURE 4.7 Pump head and efficiency curves. (Courtesy FLOWSERVE Pump Division).

4.8 NET POSITIVE SUCTION HEAD

As discussed in Section 4.4, in order to prevent cavitation occurring at the pump
impeller, NPSHA must equal or exceed NPSHR. NPSHR is dictated by the
pump design, and the pump manufacturer may be able within limits to engineer a
pump with lower values of NPSHR.
The “Thoma’s cavitation constant”, σ , defined as the ratio of the net positive suc-
tion head at the point of cavitation inception, NPSHi, to H may be calculated from
Equation (4.5).3
4
NPSHi Kn 3
   (4.5)
H 106

where values of K are given in Table 4.1.3

As one may see from Equation (4.5) and Table 4.1, NPSHi increases with H and n
and decreases with increased ηp. When applied to multistage pumps, H is the total
head per stage, so one way to reduce NPSHi would be to design a very efficient pump
with two stages.

TABLE 4.1
Values of K for Equation (4.5)3
Pump Efficiency (%) SI Units English Units
70 1726 9.4
80 1210 6.3
90 796 4.3
Pumps 35

4.9 PUMP MINIMUM FLOW RATE

As one may see from Figure 4.5, even at shut-off head the pump motor brake horse-
power transmitted to pump can be quite high, due to poor pump efficiency at that
point. The pump is cooled by the cooling water flow through the pump. If the flow
through the pump is too low, the pump can overheat. As a general rule, the minimum
flow through a cooling water pump should be no less than 40% of the flow rate at the
point of maximum efficiency. Large high horsepower pumps may have a minimum
flow limitation as high as 70% of the flow rate at the point of maximum efficiency.6
Since in the case of cooling water systems the required flow rate may vary with cool-
ing water temperature, multiple pumps operating in parallel may be required to not
only avoid overheating the pump but also to be able to operate at closer to the point
of maximum efficiency when the temperature of the cooling water is low.

4.10 AIR VENTING
As shown in Figures 3.1 and 3.5, a check valve is normally located at the discharge
of a wet pit vertical turbine-type cooling water pump to prevent the entire cooling
water system from draining down each time a pump is stopped. When the pump
stops, normal practice is to bleed air into the pump column through a float-type air
release valve or through an inverted check valve as shown in the figures. When the
pump is restarted, the air in the pump column is gradually released through this valve
to control the rise of water in the pump column to avoid a water hammer as the col-
umn of water hits the closed check valve. (See Chapter 8.) Care must be taken in
sizing the air release mechanism, since if it vents the air too slowly, the remaining air
will pass through the discharge check valve with the water, and if too quickly, a water
hammer may occur. Although a float-type air release valve or inverted check valve of
suitable size may be available, most likely the correct air flow would be dictated by
an appropriately sized orifice as shown in the figures. The mass flow rate of air may
be determined by the volumetric flow rate of the pump at rated conditions times the
density of air. Equation (4.6) from Page 3–24 of Reference 7 shows the relationship
between the mass flow rate of air as a function of the diameter of the orifice in
English units.

w  0.525 Y d12 C P 1 (4.6)

where
w = mass flow rate of air, lbm/sec
Y = expansion factor (See Page A-21 of Reference 7)
d1 = orifice diameter, inches
C = orifice flow coefficient
ΔP = pressure drop across the orifice, lbf/in2.

The size of the piping and valves up stream of the orifice should be such that the
pressure drop through them should be no more than approximately 10% of the pres-
sure drop through the orifice.
36 Engineering of Power Plant and Industrial Cooling Water Systems

In an open CCW system, the siphon through the condenser is normally main-
tained when the CCW pumps are stopped. Refilling the system and reestablishing
condenser vacuum is a long and laborious process, so the system would not be
allowed to drain down by continuing to operate the condenser vacuum priming
pumps. The condenser vacuum is maintained by continuing to operate the condenser
vacuum pumps. If it becomes necessary to allow the water to drain from the MC, it
would be refilled by throttling the condenser discharge valves to raise the hydraulic
gradient sufficient to fill the condenser tubes as air is vented through the vacuum
priming system to control the rate at which the water rises in the tubes to avoid a
water hammer. (See Chapter 8.)

4.11 PUMP LINE SHAFT SEALS

The line shaft connecting the impeller of a vertical wet pit pump may be either closed
or open. The pumps in Figure 4.3 have a closed line shaft (i.e. the shaft is enclosed
within a pipe). Pumps with long line shafts must have bearings supporting the shaft
spaced so that the natural frequency is above that of the pump’s rotational speed to
prevent vibration. The bearings are either cutlass rubber or phenol plastic. If the shaft
is enclosed, the lubricating water may come from down stream of the strainer shown
in Figure 3.5 or from a separate seal water system. If the shaft is open, the bearing is
lubricated by the cooling water passing by the bearing. Long line shafts require
screwed, split ring, or flanged shaft couplings. Prior to pump start, prelubrication is
required for the line shaft bearings, throttle bushing, and mechanical seal.

4.12 PUMP MOTORS
The required pump motor rated horsepower is defined in Equation (4.7).

BHP
MHP  (4.7)
m

where
MHP = motor horsepower
ηm = motor efficiency.
The motor efficiency is typically 95%.
The cooling water system pump motor should be sized so that the MHP require-
ment is not exceeded at any point over the normal operating range. The service factor
of a motor is an indication of how much the nameplate rating of the motor may be
exceeded for short periods of operation.
The type of motor depends on the function of the cooling water system. Safety-
related cooling water pump motors are normally Nuclear Class 1E, totally enclosed,
with a service factor of 1.0. Non-safety-related cooling water pump motors are
normally National Electric Manufacturers Association MG1, open and drip-proof,
but if the cooling water system is an outdoor installation, the pump motor should be
weatherproof. The service factor for these pumps is normally 1.15. Motor cooling
water lines (if required) should be self-draining to preclude freezing.
Pumps 37

4.13 PUMP MARGIN AND IN-SERVICE TESTING

Under provisions of 10CFR50.55a, nuclear safety-related (essential) cooling water
pumps are required to be inservice tested in accordance with Section XI of the
American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel
Code. ASME Section XI in turn invokes Reference 8. Subsections ISTB and ISTF of
Reference 8 cover inservice testing of pumps in light-water reactor nuclear power
plants before and after the year 2000, respectively. The test quantities are required to
be measured and compared to reference values that are to be measured during initial
operation or after refurbishment of the pump. The values to be measured include
pump speed, differential pressure, flow rate, and vibration amplitude. The pump is to
be operated at the nominal motor speed and the resistance of the system varied until
the measured flow equals the reference point, whereupon the pump head is deter-
mined and compared with the reference value. Alternatively, the flow rate may be
varied until the pump head matches the reference value and the flow rate is compared
to the reference value. Note that the test is solely to determine the performance of the
pump and does not speak to the flow-passing capability of the cooling water system.
(See Chapter 7.)
The test acceptance criteria include an “alert range” and an “action range”. If the
results of the test fall within the former value, the frequency of the testing is to be
doubled until the cause of the problem is determined and the condition corrected. If
the results of the test fall within the latter value, the pump is to be declared inoperable.
Figure 4.8 shows the results of inservice testing of an essential cooling water
pump. This pump was tested at a flow rate of 9,500 gal/min.
This pump had a semi-open impeller, and the line shaft was SS, while the pump
casing was CS. In addition to the gradual declining performance of the pump, one
can clearly see the impact of the seasonal change in the clearance between the
impeller and the wear ring on performance as discussed in Section 4.2.

4.14 POTENTIAL IMPROVEMENTS IN THE ENGINEERING OF

COOLING WATER PUMPS
A search of the INPO Nuclear Plant Reliability Data System over a four-year period
revealed that the majority (72%) of the pump failures were due to external leakage
which did not significantly affect operability. Of the remaining failures, approxi-
mately 40% of the failures were detected by periodic testing. The remainder were
detected through operational abnormalities or while performing normal maintenance.
Cooling water system pumps are particularly vulnerable to wear and failures due to
the corrosive and abrasive nature of the raw water being pumped.
As illustrated by Figure 4.8, the performance of vertical mixed flow cooling water
pumps can deteriorate over time. When the pump column and line shaft are of
dissimilar metals, the seasonal loss on pumping head due to difference in thermal
expansion can be significant if a semi-open impeller is employed. Although no such
problem occurs with totally enclosed impellers, one does not have the ability to
manually adjust the clearance between the impeller and the wearing ring without
disassembling the pump in order to restore the operating clearance.
 38 Engineering of Power Plant and Industrial Cooling Water Systems

ERCW Pump A-A Head at 9500 GPM

Head Req. Action Low Alert Req. Action High Adjusted 0-PI-67-18B 2-PDI-67-61 Head Trend
280.00

275.00

270.00 269.46
268.35
267.37

265.00
264.12 263.90
Head, feet of water

262.15
260.00 259.97
258.35 258.32
257.61 257.88
256.40 255.35
255.80 255.30
255.00 254.67
254.15
252.51 252.99
252.01
252.98

250.00 249.88
248.37
247.76
247.79 247.84 246.63
245.27
245.00 243.22 243.64 244.11
244.34
243.38 243.88 244.60
242.74 243.19242.74
243.09 243.22
242.27 241.16 240.62 240.87
240.00 240.46
240.64
238.12 237.38
237.15 236.70
236.02 235.54
235.00

230.00
07/03/1995 11/14/1996 03/29/1998 08/11/1999 12/23/2000 05/07/2002 09/19/2003 01/31/2005 06/15/2006 10/28/2007
Date of Test

FIGURE 4.8 Inservice test results for a cooling water pump at a flow rate of 9,500 GPM.

The following enhancements to the engineering of cooling water pumps are

proposed:

• The ability to test cooling water pumps is essential to diagnosing pump dete-
rioration and correcting problems. Therefore, provisions should be made in the
design of the cooling water system to be able to periodically test the pump.
These provisions should include means to measure the flow rate and to calcu-
late the total head of the pump and to provide a flow path capable of testing
near the normal flow rate.
• Prelubrication of vertical wet pit mixed flow pumps should be provided from
either down stream of the strainer shown in Figure 3.5 or from a separate seal
water system.
• Small water lines such as those required to provide prelubrication and cooling
water pump motor cooling should be protected by appropriate insulation
and/or heat tracing.
• If a semi-open impeller is employed, the line shaft and the pump column
should be of the same material. If the line shaft is of CS, disassembly can be
very difficult due to corrosion. Therefore, both the line shaft and the pump
column should be of SS.
• Considering the corrosive and erosive nature of RCW, serious consideration
should be given to employing SS materials for the pump bowl, impeller, col-
umn, line shaft, and discharge head. (See Chapter 6.)
Pumps 39

• Specify the appropriate heat treatment for each part to prevent stress corrosion
cracking.
• For pumps in unfavorable sump configurations, consider employing features
such as straightening vanes on the pump suction.

REFERENCES
1. Linsley, R. K. et al. Water-Resources Engineering, 4th ed., pp. 416–423, McGraw-Hill,
Inc., 1992.
2. Hydraulic Institute Standards for Centrifugal, Rotary, and Reciprocating Pumps, 14th
ed., p. 12, Hydraulic Institute, Parsippany, NJ, 1983.
3. Mays, L. W. Water Resource Engineering, 2nd ed., pp. 479–480, John Wiley & Sons,
Inc., 2011.
4. Shashi, M. Liquid Pipeline Hydraulics, CED Engineering.
5. Cameron Hydraulic Data, 16th ed., pp. 1–25, Ingersoll-Rand, Woodcliff Lake, NJ,
1981.
6. Guyer, P. J. Introduction to Pumping Stations for Water Supply Systems, CED
Engineering.
7. Crane Technical Paper No. 410, 24th Printing, Crane Co., King of Prussia, PA, 1988.
8. Operation and Maintenance of Nuclear Plants, ASME OM-2020, American Society of
Mechanical Engineers, New York, NY, 2020.
 5 Piping and Valves

5.1 PIPING APPLICATIONS
The purpose of cooling water piping is to transport water from one point in the sys-
tem to another in an efficient and reliable manner for the life of the plant. The appli-
cations for cooling water piping may be broadly categorized as either buried pipe or
pipe that is hung from pipe hangers inside a structure. A material that may be suitable
for one application may not be suitable for the other.

5.2 PIPING MATERIALS
CS piping alone, though less expensive than some other alternatives, is generally a
poor choice in the long run for either buried or hung applications when compared
with other piping materials. As will be shown in Chapter 6, CS in cooling water sys-
tems is frequently not a suitable material to last for the life of the plant and must be
replaced due to corrosion or flow blockage. As will be shown in Chapter 7, the flow-
passing capability of CS piping in cooling water systems is often poor even in rela-
tively new piping systems when compared with other materials. Indeed, some studies
have shown that due to the tuberculation found in CS piping in cooling water sys-
tems, the line size required to pass the same flow rate with the same pressure drop
may be one size smaller for corrosion-resistant materials than for CS, and the installed
cost may be only a fraction more or less than for CS. CS pipe (typically ASTM A106
Grade B or A53 Grade B) has been used in all sizes up to more than 20 ft (6.1 m) in
diameter. Steel pipe in sizes 0.5 to 12-inch (1.27 to 30.5 cm) in diameter is often a
continuous tube formed by drawing over a mandrel. Larger steel pipe up to 48-inches
(122 cm) is normally fabricated from long, narrow steel plate that is bent to shape and
welded along a spiral joint.1
Suitable alternative materials, depending on where in the plant the piping is
located, may include SS, CPP, pre-lined ductile iron (PDI) pipe, and a variety of
plastic piping materials such as FRP, PVC, and HDPE pipe.

5.3 PIPE LININGS, COATINGS, AND WRAPS

Pipe linings and coatings including CS pipe lined with epoxy, PVC, SS, or cement
mortar are available to the designer. Some linings and/or coatings may be more
appropriate for replacement of existing CS piping, while others are best suited for
new construction.

5.4 BURIED PIPING
A typical power plant has a large complement of cooling water piping buried in and
around the facility. CS piping has been the material of choice for piping systems
41
42 Engineering of Power Plant and Industrial Cooling Water Systems

buried in the plant yard. Such piping is normally coated on the exterior with a bitu-
minous coating and wrapped in fiberglass to protect against exterior corrosion.
Depending on the composition of the soil surrounding the pipe, a cathodic protection
system may be needed to supplement the exterior coating. As shown in Figure 5.1,
cathodic protection consists of a direct current power source with the positive termi-
nal connected to sacrificial anodes buried under ground. Other wires serving as cath-
odes are connected to the pipes. Any breach in the pipe coating provides a path for
the electricity to pass from the anodes through the ground, back to the pipe, and back
through the cable to the power source. The flow of electricity into the pipe limits the
corrosion, thus protecting the pipe from loss of metal. Although frequently viewed as
passive, cathodic protection systems require supervision. Excessive impressed volt-
age can cause disbanding of the pipe coating, resulting in an accelerated corrosion
rate. Plant engineers are often not familiar with these systems, and when equipment
fails (if it is discovered at all) it may be found to be obsolete and spare parts may not
be available.
Other materials such as CPP, PDI pipe, and plastic piping may be more suitable
for buried service. In recent years, there has been a trend toward using various plastic
piping systems.

5.4.1 Concrete Pressure Pipes

The three types of CPP are reinforced, prestressed, and pretensioned concrete pipes.
Reinforced concrete pipe consists of a thin steel cylinder with steel joint rings welded
to its ends. The cylinder is surrounded with one or more cages of reinforcing made
from bar, wire, or welded wire fabric. This assembly is encased in a wall of dense
concrete covering the cylinder both inside and out. The pipes are manufactured
according to American Water Works Association (AWWA) C300. Sizes covered by

FIGURE 5.1 Cathodic protection of buried CS piping.

Piping and Valves 43

the standard are 20 through 90-inch (50.8 through 229 cm). Operating pressures can
be as high as 260 lbf/in2 (1,793 kPa). Standard lengths are 12, 16, and 20 feet (3.66,
4.88, and 6.10 m).2
Prestressed concrete pipe consists of a thin steel cylinder with steel joint rings
welded to its ends. The cylinder is then lined with a dense coating of concrete and
high tensile strength wire is wound around the outside of the cylinder at a
predetermined stress. The pipes are manufactured according to AWWA C301. Sizes
covered by the standard are 16 through 96-inch (40.6 through 244 cm). Operating
pressure range for economical design is from 200 to 275 lbf/in2 (1,379 to 1,896 kPa).
Standard lengths are 16 and 20 feet (4.88 to 6.10 m).2
Pretensioned concrete pipe as shown in Figure 5.2 consists of a heavier steel cyl-
inder with steel joint rings welded to its ends. The cylinder is then lined with a dense
coating of concrete. High tensile strength wire is wound around the outside of the
cylinder at a predetermined stress, and the outside is coated with concrete. The pipes
are manufactured according to AWWA C303. Sizes covered by the standard are 10
through 36-inch (25.4 through 91.4 cm). Operating pressure range can be as high as
400 lbf/in2.(2,758 kPa). Lengths range from 24 to 40 feet (7.32 to 12.2 m).2
There has been considerable operating experience with all three of these types of
pipes, including some in nuclear seismic Category I applications. They are not
subject to either interior or exterior corrosion. However, special consideration should
be given to the joint details due to possible joint leakage.

5.4.2 Pre-Lined Ductile Iron Pipes

Pre-lined ductile iron pipe is produced in a factory by applying a high strength mortar
to the interior of ductile iron pipe using either the centrifugal process or the projec-
tion method. In the former method, the mortar is distributed evenly throughout the

FIGURE 5.2 Pretensioned concrete pipe.

44 Engineering of Power Plant and Industrial Cooling Water Systems

length of a section of pipe by moving a lance while the pipe is spinning. In the latter
method, the mortar is sprayed or slung evenly onto the pipe wall by a rapidly revolv-
ing moving head inserted through the stationary pipe at the centerline. The lining is
then cured by either storing it in a moist environment, processing it through an ele-
vated temperature curing tunnel, or seal-coating it with a solvent-based paint. The
lining produced by either of these methods produces a dense, smooth, surface that
both inhibits interior corrosion and promotes flow. PDI is produced in accordance
with ANSI/AWWA C104/A21.4. The thickness of the lining is to be not less than
1/16-inch (0.159 cm) for 3 to 12-inch (7.63 to 30.5), 3/32 (0.238 cm) for 14 to 24-inch
(35.6 to 61.0), and 1/8-inch (0.318 cm) for 30 to 64-inch (76.2 to 163 cm) pipe.
Normally acceptable operating temperatures are up to the boiling point of water
unless seal coated in which case should not exceed 150°F (65.6°C). Sections of pipe
are joined by bell and spigot with an intermediate rubber gasket placed in a groove
inside the socket at the bell end.3

5.4.3 Plastic Pipes
The two main classes of plastic pipes are thermoplastics and thermosetting.
Thermoplastic pipes soften at elevated temperatures above the useful range and
become readily formable. On cooling, the material regains its original properties.
Production by extrusion is simple, and joining and field modification exploit the
softening and hardening ability. When heated to high enough temperature and pressed
together, pipe ends fuse into a serviceable joint.4
HDPE pipe is a thermoplastic pipe manufactured by the extrusion process.
Operating pressures can range from 50 lbf/in2 (345 kPa) for large pipes to 200 lbf/in2
(1,379 kPa) for small pipes. The operating temperature limit is 140°F (60.0°C). The
pipes are manufactured according to ASTM D2104, D2239, D2447, F405, and
F449.2 Sizes covered by the standards are 0.75 through 200-inch (1.91. through
508 cm). Standard lengths range from 20 to 40 feet (6.1 to 12.2 m) with coils 1000
feet (305 m) in length.2
Among the thermoplastics, PVC pipe is the most common used for small plastic
piping at power plants. It is also manufactured by the extrusion process. Operating
pressures can range from 150 lbf/in2 to 200 lbf/in2 (1,034 to 1,379 kPa). The operating
temperature limit is 140°F (60.0°C), but the strength is defined at 23°C (73.4°F) and
decreases at higher temperatures. The pipes are manufactured according to ASTM
D1785, D2241, D2321, D2412, D2672, and D3034.2 (See also AWWA C900.) Sizes
covered by the standards are 4 through 18-inch (10.2 through 45.7 cm).2 Standard
lengths range from 20 to 40 feet (6.1to 12.2 m) with coils 1000 feet (305 m) in
length.2
Thermosetting resins cannot be reversibly cycled over a wide temperature range.4
Thermosetting plastics require reinforcement, usually by fiberglass which forms a
composite wall in which each layer contributes to the desired overall qualities of the
pipe.4 Thermosetting resin pipe reinforced with fiberglass reinforcement is manufac-
tured by wrapping reinforcement fibers unbonded around the pipe surface and the
pipe is then heated until the thermal bonding occurs.4 Operating pressures can range
from 30 lbf/in2 (207 kPa) for large pipes to 250 lbf/in2 (1,724 kPa) for small pipes.
Piping and Valves 45

The maximum operating temperature limit is 125°F (51.7°C).2 The pipes are manu-
factured according to ASTM D2996 and ASTM 2997.2 Sizes covered by the stan-
dards are 14 through 168-inch (35.6 through 427 cm). Standard lengths range from
40 to 60 feet (12.2 to 18.3 m).2
FRP pipe is a composite material made from a thermosetting polymer matrix
reinforced with fibers.4 Operating pressures can range from 30 lbf/in2 (207 kPa) for
large pipes to 250 lbf/in2 (1,724 kPa) for small pipes.2 The operating temperature
limit is 140°F (60.0°C).2 The pipes are manufactured according to ASTM D3517 and
ASTM 3754.2 Sizes covered by the standards are nominal 8 through 144-inch (20.3
through 366 cm).2 Standard lengths range from 10 to 20 feet (3.05 to 6.10 m).2 In
some installations, the cost of the fittings to join the pipe runs may equal or exceed
the cost of the pipe. Older installations of FRP pipe have experienced failures due to
a lack of uniform soil compaction, manufacturing defects, and the use of unsuitable
resins.
Considerable operating experience exists with plastic pipe in a variety of applica-
tions in industry. Plastic pipe is not generally subject to corrosion, but if the cooling
water contains particles such as sand, erosion may occur mainly at turns. Plastic
pipes are light and easy to handle, but extreme care must be taken in their installation
in buried applications, as excess soil loading and/or ovalling of the pipe can be a
problem.5

5.4.4 Stainless Steel Pipes

Alternative pipe materials are normally less expensive than SS pipes, but may be
used in buried applications. Although Type 316L has good resistance to chlorides,
overall consideration of soil variability, halide leaching, etc. suggests that an exterior
coating such as a heavy coat of halide-free paint or tape followed by a bituminous
coating is advised.

5.4.5 In Situ Lined Pipes

There are several types of in situ pipe lining processes. One is lining clean pipe with
a thin coat of epoxy. A second is lining the pipe with a thick in situ tube saturated
with epoxy that serves as a pipe within a pipe. A third is cement-mortar lining (CML).
All three of these lining processes require access points in the buried piping some
distance apart.
An epoxy coating alone lends no additional structural strength to the pipe. In
order for epoxy coating to be successful, the pipe must be sand blasted to bright
metal before applying the epoxy, a challenging task. Any failure to achieve bright
metal or any pipe imperfection such as those brought about by rolling and finishing
the pipe can result in imperfections in the lining process. Such imperfections may
result in preferential galvanic attack at that point in the pipe and/or failure of the
lining. Such coating failures frequently result in the epoxy coming off in sheets and
blocking downstream valves and/or HX tubes.
In the second lining method shown in Figure 5.3, an engineered felt in situ tube or
sock of the same diameter as the pipe and slightly longer than the distance between
46 Engineering of Power Plant and Industrial Cooling Water Systems

FIGURE 5.3 Resin-impregnated pipe lining. (Courtesy of Insituform).

access points is first impregnated with a liquid thermosetting epoxy resin and then
folded. The open end of the in situ tube is attached to an inversion tube extending
down from the surface to one end of the pipe. As the tube is filled with water, the in
situ tube is inflated by the water, causing it to be turned inside out and pressing the
resin-impregnated side of the in situ tube firmly against the inside wall of the pipe.
The water inside the in situ tube is then circulated through a boiler, causing the resin
to cure over a period of a few hours and changing the in situ tube into a hard, structur-
ally sound pipe-within-a-pipe. After the water is pumped out and the inversion tube
is removed, the ends of the in situ tube are cut off to match the ends of the pipe. This
method of lining may be applied to pipe diameters from 6 to 60-inches (15.2 to
152 cm). The thickness of the lining is a function of the diameter of the pipe and
depth of burial ranging from 0.1 to 1.45 inches (0.254 to 3.68 cm). No structural
strength is required of the host pipe, since the resulting epoxy coated pipe has con-
siderable strength, which is a function of the diameter to in situ tube thickness ratio.
Piping and Valves 47

FIGURE 5.4 Cement-Mortar lining process. (Courtesy of Ameron Corp.)

The third lining method shown in Figure 5.4 is in situ CML.

A portion of the buried pipe is excavated and removed at intervals of several hun-
dred feet. The spool piece thus removed is set aside and cleaned and lined by hand.
The remaining buried pipe is scraped clean of corrosion products and other foreign
material by pulling a “pig” through the pipe. Next, a lining machine is pulled through
the pipe, applying a 0.25 to 0.5-inch (0.64 to 1.28 cm) layer of high strength mortar
to the buried pipe, which is trawled smooth behind the lining machine in the process.
The section of pipe is then sealed and cured in a high humidity environment for sev-
eral days before the spool piece is welded back in place. The portion of the new lin-
ing damaged by the welding process is removed and the heat-effected zone is lined
by hand. Obviously, if access to the interior of the pipe is required to repair the por-
tion of the lining damaged by the weld, this process is principally employed in piping
24-inch (61.0 cm) and larger.
As discussed in Chapter 7, flow tests conducted at TVA’s Sequoyah and Watts Bar
nuclear plants in 1982 indicated that the large essential raw cooling water (ERCW)
system headers between the IPS and the nuclear plants were exhibiting Hazen-
Williams ‘C’ factors of from 75 to 85 after only approximately five years of service.
These values were substantially below the C = 100 which was assumed in the origi-
nal design. As Bowman and Guthrie6 have shown, MIC, largely a pitting corrosion
mechanism, can occur at a rate that is three times the general corrosion rate in cool-
ing water piping, and the Tennessee Valley Authority (TVA) nuclear plants were
observing MIC corrosion rates as high as 16 mils per year (MPY). This corrosion rate
would imply a useful life of less than 23 years for large headers that are typically
0.375-inch (.9325 cm) thick. As part of the corrective action to address these
deficiencies, TVA investigated the use of CML applied in situ, and this solution was
eventually implemented at TVA’s Watts Bar Nuclear Plant (WBNP).7
TVA was the first utility to apply a CML to a safety-related cooling water piping
in situ. The project was undertaken only after exhaustive investigation. Factors
considered in this investigation included operating experience with CML, the
technical feasibility of the undertaking, the ability to qualify the lining, the ability to
meet all codes and standards applicable to the ERCW system, the impact on the
existing design, potential nuclear safety issues, the licensability of the option, the
installed cost, and the impact on the plant schedule.
Miller8 reported that CML has been employed successfully in municipal water
systems for over 100 years. Field experience of the lining showed that CML was still
48 Engineering of Power Plant and Industrial Cooling Water Systems

effectively serving the purpose for which it was intended for many years. TVA sur-
veyed the use of CML not only in municipal applications but also in fossil and
nuclear power plants and discovered that CML had been successfully used in non-
safety-related applications at numerous fossil and nuclear plants7. In general, the
experience with in situ CML has been excellent when applied in accordance with the
applicable AWWA standard.9 However, experience with CML in salt water applica-
tions has not been encouraging. In situ CML should not be confused with pre-lined
pipe which is welded or bolted together in the field. The impact of applying CML on
the existing piping design is minimal, since the piping fittings and other access open-
ings are reinstalled as originally designed.
The technical feasibility of CML is a function of the size or piping to be lined and
the complexity of the system. Although pipe sizes as small as 4 inches (10.2 cm) have
been lined, TVA decided that for safety-related applications, only those lines that can
be accessed from inside the pipe should be candidates for CML. Since each valve and
fitting must be removed from the system prior to lining, CML lends itself to long runs
of large, buried piping.
The WBNP ERCW system is divided into two trains (A and B) with a WBNP
Unit 1 train A and train B and a WBNP Unit 2 train A and train B supply header and
a train A and train B discharge header serving both units. The portion of the WBNP
ERCW system lined with CML consists of the four supply headers extending from
the IPS to the plant auxiliary building and the two discharge headers from the auxil-
iary building to the cooling towers. The supply headers range in length from 5,600 to
5,900 feet (1,710 to 1,800 m) and in size from 24 to 36 inches (61.0 to 91.4 cm). The
two discharge headers are 1,200 and 1,700 feet (366 to 518 m) long. The first 153 feet
(46.6 m) of the supply header is 24-inch (61.0 cm). The balance is 30 and 36-inch
(76.2 and 91.4 cm). Discharge headers are 36-inch (91.4 cm) in diameter. The project
was divided into two halves with each half consisting of one of the supply headers
serving each unit and one of the discharge headers in the same train being lined at a
time so that a train of ERCW would be available to support construction at all times.
An interval between lining the two halves was provided to complete the curing of the
last section, closure, flushing, testing the first half, and unwatering, and opening the
first section of the second half.
The CML process is a highly specialized construction technique that should be
attempted only by those who are experienced in the practice. TVA allowed the CML
contractor the freedom to use the same methods which had proven so successful in
other applications. Every effort was made to minimize the impact on those techniques
due to the fact that the project was safety-related.
The locations of access openings were selected to excavate valves and fittings and
at high and low points in the piping and to ensure that no run of piping is longer than
a few hundred feet. Where possible, locations were selected to minimize the depth of
access holes. The excavations extended at least a foot below the pipe invert and
exposed at least 12 feet (3.66 m) of pipe.
After the piping system had been unwatered, the valves, fittings, or spool pieces
(which are referred to as closure pieces) were removed by cutting the pipe to permit
access to the piping for cleaning and lining equipment. Spool pieces removed from
straight runs of pipe were at least 8 feet (2.44 m) long. Each closure piece was
Piping and Valves 49

removed from the hole and set on the ground beside the hole. Where fittings were
removed, at least 2 feet (0.61 m) of pipe on the end or each run was exposed. After
the piping was cleaned, lined, cured, and inspected, TVA welded the closure piece
back into place. After the closure piece was welded into place and before the CML
was applied by hand to the interior of the weld joint, a vacuum box test was performed
on the weld.
After the access openings were created, the interior surface of the piping was
cleaned by dragging spring-loaded mechanical scrapers through the pipe. A mortar
mix consisting of Portland cement, Type II sand, and water was centrifugally applied
by machine to each run of pipe so as to achieve a densely packed and uniform lining.
The mortar mix was approximately one-to-one by volume, and the water/cement
ratio was kept as low as possible. The mortar application was followed in the same
pass by either a conical-shaped drag trawl or by rotating paddle trawls. Rotating
paddle trawls were found to be superior to the conical-shaped drag trawl. With a drag
trawl, the smallest piece of foreign material could be dragged through the pipe,
requiring repair by hand. As the trawling machine moved along at a rate of
approximately 4.5 ft/min (0.0229 m/s), the mortar was pumped to the machine
through a hose in front of the machine. The travel speed, hose pressure, and trawl
adjustments were set to achieve a 3/8-inch (0.952 cm) thick lining with a tolerance of
+1/8-inch (0.318 cm) and -1/16-inch (0.159 cm). In areas where machine lining was
not practical such as in fittings or closure weld joints, the lining was applied by hand.
Curing was begun immediately upon completion of a lining run. The goal of curing
was to maintain at least a 90% relative humidity inside the pipe. After the lining had
been inspected and repaired as required, the closure piece was installed by welding.
The heat-affected mortar was removed by chipping, and the weld joint was lined by
hand from within the pipe. A curing compound was used only on this weld joint
repair. Since CML does not rely upon its adhesion to the pipe but rather to the arching
action to keep the CML in compression, its use in noncircular transitions is question-
able without special design provisions.
TVA was responsible for opening and closing the holes and piping and keeping
holes unwatered except when the CML contractor was in the hole performing
cleaning lining activities. When the holes were backfilled, the same procedures were
used as in the original construction. This included controlling the quality of backfill
material and using vibrating compactors. Care was taken to ensure that the compactors
did not ride against the pipe. The backfill activity was delayed until the lined pipe and
closure piece had been cured for at least four days.
At points in the piping where the CML terminated, TVA installed termination
rings consisting of 3/8-inch (0.9525 cm) thick by one-inch (2.54 cm) wide rolled bar
stock that were fillet welded to the interior of the pipe before the CML was applied.
In this way, the end of the CML butted against termination rings rather than being
exposed to hydraulic forces. All ASME code activities were performed by TVA,
since the CML contractor was not an “N” stamp holder. (An “N” stamp holder is
authorized to perform work under ASME Section III at a nuclear facility.)
Proper curing is the most important step in the lining process. Curing was the
responsibility of the CML contractor until a section of pipe had been inspected and
accepted by TVA. Pipe end caps were used to keep the CML from drying too quickly.
50 Engineering of Power Plant and Industrial Cooling Water Systems

After the lining achieved initial set, sandbags were installed at the ends of the pipe
and water was added until it flowed steadily through the bottom or the pipe from one
end to the other. The end caps were secured on the ends of the pipe for a minimum of
four days with water being added daily in a similar manner. Any of the exterior
surfaces exposed to the sun were covered with burlap that was wetted hourly. The
closure pieces were capped, wrapped in plastic, and covered with wet burlap.
A total of 612, 13,083, and 12,238 feet (186.5, 3,988, and 3,730 meters) of 24, 30,
and 36-inch (61.0, 76.2, and 91.4 cm) piping were lined. A total of 31 access holes
were excavated to permit removal of 72 closure pieces. Approximately 200 cubic
yards (153 m3) of corrosion products were removed from the interior of the piping.
The project contractor cost in 1982 was $467,000. The cost of TVA support was
estimated to be approximately $1,000,000 which included the cost of excavating and
backfilling the access holes, cutting and welding the piping, curing the pipe, and
performing inspections.

5.5 HUNG PIPING
Unlike buried piping which is bedded and supported by the earth, hung piping is
simply supported by pipe hangers, and more rigid pipe are required to minimize the
number of hangers. In hung piping, dynamic and thermal loading are concentrated at
the hangers. If oversized CS piping is employed to compensate for the reduced flow-
passing capability, the result can be lower velocities, resulting in increased sedimen-
tation and corrosion. PDI pipe is generally not suitable for service in hung piping due
to a lack of flexibility of the cement lining in these piping systems and the increased
weight, especially at the joints where flanges are often employed. Similarly, CPP is
not normally employed in hung piping for much the same reasons. Neither PDI nor
CPP can be readily modified in the field, potentially resulting in schedule impacts
due to field changes.
Dipped and heat-fused plasticized PVC coated pipe has successfully controlled
corrosion in CS piping and fittings. Success depends on the proper procedures to
ensure a good application of the coating. When properly applied, PVC coating has
performed well due to its excellent adhesion characteristics. This tremendous
adhesion strength results in a coating providing bond strengths that may exceed the
tensile strength of the PVC itself. The coating does not delaminate or come off in
large pieces as has sometimes been the case with in situ epoxy coatings. Before the
coating is applied, the base metal is sand blasted to a near white metal finish and the
coating is seamlessly baked on to the pipe making it impervious to attack by saltwater
or chlorine. PVC coated pipe has had to be replaced in some areas subject to high
cavitation. Because PVC is a thermoplastic, it will distort under high heat and
pressure. If the lining is too thick on the flange face, it will possibly extrude into the
inside diameter of the pipe if excessive torques are used during the assembly process.
For this reason, the thickness on the flange face is kept as thin as possible. The PVC
material has a tendency to stick to itself, making disassembly of spool pieces
difficult.10
The use of CS piping clad with a corrosion-resistant alloy such as 300 series SS
or Inconel 625 might be a suitable choice for replacing CS hung pipe.11 Since the
Piping and Valves 51

allowable stresses and thermal expansion would be essentially the same as the CS
pipe, the material may be replaced like-for-like without requiring redesign and/or
reanalysis.11 For cooling water from rivers or lakes with hard water and low chloride
levels, the 300 series SS would be adequate, while those from brackish or salt water
may require a more corrosion-resistant material such as Inconel 625.7 The cladded
pipe is manufactured by a machine-gas arc welding and roll bonding process that
produces a metallurgically bonded alloy 80 to 100 mm thick internal cladding.7 The
CS piping constitutes the pressure boundary, while the cladding constitutes the cor-
rosion allowance. The root pass of field girth welds is made using the same alloy rod
as the cladding, so the surface exposed to the cooling water after welding is the same
as the cladding.7
Plasticized PVC coated pipe and alloy cladded pipe are normally flanged or
screwed fittings, since welding would destroy the coating or cladding Flanged joints
require more surrounding space for fitting up and tightening the bolts than do welded
joints.5
In light of its resistance to corrosion, SS is the preferred material for new hung
piping. Type 304 and 304L SS provide good service in many instances. In the
presence of sedimentation, algae, and oxidizing biocides such as chlorine, Type 316L
SS may be the preferred choice to deal with crevice corrosion and potential localized
high concentrations of chloride ions leading to pitting. Brackish or salt water
applications may require a more corrosion-resistant material such as Inconel 625.
Recently, plastic piping systems have been used inside power plants, but in
addition to requiring more hangers, they present a fire hazard. Figure 5.5 shows
HDPE supported by pipe hangers inside the Catawba Nuclear Station.

FIGURE 5.5 Cooling water HDPE supported by pipe hangers inside the Catawba Nuclear
Station.
52 Engineering of Power Plant and Industrial Cooling Water Systems

5.6 NUCLEAR SAFETY CONSIDERATIONS

5.6.1 Seismic Qualification
From a design standpoint, just about any buried pipe (steel, concrete, plastic, or
lined) can be seismically qualified as long as the design stresses in the piping and
joints are below the allowable stresses. Since seismic loads are mainly due to seismi-
cally imposed deformation, the more flexible pipes are easier to qualify, as flexibility
reduces the loads in the pipe. Operating experience and tests on lined pipes indicate
that most liners will adhere to the pipe and maintain their integrity while the parent
pipe is deforming.
Although prior to the TVA project no effort had ever been made to seismically
qualify CML, impressive evidence exists as to the ability of CML to withstand earth-
quake as long as there is no plastic deformation of the pipe base metal. During the
1971 San Fernando earthquakes, a 96-inch (244 cm) above-ground water line owned
by the Los Angeles Department of Water and Power that was located within three
miles of the quake epicenter suffered both vertical and horizontal displacement due
to surface acceleration. CML was undamaged except where the pipe accordioned,
where spalling did exist. A 20-foot (6.1 m) diameter steel tunnel liner owned by
Metro Water District of Southern California and buried 50 feet (15.2 m) below
ground was not damaged. The pipeline owned by the Metro Water District of
Southern California, which serves the Jenson treatment plant separated by about
3 inches (7.6 cm) in two places and then was driven back together with one section
inside the other, although the water treatment plant was largely destroyed by the
earthquake. The CML was damaged only where the pipe separated.12
The evidence suggested that unless there is plastic deformation in the pipe, the
CML will not be destroyed. Since safety-related cooling water must meet the
requirements of ASME Section Class 3, the piping base metal must meet all Code
requirements and no credit may be allowed for the CML. The CML is considered to
be a coating.
To demonstrate that CML will withstand design basis seismic events, TVA
implemented a full-scale testing program consisting of laboratory field tests and
vibration measurements.12 A total of 100 feet (30.5 m) of 30-inch (76.2 cm) diameter
pipe, 20 feet (6.1 m) of 18-inch (45.7 cm) diameter pipe, and 90-degree elbow were
lined in the field. The lining materials and procedures were the same as those to be
used in the CML projects. Cement-mortar specimens were tested for compressive,
tensile, and flexural strength, modulus of elasticity, and density. A 40-foot (12.2 m)
length of 30-inch (76.2 cm) CML was installed in a trench and as it was being
backfilled, it was subjected to a dynamic loading of 36,000 lbf (160.1 kN) at 28 hertz
from a vibratory roller as shown in Figure 5.6.
The piping was then transported to the Singleton Materials & Engineering
Laboratory (now known as SM&E). As the piping was transported the 100 miles
(161 km) from the construction site where the lining was done to SM&E where the
tests were conducted, two accelerometers were mounted on two of the 30-inch
(76.2 cm) pipes to monitor vibration experienced by the pipes during the trip.
Maximum acceleration experienced by the bottom pipe was 0.6g and that experi-
enced by top pipe was 2.lg. Both values were higher than the 0.18g safe-shutdown
Piping and Valves 53

FIGURE 5.6 Cement-Mortar lining being subjected to dynamic loading.

earthquake acceleration used for the design or TVA nuclear plants. The recorded
maximum peak-to-peak accelerations were 1.2g and 3.8g, respectively. Dominant
frequencies ranged from 15 to 70 Hz. For most large earthquakes, the dominant fre-
quencies are in the range of 0.5 to 10 Hz. Lower frequencies indicate a buried pipe
would experience fewer cycles of vibration during a real earthquake. No cracks due
to vibration were found in any of the CML after unloading, and it was concluded that
the lining had experienced more severe vibrations than from any potential earth-
quakes in terms of magnitude and number of cycles.
After the piping arrived at the laboratory, a 30-ft (9.14 m) section of 30 inch
(76.2 cm) pipe was subjected to bending, cyclic loading, and drop tests as seen in
Figures 5.7 and 5.8.

FIGURE 5.7 Cement-Mortar lining being subjected to cyclic loading and bending test.
54 Engineering of Power Plant and Industrial Cooling Water Systems

FIGURE 5.8 Cement-Mortar lining drop test.

FIGURE 5.9 Cement-Mortar lining elbow bending test.

Cracks appeared in the lining only after the applied stress exceeded the tensile
strength of the mortar. No lining failure occurred.
2-foot (0.61 m) long sections of 30-inch (76.2 cm) pipe were subjected to cyclic
loading, torsion, drop, and impact tests as seen in Figure 5.8.
Cracks appear in the pipe at 15 foot (4.57 m) drop height. A 5-foot (1.5 meter)
length of 30-inch (76.2 cm) pipe was welded to each end of the 30-inch (76.2 cm)
elbow and it was subjected to bending tests shown in Figure 5.9
The elbow was loaded to 50,000 lb (22,700 kg) without failure. When the elbow
was loaded to 54,000 lb (24,500 kg), the lining failed, and the elbow was shortened
by 5 inches, 1.5 inches permanently.
Piping and Valves 55

FIGURE 5.10 Cement-Mortar lining three-edge bearing test.

A three-edge-bearing test shown in Figure 5.10 demonstrated that the CML is

flexible.
The CML failed only after the formation of plastic hinges in the steel. The lining
underwent considerable cracking prior to separation and failure.
From these tests, it is concluded that the test loadings applied to the CML were
much more severe and broad-ranged than the design seismic loadings. Therefore, the
CML lining in buried SWS piping is seismically qualified when applied in accor-
dance with approved QA procedures.13

5.6.2 Codes and Standards

Reference 14 establishes piping in safety-related cooling water systems in nuclear
plants as “Quality Group C” and states that the requirements of this group are ful-
filled if the system is designed to ASME Section III, Class 3 requirements.14
Prestressed and pretensioned concrete pipes are not covered by ASME Section III but
equivalency may be established under provisions of Reference 14. ASME Code Case
N-155-2 addresses FRP. ASME Code Case No. N-755 addresses HDPE pipe. In
2008, the NRC approved the installation of HDPE to replace buried safety-related CS
cooling water piping. HDPE was subsequently installed in the safety-related cooling
water systems at the Catawba and Callaway nuclear plants.
Although the welding activities of the WBNP CML project were done to ASME
Section III, Class 3 requirements, the Article ND-6000 requirements presented a
challenge, since it was impractical to perform hydrostatic testing on each weld before
the CML was applied to the interior of the pipe weld joint. Neither Section III nor
Section XI specifically address the hydrostatic testing of lined pipe. The solution was
to close out the Section III activities on the basis of the hydrostatic test that was per-
formed prior to starting the CML activities and to declare the CML to be a “replace-
ment” under Section Xl and perform the test under Section Xl after the CML has
been applied. At least two code interpretations seemed to TVA to set a precedence
permitting pressure tests in this manner, and concurrence was received from the
Authorized Nuclear Inspector on this approach.
56 Engineering of Power Plant and Industrial Cooling Water Systems

5.6.3 NRC Approval of CML

On April 30, 1982, TVA presented the WBNP CML project including the seismic
qualification test results to the NRCs Advisory Committee on Reactor Safeguards
(ACRS). The ACRS’ primary concern was with the long-term durability or the CML
with the potential for the failed lining to plug the tubes of HXs in the ERCW system.
The ACRS recommended to the NRC staff that TVA develop in-service inspection
requirements for the CML. Accordingly, the NRC requested additional justification
for the modification. TVA responded with a report that addressed each of the follow-
ing potential failure mechanisms: leaching, erosion, and spalling.
The TVA report pointed out that 9,200 lb/in2 (63,400 kPa) cement mortar is
extremely dense and relatively impermeable to mildly aggressive water. Tennessee
River water has an average pH of 7.3, a chloride ion content of less than 10 ppm, and
a sulphate ion content of less than 20 ppm. Other materials that might attack the
mortar are in negligible quantities. TVA noted that there will possibly be some sur-
face leaching, but it was pointed out that Miller8 reported 1/16-inch (0.159 cm) thick
lining was almost leached away after 29 years in water with pH of 6.6 and a total
hardness of 12 ppm. TVA concluded that the CML will not lose more than 1/16 inch
(0.159 cm) over the life of the plant.
Regarding the potential for the lining to fail due to erosion, TVA pointed out that
the maximum water velocity in the headers is approximately 12 ft/sec (3.66 m/sec)
and the suspended solids are less than 20 ppm. Therefore, neither cavitation nor
abrasion is expected to occur. Factors which could possibly cause spalling are
excessive shrinkage, corrosion behind the liner, or gross distortion of the pipe. It is
noted that after the piping system is placed in service, the mortar will expand rather
than shrink. The pH of the mortar will prevent corrosion behind the liner, and gross
distortion of the piping will not occur since the piping is buried.
After a second meeting with the ACRS on August 13, 1982, the NRC issued a
favorable Safety Evaluation Report (SER)15 on the CML modification. The SER
acknowledges the good experience with CML in city water distribution systems and
in power plants. The SER states that the tests performed by TVA demonstrate the
adherence and flexibility of the freshly installed CML, and the SER concludes that
significant short-term failure of the lining is unlikely. The SER noted that as CML
ages it normally gains in strength and that rates of leaching depend upon the hardness
and acidity of the water. The NRC staff reviewed the program proposed by TVA to
periodically inspect the condition and calcium content of the CML samples and
found it to be acceptable. The SER states that the NRC concludes that it is unlikely
that CML will be significantly weakened by leaching during the plant life and that
the NRC staff concludes that CML in the ERCW piping was acceptable.

5.7 VALVES AND FITTINGS

Valve bodies and fittings are generally made of forged or cast steel, cast iron, or duc-
tile iron. The connection to piping may be flanged, screwed, or welded. For pipe sizes
under 3 inches (7.6 cm), forged steel socket welding fittings are often employed.
Elbows may be fabricated from piping by bending the pipe. Complex sections of
Piping and Valves 57

piping such as headers are normally fabricated in the shop using saddles to reinforce
fabricated tees where required.

5.7.1 Valves
A valve is a device used to prevent, limit, or permit the flow of cooling water through
a portion of the system. Swing check valves as shown in Figure 5.11 prevent the
reverse flow of the cooling water and are most often found at the cooling water pump
discharge to prevent back flow through an adjacent idle pump. A disk is hinged at the
top of the valve, and flow through the valve keeps the disk open, while reverse flow
or gravity causes the disk to swing shut, shutting off flow.
Butterfly valves are the most common valves in a cooling water system because
they are simple, relatively inexpensive, weigh less, and are easier to operate and
maintain than many other types of valves. Butterfly valves use a circular disk to
rotate through 90° to go from full closed to full open and may be stopped at any point
in between to throttle the flow through the line. There is very little pressure drop
when the butterfly valve is wide open. Most butterfly valves in cooling water system
service have elastomer lining seats. They are controlled with either a hand wheel or
are operator powered using either an electric motor or pneumatic actuator. The com-
mon body types are lug style in which the lugs are threaded on each side to receive
bolts from adjacent flanges, wafer style where the body is clamped between two
pipeline flanges, and flanged ends.
Gate valves are used for applications requiring either fully open operation (requir-
ing minimum pressure drop) or fully closed operation (requiring a tight shut-off) and
should not be used for throttling. When operating, the disk moves up or down on the
end of a threaded valve stem, driven by an electric motor or pneumatic actuator. Gate
valves are either parallel type in which a disk slides between two parallel seats or
wedge-shaped, which has two inclined seats. When fully open, the disk is fully with-
drawn, making it possible for a pipe cleaning pig to pass unobstructed. The gate

FIGURE 5.11 Swing check valve. (Courtesy Crane Co.)

58 Engineering of Power Plant and Industrial Cooling Water Systems

FIGURE 5.12 Globe valve. (Courtesy Petteri Aimonen).

valve has either a metal or rubber seat in a depression in the valve bottom. Since for
cooling water systems the potential exists for objects such as pebbles or mollusks to
hinder complete closure of the valve, a rubber seat may be required if tight shut-off
is required. Gate valves are more expensive than butterfly valves and require more
physical space.
Globe valves as shown in Figure 5.12 are used to regulate flow by passing the flow
through a spherical body in which the flow must pass through an internal baffled
opening, the size of which is adjusted by a movable plug that can be screwed up or
down. Typically, automated globe valves (control valves) use a smooth rather than
threaded stem which is opened and closed by an actuator assembly that may be air-
operated. Globe valves exhibit a relatively high pressure drop.
Ball valves are sometimes used in smaller lines where tight shut-off is impor-
tant. A ball with a hole in it controls the flow. When the hole is aligned with the
direction of flow, the pressure drop through the valve is insignificant, and when the
hole is positioned perpendicular to the direction of flow, the flow is shut off. By
positioning the ball in any intermediate position, the cooling water flow rate may
be controlled by changing the pressure drop through the valve. Ball valves are nor-
mally operated manually by a handle attached to a stem that controls the position
of the ball.

5.7.2 Fittings
Figure 5.13 shows a sampling of the cooling water system pipe fittings that are com-
mercially available. Long radius elbows are generally preferred because they require
less pressure drop than short radius; however, they require more space.

5.8 PIPING DESIGN WALL THICKNESS

The required wall thickness of commercial grade piping under the ASME B 31.1
Power Piping code, is shown in Equation (5.1).16
Piping and Valves 59

FIGURE 5.13 Pipe fittings. (Courtesy Crane Co.)

P Do
tm   A (5.1)
2  SE  Py 

where
tm = minimum required wall, in
P = internal pipe pressure, lbf/in2
Do = outside diameter, in
S = maximum allowable stress, lbf/in2
E = joint efficiency (1.0 for seamless pipe)
y = temperature coefficient (0.4 for cooling water piping)
A = corrosion allowance, in (See Chapter 6).

Additional pipe wall thickness may be required to withstand external loads such
as gravity, thermal loads, torsion, and seismic effects, etc.
The formulae for ASME Section III, Class 3 piping are the same as Equation (5.1).

REFERENCES
1. Linsley, R. K. et al. Water-Resources Engineering, 4th ed., p. 378, McGraw-Hill, Inc.,
New York, NY, 1992.
2. Chandley, C. A. et al. Essential Raw Cooling Water (ERCW) System Piping Materials,
memorandum MEB ‘810413022, Tennessee Valley Authority, Knoxville, TN, 1981.
60 Engineering of Power Plant and Industrial Cooling Water Systems

3. Bonds, R. W. Corrosion Control – Cement-Mortar Lining for Ductile Iron Pipe, Ductile
Iron Pipe Research Association, Birmingham, AL, 2017.
4. O’Keefe, William. Corrosion-resistant Piping for Utility and Industrial Power Plants.
Power, Vol. 125, No. 4, pp. S1–S24, April 1981.
5. Chandley, C. A. et al. Nonsafety-Related Raw Water Systems Piping Materials, memo-
randum MEB ‘820107011, Tennessee Valley Authority, Knoxville, TN, 1982.
6. Bowman, C. F. and P.V. Guthrie, Jr. Corrosion in Carbon Steel Service Water Piping.
1994 ASME Pressure Vessel and Piping Conference, Phoenix, AZ, ASME, 1994.
7. Bowman, C. F. In Situ Cement-Mortar Lining of Safety-Related Service Water Piping
Systems. International Joint Power Conference, 94-JPGC-NE-6, 1994.
8. Miller, W. T. Durability of Cement-Mortar Linings in Cast-Iron Pipe. Journal AWWA.
American Water Works Association, Denver, CO, June 1965.
9. AWWA Standard C602-76. Cement –Mortar Lining of Water Piplines – Inch and
Larger – In Place, American Water Works Association, Denver, CO, 1976.
10. Robinson, L. K. and P. C. Fritchman. Lining of Raw Water Piping with Plasticized PVC,
EPRI Service Water Systems Reliability Improvement Seminar, Electric Power Research
Institute, New York, NY, 1994.
11. Chakravarti, Bhaven. Use of Clad Piping Products for Solving Nuclear Plant Service
Water Systems Corrosion, 91-JPGC-NE-12. International Power Generation
Conference, ASME, 1991.
12. Hubble, J. D. Operating Experience for Alternative Raw Water Piping Materials,
Memorandum MEB 810720032, Tennessee Valley AuthorityKnoxville, TN, 1981.
13. Sun, C. N., et al. Full Scale Testing and Qualification of Cement-Mortar Lined Carbon
Steel Pipe, CEB-82-8, Tennessee Valley Authority - Division of Engineering Design –
Civil Engineering Branch, Knoxville, TN, 1982.
14. Regulatory Guide 1.26: Quality Group Classifications and Standards for Water, Steam,
and Radioactive-Waste-Containing Components of Nuclear Power Plants, Rev. 4,
Nuclear Regulatory Commission, Washington, DC, 2007.
15. Watts Bar Nuclear Plant SSER 4. Report of the Advisory Committee on Reactor
Safeguards, Section 19, pp. 19–1 to 19–2, 1982.
16. ASME B31.1-2001. Power Piping: ASME Code for Pressure Piping, B31.1 – American
National Standard, The American Society of Mechanical Engineers, New York, NY,
2001.
 6 Corrosion and Fouling

6.1 CORROSION OF COOLING WATER PIPING

6.1.1 Background
In the 1970s, a number of articles appearing primarily in the Journal of the American
Water Works Association recognized the problem of tubercle deposits occurring in
cast iron potable water distribution lines. It was recognized that these tubercles
resulted from pitting corrosion, because they consisted largely of iron oxides and
were found overlying pits. Past studies suggested that microorganisms accelerated
the corrosion reactions and enhanced tuberculation. However, these early studies
were primarily concerned with the problems associated with red water stemming
from the oxidized iron released into potable water supplies during periods of hydrau-
lic upsets and high flow demand. The iron content was at times sufficient to stain
ceramics and clothing. Although screening showed no evidence for the presence of
bacteria as a public health concern, an increase in heavy metals accompanied mate-
rial loosened from the tubercle exterior during hydraulic stress. The potential was
also recognized for esthetic degradation, including red water and taste problems. The
loss of hydraulic capacity was also noted. The early publications did not focus on the
impact that the corrosion was having on the structural integrity of the piping.

6.1.2 Tennessee Valley Authority Investigation

During preoperational testing of the Emergency Equipment Cooling Water (EECW)
system at the Browns Ferry Nuclear Plant (BFNP) during the summer of 1976,
certain HXs were found to be receiving inadequate cooling water flow due to a
buildup of foreign materials on the interior of the CS piping servicing the equipment.
A study was undertaken by TVA to determine the pervasiveness of this problem in
the TVA system and to develop recommended practices to mitigate its effects in the
design of future power plants.1 A CS raw water piping sampling program was initi-
ated to determine the extent of the problem in TVA power plants existing at that time.
Approximately 50 piping samples were removed from nine different power plants
and analyzed by the TVA Central Laboratories to determine the chemical composi-
tion of the buildup, the average pipe inside diameter reduction, the average pipe wall
reduction, and the maximum pipe wall thinning. Although large differences were
found in some of the above parameters between samples removed from the various
plants, the problem was found to be widespread. In fact, difficulty was frequently
encountered in locating original pipe samples in many of the plants because much of
the CS raw water piping had become unserviceable and had been replaced.
One foot long (0.305 m) pipe samples were removed from not only CS cooling
water lines but also fire protection lines, etc., in both flowing and stagnant service
from existing TVA coal-fired power plants including Colbert (C), Gallatin (G), John

61
62 Engineering of Power Plant and Industrial Cooling Water Systems

Sevier (JS), Watts Bar (WB), Widows Creek (WC), Kingston (K), and Cumberland
(CU) and nuclear plants including BFNP and SNP. The sections of piping removed
from the various raw water systems were kept hydrated and delivered for analysis to
the TVA Power Service Center Laboratories in Chattanooga, Tennessee. A measured
length of a piping sample was sealed at one end and filled with water. The volume of
water contained in the sample was compared with the original volume as calculated
from the nominal dimensions of a new pipe. The percent volume occupied by the
deposit thus determined represented the average loss in pipe cross-sectional area and
was related to an average decrease in pipe diameter. A 4-inch (10.2 cm) length of
each pipe was split lengthwise, scraped, and acid cleaned in inhibited hydrochloric
acid. The sections were weighed and compared to the weight calculated for new pipe
of nominal size. The change in pipe weight was then directly related to the change in
pipe wall thickness. A visual inspection of each piping sample was made after
cleaning to identify the deepest pit or area of maximum wall thinning. The remaining
wall thickness in the deepest pit was measured with a dial micrometer with a
1/32-inch (0.079 cm) diameter anvil. The deposit in each sample was analyzed for
various constituents.
Figure 6.1 is illustrative of the amount of corrosion products and other material
found inside of the piping. Although such pictures are very common in the technical
literature today, when the piping was removed and examined by the author in the late
1970s the implications were staggering.
Observed and calculated data for the piping samples are shown in Table 6.1. The
data in Table 6.1 is sorted by the number of years that the piping has been in service.
Samples that had been in constant flow service are noted as “C” and those that had
been in mostly stagnant flow service are noted as “S”.
Figure 6.2 shows the average and maximum wall thinning from Table 6.1. As one
may see, whereas the average wall thinning appeared not to be a strong function of
age, the maximum wall thinning (i.e. pitting) was. The average ratio between the
average and maximum wall reduction was approximately 3.

FIGURE 6.1 3-inch pipe removed from widows creek fossil unit 8.
Corrosion and Fouling 63

TABLE 6.1
Analysis of Samples Taken from Deposits in Raw Water Piping
Ave. Max. Ave. Pitting

Wall Wall Corr. Corr. Pit/Ave % Deposit Analysis

Sample Flow Age Red. Red. Rate Rate Ratio Fe3O3 SiO2 S Mn3O2
ID Cond. (Years) (in) (in) (MPY) (MPY)
SNP-1 C 2.0 0.010 0.036 5.10 18.00 3.53 87.5 0.4
CU-7 C 5.0 0.014 0.057 2.80 11.40 4.07 89.0 9.0 0.4 1.9
CU-6 C 5.0 0.025 0.019 5.00 3.80 0.76 81.2 17.0 0.4 2.7
CU-2 S 5.0 0.011 0.064 2.20 12.80 5.82 88.9 8.0 0.6 1.5
CU-8 C 5.0 0.023 0.055 4.60 11.00 2.39 78.0 14.0 0.3 1.5
CU-4 C 5.0 0.014 0.027 2.80 5.40 1.93 85.2 9.0 .05 1.4
CU-1 C 5.0 0.020 0.014 4.00 2.80 0.70 88.5 8.0 0.4 1.8
CU-5 C 5.0 0.021 0.075 4.20 15.00 3.57 74.0 21.0 0.6 2.0
WC-17 C 10.0 0.035 0.064 3.50 6.40 1.83 86.0 7.0 0.1 2.5
K-4 S 12.0 0.018 0.038 1.50 3.17 2.11 82.5 6.6 1.6
K-5 S 12.0 0.033 0.068 2.75 5.67 2.06 79.8 4.5 1.0
K-6 S 12.0 0.032 0.07 2.67 5.83 2.19 71.5 4.3 1.3
WC-4 C 12.1 0.014 0.046 1.12 3.80 3.38 90.4 0.5
WC-23 C 13.8 0.019 0.095 1.38 6.91 5.00 81.0 8.1 2.4
WC-22 C 13.8 0.018 0.053 1.31 3.85 2.94 79.1 6.9 2.4
WC-21 C 13.8 0.035 0.067 2.55 4.87 1.91 65.0 20.1 3.7
WC-24 C 13.8 0.017 0.045 1.24 3.27 2.65 56.3 27.4 5.2
WC-19 C 17.0 0.039 0.063 2.29 3.71 1.62 70.0 10.0 0.1 3.5
WC-20 C 17.0 0.031 0.059 1.82 3.47 1.90 87.0 4.0 0.1 3.5
WC-16 S 17.0 0.021 0.062 1.24 3.65 2.95 90.0 6.0 0.1 3.0
WC-18 C 17.0 0.033 0.07 1.94 4.12 2.12 78.0 7.0 0.1 2.6
JS-2 S 17.7 0.020 0.061 1.11 3.45 3.10 75.8 0.5 2.2
G-1 C 17.8 0.022 0.077 1.26 4.33 3.44 88.1 0.8
G-2 S 19.5 0.028 0.138 1.44 7.08 4.93 78.2 9.9 1.9 1.5
G-3 S 19.5 0.024 0.16 1.23 8.21 6.67 74.8 10.3 1.8 1.6
G-4 S 19.5 0.024 0.092 1.23 4.72 3.83 82.6 8.3 2.5 1.4
G-5 S 19.5 0.031 0.102 1.59 5.23 3.29 75.5 8.2 2.2 1.4
JS-1 C 21.5 0.016 0.079 0.72 3.67 5.10 3.7 0.1 72.4
C-1 C 22.2 0.027 0.073 1.20 3.29 2.74 82.9 0.3
JS-6 S 23.0 0.045 0.086 1.96 3.74 1.91 89.5 7.0 1.1 1.4
K-8 S 23.0 0.007 0.042 0.30 1.83 6.00 79.4 4.8 1.1
JS-7 C 23.0 0.048 0.096 2.09 4.17 2.00 69.0 9.0 0.4 3.7
JS-8 C 23.0 0.028 0.04 1.22 1.74 1.43 44.0 13.0 0.3 4.2
K-7 S 23.0 0.017 0.143 0.74 6.22 8.41 81.8 5.4 1.3
JS-3 S 23.0 0.024 0.044 1.04 1.91 1.83
JS-4 S 23.0 0.040 0.046 1.74 2.00 1.15 91.0 6.0 1.3 0.6
JS-5 S 23.0 0.061 0.096 2.65 4.17 1.57 85.6 6.0 0.7 1.3
K-10 S 23.0 0.025 0.128 1.09 5.57 5.12 90.3 4.2 0.9
K-9 S 23.0 0.022 0.044 0.96 1.91 2.00 80.2 3.4 0.8
WC-2 C 23.8 0.030 0.122 1.24 5.13 4.12 38.6 0.6
WC-15 S 25.0 0.006 0 0.24 0.00 0.00 63.4 11.0 .4 3.2
WC-6 C 25.0 0.026 0.065 1.04 2.60 2.50 89.0 133.0 0.4 3.1
WC-13 C 25.0 0.022 0.036 0.88 1.44 1.64 77.9 7.0 0.3 2.4
WC-7 C 25.0 0.018 0.06 0.72 2.40 3.33 80.8 8.0 0.5 2.3
WC-8 C 25.0 0.016 0.129 0.64 5.16 8.06 72.3 15.0 0.8 2.8
WC-10 C 25.0 0.015 0.052 0.60 2.08 3.47 80.0 7.0 0.7 1.6
WC-11 C 25.0 0.054 0.095 2.16 3.80 1.76 72.5 7.0 0.4 1.9
WC-9 C 25.0 0.012 0.045 0.48 1.80 3.75 71.5 18.0 0.9 2.8
WC-12 C 25.2 0.027 0.046 1.07 1.83 1.70 77.8 5.0 0.4 1.5
WB-7 C 33.0 0.028 0.079 0.85 2.39 2.82 90.0 7.0 0.8 1.3
(Continued)
64 Engineering of Power Plant and Industrial Cooling Water Systems

TABLE 6.1 (Continued)

Ave. Max. Ave. Pitting

Wall Wall Corr. Corr. Pit/Ave % Deposit Analysis

Sample Flow Age Red. Red. Rate Rate Ratio Fe3O3 SiO2 S Mn3O2
WB-6 C 33.0 0.024 0.057 0.73 1.73 2.38 81.8 8.0 0.3 1.5
WB-2 C 34.1 0.035 0.1 1.02 2.93 2.88 84.4 1.0
WB-3 C 35.0 0.016 0.057 0.46 1.63 3.56 93.7 4.0 0.3 1.0
WB-5 C 35.0 0.011 0.019 0.31 0.54 1.73 89.7 5.0 0.5 1.2
WB-4 C 35.0 0.018 0.13 0.51 3.71 7.22 92.5 5.0 0.4 1.2
Average 3.05 78.1 11.6 0.6 3.5

C = Constant flow.
S = Normally stagnant flow.

0.18
0.16
0.14
Wall Thining (in)

0.12
0.10
Ave.
0.08
Max.
0.06
0.04
0.02
0.00
0 5 10 15 20 25 30 35 40
Years Service

FIGURE 6.2 Pipe wall thinning.

From Table 6.1, it can be seen that JS-5, WC-1, JS-7, and JS-6 had the highest
values of average wall reduction. Samples JS-5, JS-6, JS-7, and WC-11 all had vary-
ing degrees of exterior corrosion. This exterior corrosion also affects the maximum
wall thinning calculations so that the maximum thinning for these samples would
have been somewhat reduced if the exterior corrosion had been controlled. Varying
degrees of exterior corrosion were noted on all these samples with JS-5 being the
worst. Control of exterior corrosion would allow these values to be discarded. Data
from stagnant lines were seen to fall within the same areas as the continuously flow-
ing samples. Large differences in the average wall reduction are seen in samples
removed from a given site and even between samples removed from a given
pipeline.
Samples WC-15, WC-20, WB-8, WB-9, and WB-10 were removed from galva-
nized lines. Galvanizing appears to be more effective in controlling maximum wall
thinning than in controlling average wall reduction. Much more data from galvanized
Corrosion and Fouling 65

lines would be necessary before any conclusions could be made, however. The gal-
vanized sample data should be used with caution since the maximum wall thinning
values are calculated from the remaining wall thickness measured by a micrometer.
The calculated values are underestimates of the actual thinning since the assumed
initial wall thickness of the pipe does not account for the additional wall thickness
due to the zinc coating. This results in negative values of maximum wall thinning for
some samples. WB-8 was found to be a unique galvanized sample. Removing the
insulation from the piping exterior to obtain the sample revealed a 0.4-inch (1.02 cm)
diameter hole rusted through the pipe wall. It is speculated that the zinc coating was
defective at this point, exposing the CS to the stagnant water in the line. The corro-
sive galvanic attack was accelerated due to the small area of CS pipe exposed to the
galvanized portion of the line.
As seen on Figure 6.2, large variations in maximum wall thinning were found
with the variations in data for samples from one site or from one pipeline at the same
site and even greater than that for the average wall thinning data. Values of maximum
wall thinning up to 0.160-inch (0.406 cm) were found in the observed data. The larg-
est values of maximum wall thinning were seen in some of the 8-inch (20.3 cm)
samples at G (G-2 and G-3) and the 6-inch (15.2 cm) samples from K (K-7 and
K-10). Tubercles approaching 2 inches (5.1 cm) in height were found in these sam-
ples. It was found that areas of maximum wall thinning were usually found beneath
the large tubercles, indicating that maximum wall thinning is a function of tubercle
size. The average diameter reduction of these samples was less than some of the
smaller diameter samples due to only isolated instances of the large tubercles.
Figure 6.3 shows a plot of the average and pitting corrosion rate for each specimen
as determined by the method described above.
As one may see from Figure 6.3, both the average corrosion rate and the pitting
corrosion rate appear to decline with age, suggesting that cleaning the pipe without
adding an inhibitor may accelerate the rate of corrosion. The average corrosion rate
was 1.7 MPY.

20
18
16
Corrosion Rate (MPY)

14
12
10
Ave.
8
6 Pitting
4
2
0
0 5 10 15 20 25 30 35 40
Years Service

FIGURE 6.3 Average and pitting corrosion rate.

66 Engineering of Power Plant and Industrial Cooling Water Systems

The scope of the 1979 TVA study was not intended to predict a 40-year design
value of average wall reduction or maximum wall thinning (i.e. corrosion allowance
“A” in Equation 5.1) or the impact on system design of the reduced wall thickness but
only to report the observed data and to make any comments regarding trends or pecu-
liarities. It was observed that the samples having large values of average wall reduc-
tion also had varying degrees of exterior corrosion. If all samples are considered,
average wall reductions reaching 0.0625 inches (0.159 cm) were seen. Discarding
samples with obvious extreme exterior corrosion drops the maximum value of the
average wall reduction to 0.040 inches (0.102 cm). Values of maximum wall thinning
up to 0.160 inches (0.406 cm) were found in the observed data. It is noted that only
one sample was found with through-wall leaks, since they would normally have
already been replaced. Therefore, the maximum wall thinning could be much higher
than indicated by the 1979 study. No significant differences were observed in the
corroded condition of horizontal versus vertical runs of pipes as long as the pipes
were completely full of raw water.
In 1980, some of the results of the TVA investigation focusing on the flow-passing
capability of the piping were reported by Bowman and Bain.2 (See Chapter 7.)
Figure 6.4 from Reference 2 shows the effective decrease in diameter of the pipes
due to material buildup as determined by the method described above as a function
of years of service.
One may see that the degree of material buildup varied greatly from sample to
sample, but generally it appeared to be progressive with the age of the pipe. In
stagnant, continuously pressurized piping systems, the rate of material accumulation
is generally lower than that found in raw water piping systems where water flows in
a continuous or nearly continuous manner, thereby replenishing the oxygen supply
which can then induce further corrosion. However, data from some stagnant lines fall
within the same areas as the data from flowing lines.

FIGURE 6.4 Average reduction in pipe diameter vs. years of service.

Corrosion and Fouling 67

From the scatter of data seen on Figure 6.4, it can be seen that age was not the only
parameter influencing corrosion product buildup. Large variations in buildup are
seen for piping removed from a given site at a given age (e.g. WC and WB) and in
some cases large variations can be seen from samples removed from a single pipeline
(e.g. WC-21 through WC-24 and G-2 through G-5). It was found that the average
buildup in 8-inch (20.3 cm) diameter piping (G-2 through G-5) and 6-inch (15.2 cm)
diameter piping (K-7 through K-10) was on the same order of magnitude as the
buildup in the 2, 3, and 4-inch (5.1, 7.6, and 10.2 cm) lines taken at JS and WC.
Buildup did not appear to be dependent on pipe diameter.
Large differences in the appearance and consistency of the corrosion product
buildup were found. In some cases, materials other than corrosion products were
found on the pipe interior. At JS, some of the samples were found to have a large
amount of manganese deposit. Sample WC-24 was found to have a higher level of
silica than other samples. Most of the samples had a relatively uniform buildup with
a very rough surface like that shown in Figure 6.1. However, some samples such as
K-10 had almost no average buildup but had large, randomly spaced isolated tuber-
cles. The samples removed from WB appeared to not be any worse than those
removed from WC and JS even though the WB piping had been installed approxi-
mately 20 years earlier. However, It should be noted that WB was out of service for
a significant period of time (10–15 years) and the status of the piping during that
period is not known. The effective age of the piping at WB could actually be less than
that shown. It can be seen, in most cases, that the diameter reduction in galvanized
lines is less than that of the other samples. It can be seen from Figure 6.2 that age
effect is only one part of the explanation for material buildup and that the worst
buildup observed actually occurred between 20 and 25 years of service.
Reference 1 concluded that although the appearance and consistency of the
buildup on the inside wall of the pipe samples varied, iron oxide was virtually always
the principle constituent, and the primary mechanism was always corrosion of the
steel piping by the aerated raw water with redeposition of the corrosion products onto
the inside wall of the pipe in the form of irregular tubercles. The entire process was
determined to be induced or at least influenced by sulfur-reducing bacteria (SRB).
Corrosion of raw water piping and the resultant redeposit of ions of corrosion product
onto the inside of the pipe were found to a significant degree at all plants that were
sampled. Under each tubercle was almost invariably found an area of reduced pipe
wall thickness. The results of the study indicated that at the end of the 40-year life of
a plant in the TVA system, CS raw water piping will experience an average reduction
in the inside diameter of 0.40 inch (1.02 cm), an average reduction in the pipe wall
thickness of 0.065 inch (0.165 cm), and a maximum wall thinning of at least 0.160
inch (0.406 cm) due to corrosion. As a result of the 1979 TVA study (Reference 1)
TVA issued References 3 by J. R. Alley as guidance for design engineers in dealing
with CS piping in raw water service.
Reference 1 also presented the results of pressure drop tests performed in straight
sections of CS raw water piping at three TVA power plants. These tests conducted to
determine the impact of corrosion products buildup on the interior pipe wall on the
flow-passing capability of the pipe are addressed in Reference 2 and in Chapter 7.
68 Engineering of Power Plant and Industrial Cooling Water Systems

6.1.3 Industry Response to MIC

TVA immediately notified the NRC of the flow deficiencies discovered at the
BFNP and the results of subsequent investigations at SNP and WBNP through
nonconformance reports, as noted in Chapter 1. TVA also authorized the publica-
tion of Reference 2 in 1980, which presented the results on Table 6.1 above but
focused mainly on the problem of flow deficiencies. In 1981, TVA provided the
NRC with a copy of Reference 1 along with their final report on the nonconfor-
mance reported in 1980 (NCR No. SQNNEB8035). The NRC did not formally
notify the nuclear power industry of the MIC problem until 1985 with the issuance
of Reference 4 which only alerted licensees of the potential for through-wall leaks
in butt-welded SS piping in raw water service. The basis for this IN was a condi-
tion reported by Carolina Power and Light (CP&L) at their Robinson Nuclear
Plant in which minor pinhole leaks were found in the heat affected zones of cir-
cumferential welds jointing 6-inch (15.2 cm) diameter, schedule 10, 304 SS pip-
ing that provides service water to the four containment chilling units. Subsequent
visual inspection of the system revealed leakage at a total of 54 weld joints. CP&L
determined that the root cause of the problem was the result of MIC. The NRC
noted to the industry that the CP&L experience was only one of a number of simi-
lar incidents reported to them. The NRC did not warn the nuclear power industry
of the MIC problem in CS cooling water piping until 1989 with the publication of
Reference 5.
Although INPO issued Reference 6 in 1984, the first significant response to the
call for a response to the problem of MIC was Reference 7 by G. Licina, sponsored
by EPRI. Reference 7 provides a discussion of MIC along with 15 case histories. For
each case investigated, Reference 7 reported the operating history, the method of
discovery, and a characterization of the material in the pipe. The water chemistry was
reported for some cases. The pictures of the piping samples are similar to Figure 6.1
and many others presently found in the literature on MIC, including some of MIC
attacks on both CS and SS welds. Licina noted that in CS pipe, MIC may result in
both general corrosion and random pitting, causing the formation of tubercles
anywhere around the circumference of the pipe and on horizontal or vertical surfaces.
In 1993, EPRI completed their Service Water System Corrosion and Deposition
Sourcebook, a comprehensive reference document. EPRI continues to support their
members through additional publications and through regular meetings of their
Service Water Assistance Program (SWAP).
In 1980, TVA undertook a significant program of material upgrade to address the
corrosion problem. In 1982, the first through-wall pit in a cooling water system was
discovered in a 3-inch (7.6 cm) CS pipe in the BFNP EECW system. The report
concluded that the apparent cause of the pitting was the buildup of biological material
attributed in part to the presence of Asiatic clams in the system. Multiple leaks were
also discovered in the fire protection system at BFNP that year. The TVA began a
program of ultrasonic testing (UT) inspection of 11 EECW system CS piping areas
at BFNP. In 1989, UT inspections conducted on the BFNP CS fire protection system
piping indicated unacceptably high corrosion rates.
Corrosion and Fouling 69

In 1986 in response to Reference 4, a leak was found in a SS butt weld in the

stagnant auxiliary feedwater supply line of the WBNP ERCW system. A metallurgical
evaluation showed that a condition similar to that reported to CP&L existed. As a
result of the discovery in the ERCW system at WBNP, inspections at BFNP and SNP
revealed similar leaks in applicable corresponding systems at those plants. The
radiographic testing (RT) of SS piping in the EECW system at BFNP indicated only
minor MIC damage.
Figure 6.5 shows a 6-inch (15.2 cm) SS butt weld from SNP sectioned to show
MIC damage. Nondestructive examination (NDE) of the damage to SS butt welds in
the ERCW system at SNP suggested that, unlike MIC damage in CS resulting in
relatively benign bowl-shaped pits, MIC pits in SS can lead to subsurface cavities in
the butt weld and heat-affected zone that threaten the structural integrity of the
piping. In addition, the large number of through-wall pits implied a very aggressive
attack. When welds were removed and replaced, in some cases leaks appeared in the
new welds within six months.
In SS piping, the MIC damage may be characterized as pits oriented in the locale
of the butt welds. As may be seen in Figure 6.5, the pits are quite insidious because
they exhibit very small entry holes with much larger subsurface cavities which
frequently proceed through-wall with a small exit hole. The resulting leaks are on the
order of a few drops per minute. The pits occur in random fashion around the
circumference of the weld. The corrective action developed by Pavinich and Deardorff
included the development of a methodology to assess the damage existing in the butt
welds of SS piping to assure that the structural integrity of the piping was not vio-
lated before the damaged pipe was identified and replaced.8,9,10,11
Austenitic SS exhibits significant toughness properties, resistance to fatigue
damage, and resistance to ductile tearing. Therefore, the failure mechanism assumed
for gross structural failure under seismic loading is plastic collapse under limit load
conditions. ASME Section III design margins against limit load plastic collapse may

FIGURE 6.5 Stainless steel butt weld showing MIC damage.

70 Engineering of Power Plant and Industrial Cooling Water Systems

be calculated. The approach was an appropriate adaptation for raw water piping of
the procedures found in ASME Section XI (IWB-3640) for evaluating flaws in
austenitic SS piping in ASME Section III, Class I systems, which provides convenient
but overly conservative means of evaluating MIC damage. In this approach, all MIC
indications must be added together in the circumferential direction and evaluated as
a single flaw oriented in the direction of the primary bending to produce the weakest
net sectional properties. Additional refinements have been developed to eliminate
this conservatism as required on a case-by-case basis. The approach taken is to
establish a screening criteria of allowable total length of MIC defects based on the
IWB-3640 methodology for a given line size and enveloping load magnitudes for a
given system. All welds with a total MIC damage indication of less than this value
are structurally adequate. Welds with greater MIC damage were evaluated
individually, first by using the methodology with location-specific loads and then
with the less conservative methods if required. It should be noted that all methods of
evaluation assumed that all MIC indications were completely through-wall and make
no allowance for sound metal that may exist above the MIC cavities. This approach
was required because existing methods of NDE did not permit reliable evaluation of
MIC damage in the radial direction at the time the flaws were evaluated.
In 1987, TVA began investigating corrosion rates by means of corrosion coupons.
Although this technique cannot predict the long-term corrosion rate, it is useful to
determine the relative short-term corrosion rate as a function of time, temperature,
and various water treatments. TVA conducted tests at the SM&E Laboratory during
which CS coupons were exposed in a once-through service water flow loop drawing
water from the Tennessee River at Knoxville, Tennessee. The results of these tests are
shown in Table 6.2.
These data are shown in Figure 6.6.
Blackburn and Mullin12 reported on the corrosion rates measured in a condenser
waterbox at the BFNP in 1987–1989. Figure 6.7 shows the corrosion rates measured
at BFNP by an array of different durations of coupon exposure. As with Figures 6.3
and 6.6, the same trend of lower corrosion rates with time is observed. Although the
magnitude of corrosion as a function of duration of coupon exposure rates varies
considerably between the three studies, the data consistently suggest that the average
corrosion rate is a function of service life and that the rate decreases with time.

TABLE 6.2
Coupon Corrosion Tests in Cooling Water
Start End Time Corrosion Rate
Date Date (Days) (MPY)
10/22/1987 11/23/1987 32 5.35
11/24/1987 1/4/1988 41 3.60
1/8/1988 2/12/1988 35 5.94
11/24/1987 3/21/1988 120 2.78
2/24/1988 4/7/1990 773 1.59
11/22/1987 4/7/1990 898 1.86
Corrosion and Fouling 71

5
Corrosiom Rate (MPY)

0
0 200 400 600 800 1000

Duration of Cupon Exposure (Days)

FIGURE 6.6 Coupon corrosion tests in cooling water at SM&E.

13

12

11
Average Corrosion Rate (MPY)

10

5
30 40 50 60 70 80 90 100 110 120 130
Duration of Coupon Exposure (Days)

FIGURE 6.7 Corrosion rate vs. temperature in BFNP main condenser.

72 Engineering of Power Plant and Industrial Cooling Water Systems

Figure 6.8 shows the corrosion rates measured at BFNP by 60-day coupons as a
function of the service water temperature. This study suggests that the corrosion rate
increases with increased cooling water temperature.
Although the average corrosion rates reported above are useful in assessing the
various factors that influence corrosion, the pitting corrosion rate is the parameter
that is of more interest to the designer. Unfortunately, much of the published litera-
ture reports on corrosion monitoring performed with weight loss coupons which nor-
mally do not reflect pitting corrosion rates. Figure 6.3 shows the corrosion rate
calculated based on the deepest pit observed in the piping samples removed from the
nine different TVA power plants as discussed above. When compared with average
corrosion rates, pitting corrosion appears to occur at about three times the average for
the same specimens. Although the general trend of corrosion rates declining with
service life appears to hold true for pitting corrosion, caution should be exercised in
drawing conclusions regarding an appropriate pitting corrosion rate based on
Figure 6.3. Piping specimens that have experienced through-wall pits are not included
in this sampling, since they have long since been replaced. Through-wall leaks began
to occur in the EECW and fire protection piping at BFNP after approximately 13
years of service. The first leaks occurred in 3-inch (7.6 cm) standard wall piping,
which implies a pitting corrosion rate of 16.6 MPY. These initial leaks were followed
by 80 to 100 additional leaks over the next few years before several hundred feet of
piping were replaced in 1988. In 1993, a through-wall leak occurred in a 16-inch
(40.6 cm) standard weight pipe at BFNP, implying a pitting corrosion rate of 16.3
MPY. WBNP reported leaks in small-bore [2 to 6-inch (5.1 to 15.2 cm)] standard
weight piping after about 14–17 years of service resulting in calculated pitting

20

18
Average Corrosion Rate (MPY)

16

14

12

10

0
50 55 60 65 70 75 80 85
Average Service Water Temperature (F)

FIGURE 6.8 Corrosion rate vs. temperature in BFNP main condenser.

Corrosion and Fouling 73

corrosion rates of 8 to 16 MPY. In 2001, several leaks were discovered in the 12-inch
(30.5 cm) standard wall [0.375-inch (0.9525 cm)] WBNP fire protection supply
header from the IPS after 23 years of service for a pitting corrosion rate of 16.3 MPY.
The entire header was subsequently replaced.
In 1993, TVA initiated a comprehensive assessment, repair, and replacement of
the raw water piping systems at the SNP. The results of the RT data for the SS butt
welds showed a large percentage containing small, randomly distributed MIC defects.
None of the 55 butt welds evaluated failed to meet the structural integrity criteria.
The SS butt welds inspected in the ERCW system that showed indications of MIC
were either monitored or replaced with Inconel 625 filler metal. Figure 6.9 shows the
location of the only socket weld inspected that showed possible indications of MIC.
An analysis of this weld at SM&E concluded that socket welds are less prone to MIC
due to the location of the fillet weld. Since full penetration does not normally occur,
the weld is not in contact with the cooling water.
Visual inspection of the interior of the SS butt welds where MIC damage had
occurred showed that a small tubercle existed at the interior of the pipe at the weld
joint. A water treatment program to eliminate or minimize the extent of attack at the
SS butt welds by preventing the formation of these tubercles is discussed in
Section 6.4.1.
Figure 6.10 shows the typical condition of the CS piping encountered at SNP.
Random leaks have occurred in low flow areas of 2-inch (5.1 cm) and smaller cooling
water and fire protection piping, and in some cases, when the corrosion products
were removed, pits were found to be through-wall. CS piping was inspected by UT.
The “as found” minimum pipe wall thickness was compared with the minimum wall
thickness required to provide structural integrity. If the thickness was less than the
required value, a specific minimum wall thickness calculation using point-specific
stresses was performed. The UT data indicated random pitting throughout the CS
systems. The more severe pits that were detected were small in diameter and clearly

FIGURE 6.9 Socket weld with MIC damage at SNP.

74 Engineering of Power Plant and Industrial Cooling Water Systems

FIGURE 6.10 Three-inch carbon steel RCW pipe from sequoyah nuclear plant.

isolated with considerably thicker base metal surrounding each pit. The study con-
cluded that gross degradation leading to structural failure of CS piping is unlikely,
but random leaks could reasonably be expected. The average corrosion rate in the
cooling water systems at SNP was determined to be 4 MPY after approximately
17 years of service. However, some locations exhibited corrosion rates as high as
13 MPY. Although this effort provided useful information on the condition of the
system, it did not preclude the possibility that pitting corrosion at even higher rates
were occurring at other points in the system.
In 1994, Bowman and Guthrie13 summarized the results of the TVA studies of CS
corrosion in cooling water systems and the implications for piping system designers
in selecting an appropriate corrosion allowance “A” in Equation 5.1 in the future.
Tuovinen, et al,14 recognized that tubercles resulted from pitting corrosion because
they consisted largely of iron oxides and were found overlying pits. Past studies sug-
gested that microorganisms accelerated the corrosion reactions and enhanced tuber-
culation. The bacteria found in most cooling water systems make them particularly
vulnerable to corrosion. Touvinen conducted an experiment in which cast iron cou-
pons were immersed in various test solutions for four weeks. These solutions included
tap water, distilled water, and a liquid medium inoculated with samples of red water
and ground tubercles. Although coupon surface reactions took place under all assay
Corrosion and Fouling 75

conditions, the corrosion was twice as great in the inoculated medium. This led
Touvinen to conclude that bacteria could be involved in the mechanism of corrosion.
There was some hope that most of the iron showing up in the tubercles had been
scavenged from soluble iron found in the raw water, as was the case with the silica
and manganese found in the deposits. However, that hope was refuted by a study
conducted by Vermont Yankee Nuclear Power Corporation and published by Metell15
in1986. In that study, the author conducted tests on removable spool pieces of CS
pipe over a five-year period. The masses of the pipe spools were determined before
and after the test. The deposits were removed and the pipe was weighed once again
to yield the mass of the pipe after the test and the mass of the deposits. A volume/
density test and molar/mass balance showed that the deposits were composed of 10%
base metal, 90% oxygen and water on a volume basis. A comparison was made
between the oxide deposit based on the base metal loss and the actual oxide deposit.
The good agreement between these values supports the conclusion of the study that
the oxide was almost exclusively from corrosion and not deposition of soluble iron
from the source stream. Further, the study observed that the deepest pits had the
largest tubercles, that the iron lost from the pipe wall ended up in the overlying
tubercle, and that the formed tubercle is approximately ten times larger than the
deepest pit.. Metell observed that mechanical cleaning to remove tubercles to improve
system flow characteristics promotes corrosion of the pipe wall.
In 1986, the American Nuclear Society published Reference 16 as an American
National Standard for nuclear safety cooling water systems in light water reactors.
The Standard stated that these systems should be constructed of corrosion-resistant
materials, be protected by corrosion inhibitors or cathodic protection, or have
sufficient wall thickness allowance to ensure the integrity of the system over the
design life of the plant. Considering the fact that by that time, construction of virtually
all of the nuclear plants in the United States had been completed, this document,
though useful for future plants, was rather like closing the barn door after the horse
had escaped.
In 1990, the NRC published Reference 17. For Code Class 3 piping, a licensee is
required to perform code repairs or request NRC to grant relief for temporary non-
code repairs on a case-by-case basis regardless of pipe size. Because of the rather
frequent instances of small leaks in some Class 3 systems such as cooling water
systems, the NRC provided guidance that would be considered by the NRC staff in
evaluating relief requests for temporary non-code repairs of code Class 3 piping. The
guidance consisted of assessing the structural integrity of the flawed piping by a flaw
evaluation and assessing the overall degradation of the system by an augmented
inspection. The document concluded that “temporary non-code repair of Class 3
piping that cannot be isolated without a plant shutdown is justified in some instances.
The rather frequent instances of small leaks in some Class 3 systems, such as service
water systems, could lead to an excessive number of plant start-up and shutdown
cycles with undue and unnecessary stress on facility systems and components if the
facilities were to perform a code repair when the leakage is identified.”17
In 1991, the ASME published Reference 18 as a guide for determining the remain-
ing strength for corroded pipelines (not power piping, e.g. B31.1). Unlike MIC in
butt-welded SS where the damage is preferential to the transverse axis of the pipe
76 Engineering of Power Plant and Industrial Cooling Water Systems

(i.e. the weld) making the pipe more vulnerable to seismic and dead weight loads, the
MIC damage in CS is random and catastrophic failures in the longitudinal direction
may be possible. Although Reference 18 is not applicable to power piping, per se, the
document contains guidance as to how to determine the maximum allowable longi-
tudinal extent of corrosion.
In 2007, INPO issued a Significant Event Report on inadequate design resulting
in cooling system leaks at several nuclear plants.

6.1.4 MIC Tubercles
In addition to the role that microorganisms play in corrosion of CS service water,
general corrosion and the differential aeration cells set up by the tubercles are also
factors.
General corrosion, present when CS is exposed to an oxidizing agent such as
aerated water, can increase when an oxidizing biocide such as chlorine is used to
control bacteria. (See Chapter 12.)1 This requires the designer to balance the benefits
that the biocide may provide in controlling the pitting corrosion from the MIC against
the increase in general corrosion.
Although differential aeration cells can be associated with any form of crevice, in
raw water they are most commonly associated with tubercles. Figure 6.11 shows a
typical tubercle in CS pipe.19
All tubercles have five structural features in common:

1. Outer crust
2. Inner shell
3. Core material
4. Fluid-filled cavity
5. Corroded floor.20

FIGURE 6.11 Tubercle in carbon steel piping.

Corrosion and Fouling 77

As discussed by Metell15 and Herro,19 much of the tubercle material is formed of cor-
rosion products from the underlying metal. The tubercle often contains an inner cav-
ity filled with fluid. As tubercles increase in size with age, free oxygen within
decreases and differential aeration cell activity is promoted. Anions concentrate
within the tubercle, lowering the pH of the fluid inside the tubercle even if the bulk
fluid pH is alkaline. Since both differential aeration cells and MIC cause pitting cor-
rosion, often occurring at the same site, distinguishing between the two in service
water is often difficult. MIC is considered by many experts to be the most serious
form of corrosion in service water systems. In the case of MIC, several types of bac-
teria often work together to attack the pipe. Slime-forming bacteria (psudemonas)
produce a biofilm which becomes the host for iron-reducing bacteria, which are aero-
bic. This iron-reducing bacteria then can provide a covering for SRB, which are
anaerobic. SRBs are often found inside tubercles and produce the characteristic odor
of hydrogen sulfide when the tubercle is penetrated while still wet. This discussion is
an oversimplification of the subject. All or none of the above may occur in a specific
case.

6.1.5 Potential Improvements to Combat Corrosion in Cooling Water

Systems
The designer must be concerned with the impact that corrosion will have on the
structural integrity and serviceability of CS cooling water piping. Since ASME
Section III is a design code, it does not provide explicit rules for evaluation or
acceptance of degradation due to corrosion. However, the design should retain the
structural margins inherent to the design requirements stipulated in Section III by
defining acceptable degradation due to anticipated corrosion. Subsection ND-3613.1
states “When corrosion or erosion is expected, the wall thickness of piping shall be
increased over that required by other design requirements”. In determining the mini-
mum thickness of pipe wall required for the design pressure, Equation 3 of ND-3641
requires the designer to specify an additional thickness “A”, which is to include an
appropriate corrosion allowance to consider all forms of corrosion including pitting
corrosion. (See Equation 5.1.) When considering sustained (deadweight, thermal
expansion, etc.) and occasional (seismic, etc.) loads, Code pipe stress equations in
Subsection ND-3650 stipulate the nominal wall thickness without any consideration
for corrosion. For typical piping systems under normal or service level “A” plant
conditions, axial stress usually does not control structural design margins. However,
for service water piping designs with either a large corrosion allowance and/or a
conservative size applied, moment loading may dominate the net required wall thick-
ness, particularly for occasional loads. Long-term maintenance of design structural
margins can be assured through in-service monitoring.
Figures 6.6 and 6.7 suggest that as the interior surfaces of CS pipes become coated
with oxides, the average corrosion rate steadily declines with exposure time. Figure
6.8 suggests that the rate of MIC corrosion is influenced by the temperature of the
cooling water as well as water quality and flow velocity. The results of any corrosion
monitoring program must be interpreted in light of these factors. When considering
an additional thickness to be applied to the pipe wall required for the design pressure,
78 Engineering of Power Plant and Industrial Cooling Water Systems

pitting corrosion should be considered. The pitting corrosion rate can be as much or
more than three times the average corrosion rate. All three TVA nuclear plants have
experienced through-wall leaks in CS piping. Based on the pipe wall thicknesses and
service life of piping, a pitting corrosion rate of 13–16 MPY or more for untreated
raw water is not unreasonable. Data obtained from an effective monitoring program
can be beneficial in making reasonable predictions as to pipe integrity and operability
at a given site. (See Section 6.4.)
The designer should seek to minimize the impact of MIC in designing the service
water system. The system operating water velocity is one parameter that the designer
should consider to help reduce the impacts of corrosion. Metell15 reported that fast
moving streams appear to interrupt tubercle formation, and that stagnant and low
velocity water appears to deprive the bacteria of required oxygen, reducing growth.
He stated that moderate velocities [2 to 7 ft/sec (0.61 to 2.1 m/s)] encourage bacteria
growth. Figures 6.12 and 6.13 show 24-inch (61.0 cm) headers in the SNP ERCW
system. In Figure 6.12, the normal cooling water velocity exceeds 10 ft/sec (3.05 m/s),
whereas the flow in the redundant header shown in Figure 6.13 is normally zero.
Exceedingly high normal velocities (over 10 ft/sec) (3.05 m/s) can lead to erosion in
CS cooling water piping.
Normally, the sedimentation that would be associated with velocities below
3 ft/sec (0.91 m/s) would promote the growth of the anaerobic SRB and so most
experts recommend that normally flowing systems be designed for at least 3 ft/sec
(0.91 m/s) to keep solids in suspension. Provisions for chemical treatment such as
injection points, special provisions to make sure that the chemicals will reach all
parts of the system, and provisions for side-stream monitoring or spool pieces should
be included in the system design.

FIGURE 6.12 24-inch cooling water header where velocity exceeds 10 ft/sec.
Corrosion and Fouling 79

FIGURE 6.13 24-inch cooling water header with normally zero velocity.

Depending on the desired service life, standard wall pipe thicknesses may not be
adequate for CS piping. This is especially true for piping 3-inch (7.6 cm) and smaller.
Socket welded SS pipe may be employed that is inherently much less susceptible to
MIC. If SS piping is used for larger piping requiring butt welds, an effective biocide
treatment designed to prevent the formation of MIC nodules over the welds is imper-
ative. (See Section 6.3.1.) For larger piping, CS may be acceptable for hung piping
systems as long as the size of the piping is increased to account for the increased
pressure drop due to MIC and the pipe wall thickness is increased to account for the
anticipated rate of pitting corrosion. (See Chapter 7.)

6.2 FOULING IN COOLING WATER PIPING

6.2.1 Background
Biofouling is the undesirable accumulation of biotic deposits on a surface. The
deposit may contain micro- and/or macro-organisms. Microfouling is the develop-
ment of a biofilm such as pseudomonas, on the surface in conjunction with inorganic
particles such as suspended solids, scale, or corrosion products.21 As such, micro-
fouling is often a precursor to MIC. Macrofouling is the existence of biological crea-
tures such as mollusks or aquatic weeds such as seagrass or milfoil. Suspended solids
such as mud and sand or fish such as gizzard shad are also sources of fouling in
cooling water systems.
Biofouling that is often prevalent in cooling water systems can be the source of
many problems. Microfouling per se is not normally an impediment to flow as is
corrosion, but when it appears on the inside diameter of HXs, heat transfer is impeded.
Macrofouling is often the more serious impediment to flow, as it can not only block
the entrance to HX tubes but also obstruct the IPS. Suspended solids can also be a
problem in HXs, normally collecting and sometimes completely blocking the lower
tubes in a HX.
80 Engineering of Power Plant and Industrial Cooling Water Systems

For many years, fouling had been acknowledged to be an inevitable problem in

cooling water systems, but with the advent of commercial nuclear power plants,
fouling became unacceptable when it challenged the ability of nuclear power plants
to shut down safely. (See Section 1.8.) The result was an industry-wide effort to
identify biofouling and develop solutions to the problem. Some biofouling problems
such as fish, seagrass, and sediment are quite site specific, while others like
microfouling and macrofouling such as certain mollusks are quite widespread and
are treated in more detail herein.

6.2.2 Asiatic Clams
The Asiatic clam (corbicula fluminae), an exotic freshwater bivalve mollusk, was
first discovered in North America near the mouth of the Columbia River in 193822
and has since spread throughout the United States. Unless precautions are taken,
Asiatic clams often enter the cooling water system as adults or larvae when the sys-
tem was initially flooded or soon after it began operation. Once in a cooling water
system, Asiatic clams can grow rapidly, as much as 1/10-inch (0.254 cm) per month,
and may grow to be 2 inches (5.1 cm) in size and can produce veligers throughout
the year.23 Adult Asiatic clams cannot attach to a surface and have limited mobility
with a single “foot”, so they are frequently found in low areas or areas of low veloc-
ity. As documented in Section 1.8, they have been a major concern to the NRC, as
they have been the cause of numerous challenges to safety-related cooling water
systems that are used only infrequently. The clams settle in upstream areas and grow
until the system is called upon to operate. They are then sometimes swept up in the
cooling water flow until they obstruct an important piece of equipment such as a
valve or HX. Asiatic clams elude periodic chemical treatment regimens, as they
sense the presence of the chemical and can close up until the danger passes. However,
veligers smaller than 1/32-inch (0.079 cm) have not yet developed a protective shell
and are thus susceptible to biocides. Asiatic clams spawn when the cooling water
temperature is above 60°F (15.6°C) and have been reported to spawn well below
65°F (18.3°C).23

6.2.3 Zebra Mussels
Unlike the Asiatic clam, the zebra mussel’s byssal threads can attach to almost any
surface.24 Zebra mussels have spread rapidly, as their planktonic (drifting) veliger
stage remains in the water for several weeks after release, allowing it to be carried
great distances before settlement. By 1991, zebra mussels were found in the TVA’s
Kentucky Lake near where the Tennessee, Ohio, and Mississippi Rivers converge. In
1994, EPRI predicted that zebra mussel infestations would eventually be found in
70% of the states. Presently, they may be found in 28 states or over 50% of the states.
Zebra mussels can be found where the pH is greater than 5.5 and the water tempera-
ture is between 54 and 90°F (12.2 and 32.2°C).24 Since they feed on suspended sol-
ids, zebra mussels have brought about great change in the ecology of bodies of water
such as the Great Lakes, resulting in a shift in species composition due to the increas-
ing clearing of the water.25
Corrosion and Fouling 81

Unlike the Asiatic clam, the zebra mussels attach to hard surfaces by byssal
threads to almost any surface.26 The zebra mussel can grow up to 2 inches (5.1 cm)
in length27 and are a serious hazard to cooling water systems, as they may be found
on IPS trash racks and forebays, system piping, and HXs.24 Fortunately, zebra mus-
sels may be controlled by a combination of monitoring and chemical treatment or
heat treatment. Sodium hypochlorite in doses of less than 1.0 ppm is effective in
controlling zebra mussels.28 However, areas that may be up stream of the biocide
injection point such as IPS trash racks and forebays pose a challenging problem.

6.3 CHEMICAL AND HEAT TREATMENT

6.3.1 Chemical Treatment
Continuous chlorination with sodium hypochlorite of less than 1.0 ppm is widely
used at power stations to control mollusks. Chlorination dosing at low levels not only
stops the settlement of larvae but also reduces drastically the growth of microfouling
and kills or removes the biofouling in the long term more effectively than is achieved
with intermittent dosage.22 Chlorine concentration tests indicate that there is hardly
any difference between chlorine concentrations of 0.5 and 1.0 ppm in controlling
biofouling.25 Continuous or long-term molluscicide treatment is more effective than
short-duration intermittent treatment because of the mollusks’ inability to avoid
long-term exposure.29 However, chemical controls are more effective at preventing
the settlement and growth of mollusks than mitigating the effects of established
colonies. No chemicals have proved able to dissolve or remove the byssal threads
which zebra mussels use to attach themselves to the IPS and piping.24
Since 1979, TVA has had good experience controlling Corbicula, except where
mechanical or operational problems were experienced that interrupted chlorina-
tion.30 Bleach feed systems are often plagued with frequent down time due to the
corrosivity of the chlorine bleach and the pumps becoming air locked from off-
gassing of the bleach. Problems with the equipment resulted in long periods of no
treatment and some periods of over treatment.31 TVA recommended that all incom-
ing water to the cooling water systems should be strained. Straining is performed
by automatic backwash type straining units located immediately upstream or
downstream of the main pumping units of the system (i.e., at the source). Strainers
have 1/32-inch (0.079 m) wire mesh baskets and are designed for periodic or con-
tinuous backwashing. TVA initially elected to use sodium hypochlorite generated
onsite as needed to be injected upstream or immediately downstream of the strain-
ers to kill the veligers before they get a chance to develop protective shells.
Secondary water sources (such as jockey pumps, normally open interconnections
with other water systems, etc.) also should be strained and chlorinated. If the
incoming water has already been chlorinated, no additional injection is necessary.
The chlorine level throughout most cooling water systems and at the system dis-
charge is to be maintained at a total residual chloride (TRC) of 0.6 to 0.8 ppm dur-
ing the entire clam spawning period. The clam spawning period as defined here is
that period of time when the system inlet temperature normally exceeds 60°F
(15.6°C).30
82 Engineering of Power Plant and Industrial Cooling Water Systems

System design must be considered in conjunction with plant chemical discharge

limits in defining the exact location of chemical injection. It should be remembered
that the cooling water systems that are chlorinated often constitute only a small flow/
volume in comparison with the CCW system in a power plant and if the cooling
water discharge of the system is mixed with CCW before being discharged to the
environment, the impact of the chlorine on the environment may be minimal.
For safety-related cooling water systems, chlorination should be continuous to
protect butt-welded SS piping from MIC attack by depriving it of pseudomonas
bacteria that enable MIC to initiate at the welds. Additionally, veligers that enter the
plant in the winter can grow shells before chlorination resumes. Experience has
shown that when the chlorination is interrupted even though the cooling water is
strained, control of mollusks may be lost. Once the veligers settle in the system, they
can grow a shell and when chlorination is resumed, they are able to close and avoid
being killed.30

6.3.2 Coatings for Zebra Mussel Control

Laboratory tests of silicone elastomer-based “foul release” coatings have shown
favorable performance over a 24-month test period for the Diablo Canyon power
plant located on the Pacific Ocean.31 Two coatings containing toxic compounds that
were deemed acceptable for use in the marine waters of Long Island Sound,
New York, were cuprous oxides and copper epoxies. The single nontoxic coating
found to be effective is silicone rubber impregnated with silicone oil expected to last
five years. The silicone coating was also expected to be effective in controlling zebra
mussels in freshwater environments.32

6.3.3 Heat Treatments for Zebra Mussel Control

European research over the years has identified several lethal combinations of tem-
peratures and durations for zebra mussels ranging from 91°F (32.8°C) for five hours
to 97°F (36.1°C) for one hour with instant mortality at 104°F (40.0°C).27 This range
of intake temperatures is confirmed by experience in the United States where a tem-
perature of 92°F (33.3°C) for 2.5 hours was found to be 100% effective.33 At Detroit
Edison’s Harbor Beach power plant, an oxygen scavenger (sodium sulfide) was used
in conjunction with thermal treatment to asphyxiate and heat zebra mussels over a
several day period.34 The heating of cooling water has been accomplished by recircu-
lating CCW and by the use of auxiliary boilers.

6.4 TVA QUALIFICATION TESTING OF CHEMICAL TREATMENT

6.4.1 Corrosion Inhibitor Testing
Tests were conducted at SNP in order to make the results directly applicable to the
SNP ERCW system. Zinc polyphosphate was tested by TVA as the type of corrosion
inhibitor most likely to be both environmentally acceptable and cost-effective in their
 Corrosion and Fouling 83

applications. Recognizing that cooling water systems were to be chlorinated for

Asiatic clam control, sodium hypochlorite was tested as a potential corrosion
inhibitor along with zinc polyphosphate. The test stand included a series of three
1-foot long (0.305 m), 1-inch (2.54 cm) CS spool pieces arranged in parallel, each
carrying approximately 15 gal/min (0.95 l/s). Zinc polyphosphate was injected into
one line at approximately 1.5 ppm. Sodium hypochlorite was injected into the second
line at a rate intended to yield a chlorine residual of 0.6 to 0.8 ppm at the outlet of the
line. The third line served as a control with no treatment of the cooling water. A
sample spool piece of pipe was removed from each of the three lines on four occasions
over a period of about one year. The volume reduction of the interior of each sample
was measured as described in Section 6.2.1 to determine the extent of corrosion
products buildup.1
The sample spool pieces were removed after 1, 3, 6–1/2, and 11–1/2 months of
service, and results are plotted in Figure 6.14.1
The samples taken at one month and at three months showed rapid accumulation
of deposits. Later samples showed a slower accumulation and in some cases, an
apparent reduction in deposits. All of the test samples had numerous small tubercles
which were similar in appearance to one another and to the large tubercles observed
in the sampling program described in Section 6.4. The earlier samples had numerous
tubercles up to about 1/16-inch (0.159 cm) high but also had sizeable areas which
were completely covered with tubercles up to about 1/8-inch (0.318 cm) high. The
average diameter reduction in the later samples was on the order of 1/16-inch (0.159)
in the untreated line and the corrosion inhibited line and on the order of 1/10-inch
(0.254 cm) in the chlorinated line.1

0.12

0.10
Diameter Reduction, inches

0.08

0.06

0.04

No Treatment
0.02 Chlorination
Zinc Polyphosphate

0.00
0 2 4 6 8 10 12
Time Aftrer Start of Test, months

FIGURE 6.14 Results of corrosion inhibitor test as measured by diameter reduction.

84 Engineering of Power Plant and Industrial Cooling Water Systems

The observed high initial rate of buildup and lower rate later in the test are consis-
tent with the sample data described in Section 6.1.2 and plotted in Figure 6.4. The
study concluded the following:

• The rate of buildup was initially very rapid but had slowed dramatically by the
end of the test.
• The chlorinated line experienced more rapid buildup of deposits than the other
two lines.
• Zinc polyphosphate in the concentrations tested had little if any beneficial
effect under the test conditions.
• Corrosion-resistant materials would be needed in many instances instead of CS
because of the serious corrosion problems with CS and the lack of effective-
ness of corrosion inhibitors.1

6.4.2 Biocide Testing
Prior to implementing the proposed cooling water system chemical treatment regi-
men, TVA sponsored testing by various organizations to ensure the feasibility of the
proposed regimen. The questions that needed to be answered were as follows:

• Is the regimen harmful to the environment?

• Will the regimen damage the cooling water system components?
• Is the regimen efficacious?

Toxicity tests were conducted by the Ecology Laboratory of the Oak Ridge National
Laboratory (ORNL) to provide the necessary technical information to support regu-
latory approval for the use of the biocide in the ERCW system at SNP. The biocide
tested was Actibrom 1338, a proprietary bromine and dispersant mixture manufac-
tured by Nalco Corporation. The tests were conducted with the Actibrom 1338 mixed
with sodium hypochlorite at a 1:1 molar ratio. Tests conducted included both a
Ceriodaphnia seven-day survival and reproduction test and a fathead minnow larval
growth test also of seven days’ duration. The Actibrom 1338/sodium hypochlorite
mixtures (1:1 molar ratios) did not result in increased mortalities at concentrations in
excess of 30 ppm.
Qualification tests were conducted by the SM&E Laboratory to evaluate the
potential adverse effects of candidate biocides on the material of construction of the
wetted parts of a typical cooling water system and to ensure that the introduction of
the biocide into the system does not cause the corrosion products buildup on the CS
piping to become detached from the surface of the pipe wall. A test was conducted in
which three samples of 2.5-inch (6.4 cm) CS pipe and three samples of 6-inch
(15.2 cm) SS pipe all containing corrosion products including tubercles were sub-
jected to a 30-day exposure. One sample each of carbon and SS was exposed to
untreated raw water, one each was exposed to a raw water solution treated with a
solution of generic sodium bromide activated by sodium hypochlorite, and one each
was exposed to raw water treated with Nalco Actibrom 1338 mixed with sodium
hypochlorite. The biocides were initially injected at 7.5 ppm of free available
Corrosion and Fouling 85

chlorine with a 1:4 molar ratio of bromine to chlorine. A visual and tactile inspection
of the SS specimens at the end of the test revealed an absence of slime on the surfaces
of biocide-treated specimens, but no significant amount of corrosion products was
removed from the CS specimens.35
Corrosion tests were conducted on materials typically found in ERCW systems to
determine the relative corrosivity of candidate biocides on these materials in a sterile
solution of simulated raw water. The primary purpose of this test was to determine if
the addition of bromine or bromine with a dispersant increases the corrosion rate
when compared with sodium hypochlorite alone. The test was accelerated somewhat
by employing a biocide level of 7.5 ppm, which was five times the anticipated level.
Biocides containing bromine did not appear to be more corrosive than sodium
hypochlorite alone. In fact, the addition of commodity sodium bromide resulted in a
corrosion rate decrease for all CS specimens tested, accompanied by an increase in
the amount of passivated surface area. The slight increase in the corrosion rate with
Actibrom 1338 appeared to be due to the dispersant in this solution acting to remove
corrosion products from the surface of the CS. All copper alloys tested exhibited
good protection in the simulated raw water without biocide and significant corrosion
rates when exposed to sodium hypochlorite. All of the copper alloys showed a
decrease in corrosion rate when bromine was present in the biocide. All of the
austenitic SS specimens and weld coupons exhibited very low corrosion rates.
However, all of the austenitic SSs except AL-6XN exhibited some susceptibility to
crevice attack in one or more of the solutions, and some of the austenitic stainless
welded coupons exhibited minor crevice attack in the solutions containing
bromide.36
A test was conducted to evaluate the potential degradation effects on Buna-N
rubber materials which are used as O-rings in the ERCW system. Some of the O-rings
tested were sandwiched between corrosion coupons used in the test described above
while others were stretched over the coupon assemblies to provide the compressive
force for the sandwiched O-rings. Physical property measurements were selected for
their response to potential degradation, due to exposure to the biocides and included
hardness, tensile strength, elongation, and modulus. The results of these tests,
performed in accordance with the appropriate industry standard, showed no physical
property degradation as a result of exposure to the biocide.37
Tests were conducted by the University of Tennessee’s Institute for Applied
Microbiology (IAM) to determine the effectiveness of various biocides at controlling
sessile and/or planktonic bacteria populations and on controlling MIC. An initial test
was conducted on MIC nodule on a 6-inch (15.2 cm) SS weld removed from the SNP
to determine if a biocide consisting of chlorine and a dispersant could penetrate the
nodules and provide growth inhibition of the corrosive microorganisms without
dissolution or removal of the nodule. The test was conducted in a re-circulating
system in which the test water was replaced with fresh lake water treated with biocide
at approximately 48-hour intervals to maintain a total available chlorine concentration
of approximately1 ppm. During the test, the pH varied between 7.4 and 8.75. No
significant reduction in the size of the nodules was observed. Biological tests
conducted by IAM concluded that although there remained a very diverse population
86 Engineering of Power Plant and Industrial Cooling Water Systems

of organisms in the nodules, their numbers were very low and those that were present
were in a dormant condition.38
A series of tests was conducted by IAM which were specifically designed to
determine the biocide composition, dosage, and application time required to be
effective controlling MIC-causing bacteria.38 In six identical test stands, CS and SS
specimens were exposed to a re-circulating sterile medium. Three of these test loops
were inoculated with a mixed consortium of six types of bacteria including FeRed,
FeCcpt, SRB, and Ps taken from TVA nuclear plants. Within a few days, sessile
bacterial colonies were formed as tubercles on the CS specimens in the inoculated
loops (tubercles did not form on the SS specimens). After the colonies were
established, two of the three inoculated loops were treated with a biocide, with the
third serving as a control. The first test loop was treated with NaOCI alone, the
second was treated with commodity NaBr/NaOCI. The same biocides were added to
two of the sterile loops with the third sterile loop left as a control. Direct electrical
measurements were made of the corrosion occurring in mohms, and microbes were
recovered after each test and the number of colony-forming units (CFU) was
determined. The activity of the biofilms was determined by measuring the C-14
acetate uptake by the microbial lipids on the specimens.
Two separate experiments were conducted. In the first experiment, both biocides
were fed to maintain a residual halogen of 2 ppm for one hour. This was followed in
24 hours with an additional exposure of 2 ppm for two hours. The NaBr/NaOCl
biocide was mixed at a 1:4 molar ratio and pH was maintained at 8.5. This treatment
regimen proved to be totally ineffective for both biocides. At the end of the second
biocide injection, the corrosion rates for the inoculated specimens that were treated
with biocides were higher than the control, and the viable organisms recovered from
the specimens showed no evidence of damage to the bacteria and in many cases there
was an increase in viable cells. These results suggest that ineffective or sub-lethal
pulses of biocide actually stimulate the bacteria. Each of the specimens, with and
without biocide additions, developed a corrosion products layer over the specimen
which when removed revealed a smooth surface. In contrast, those specimens that
were inoculated with bacteria exhibited severe pitting on the metal surface after
removal of the biofilm.
A second experiment was designed in which both biocides were fed at slug dose
of 16 ppm for two hours followed by a continuous residual halogen of 2 ppm for 24
hours. The NaBr/NaOCl biocide mix was changed to a 1:1 molar ratio, and the pH
was kept at 8.5. This second treatment regimen had a marked biological effect.
Although the presence of oxidants in the system increased the general corrosion rate
both with and without bacteria present, the test results shown on Figure 6.15 indicate
that both biocides were effective in killing bacteria, and that the NaBr/NaOC1
biocides may be slightly more effective on sessile (coupon) bacteria than NaOC1
alone. This result is to be expected since the pH for the test was selected to be
prejudicial against chlorine alone. The data also suggest that both biocides are
effective against a broad spectrum of bacteria. Figure 6.15 suggests that although a
biocide may lower the number of CFUs below 106/ml, thus rendering the biofilm
inactive, even bacteria that were suppressed below detectable levels were able to
partially recover within 24 hours after termination of biocide treatment. Micrographs
Corrosion and Fouling 87

22
sterile
20 unknown
HCl/Br
18 NaClO
no biocide
16

14
dpm (thousands)

12

10

0
0 10 20 30 40 50
time (hrs)

FIGURE 6.15 Activity of biofilm measured as C14-acetate.

of the specimens tested showed that pitting corrosion characteristic of MIC is greatly
reduced when an effective biocide is employed support the numerical data.38

6.5 TVA EXPERIENCE WITH CHEMICAL TREATMENT

6.5.1 Sequoyah Nuclear Plant
The ERCW system at TVA’s SNP employs automatic backwash strainers as depicted
in Figure 3.6 near the discharge of each of the 8 ERCW pumps. These strainers filter
out objects larger than 1/32-inch (0.079 cm). SNP was the first TVA nuclear plant to
implement a biocide treatment of their safety-related ERCW system. As discussed in
Section 6.3.1, continuous biocide treatment is important. For example, in March
1982, while conducting a surveillance test of the ERCW system at SNP, a flow
decrease to the containment spray HX from 100% to 31% of rated capacity was
noted. The pipe was opened upstream of the manual inline strainer, revealing approx-
imately 15 gallons (0.0568 m3) of clam shells that were restricting the flow. Under
normal operating conditions, the 18-inch (45.7 cm) header supplying the HX is stag-
nant except for a 1-inch (2.54 cm) mini-flow line around the HX. The mini-flow line
was found to be clogged. The ERCW was strained but only periodically chlorinated
in the summer of 1981. This condition was conducive to clam growth. Subsequently,
steps were taken to ensure flow through the mini-flow line and chlorination by repair-
ing the hypochlorite injection systems. Continuous chlorination as described above
was practiced in 1982 with complete control of Asiatic clams.30
88 Engineering of Power Plant and Industrial Cooling Water Systems

FIGURE 6.16 Halogen residual and corrosion rates in the SNP ERCW system.

Injection of the chlorine/bromine biocide Nalco Actibrome 1338 began at SNP in

January 1989. Figure 6.16 shows the residual halogen and corrosion rates before and
after the change. As one may see, shortly after the change to Actibrome 1338, the
corrosion rate in the system as measured by corrosion coupons and the required
residual halogen dropped dramatically. In addition to greatly reducing the CS corro-
sion rate, the biocide was very effective in cleaning up the system so that there is less
demand for oxidant while still maintaining a free, available halogen residual.
Although qualification tests conducted by the SM&E Laboratory concluded that
the introduction of a biocide into the system would not cause the corrosion products
buildup on the CS piping to become detached from the surface of the pipe wall,35 one
might conclude from the results of its implementation at SNP that the biocide gradu-
ally cleans up the system. The improvement is likely due to not only the hyprobro-
mous acid but also the dispersant in the Actibrome 1338. As discussed in Section 7.4,
flow and pressure drop tests conducted on the ERCW system headers between the
IPS and the auxiliary building support the suggestion that biocide treatment over a
long period of time tends to clean up the system, since the pressure drop in the piping
decreased.39,40 However, plant personnel reported a deposit of iron oxide and silt in
the test water box and on the tubes of some HXs that was not there on prior inspec-
tions, which may result in the loss of heat transfer capability.41

6.5.2 Watts Bar Nuclear Plant

The ERCW system at TVA’s WBNP employs automatic backwash strainers as
depicted in Figure 3.6 near the discharge of each of the 8 ERCW pumps. These
strainers filter out objects larger than 1/32-inch (0.079 cm). WBNP injected sodium
hypochlorite on a continuous basis any time the temperature of the cooling water was
above 60°F. In December of 1992, TVA implemented a chemical treatment program
Corrosion and Fouling 89

FIGURE 6.17 WBNP carbon steel corrosion rate vs. cooling water temperature.

at WBNP consisting of bromochlordimethylhydantoin (BCDMH), a granular form

of bromine/chlorine, used to reduce the concentration of bacteria that can cause MIC.
BCDMH was chosen because it is more effective at pH > 7.5 and is a dry chemical
that is safer to work with and is a more reliable feed system. Cleanup of the ERCW
system was accomplished by continuous application of polyphosphate to soften and
remove MIC nodules over time and a copolymer dispersant to maintain sequestered
small particles in suspension to be removed by the flowing water. As shown in Figure
6.17, CS corrosion rates were controlled by continuously injecting a zinc-based cor-
rosion inhibitor in addition to the polyphosphate. The chemicals were injected into
the IPS in front of the traveling screens to eliminate problems with overcoming the
high ERCW pump discharge pressure.41
The BCDMH was injected continuously for three weeks twice each year during
the clam spawning season. DGH-QUAT, a non-oxidizing biocide, was used for 12 to
24 hours on a quarterly basis to flush out normally stagnant lines to control Asiatic
clams and zebra mussels.41

6.5.3 Browns Ferry Nuclear Plant

At BFNP, the strainers at the discharge of the cooling water pumps strain to 1/8-inch
(0.318 cm) which permits veligers that have already developed a shell to enter the
system. Sodium hypochlorite was fed at BFNP with a target TRC of 0.2 to 0.8 ppm
when the cooling water temperature exceeded 50°F (10°C) to control Asiatic clams
and MIC in butt welds. The hypochlorite feed system was frequently out of service.
Problems with the equipment resulted in long periods of no treatment and some peri-
ods of overtreatment with residuals as high as 6 ppm. Despite the treatment program,
Asiatic clams continued to foul HXs. In 1995, BFNP implemented a different chemi-
cal treatment program for the cooling water systems. For biological control, bromine
90 Engineering of Power Plant and Industrial Cooling Water Systems

was added to sodium hypochlorite to produce hyprobromous acid and fed four hours/
day with a halogen residual (0.1–0.5 ppm). One reason that bromine activated by
sodium hypochlorite was preferred over straight chlorine is that it is less corrosive
than chlorine. Other advantages include better performance at elevated pH and less
persistence in the environment. However, hypobromous acid fed for a few hours at a
time has very limited effect on macrofouling by clams and mussels. The chemical
treatment program also included phosphate (1–1.8 ppm), zinc (0.2–0.25 ppm), and
Copolymer (0.28–0.35 ppm) fed continuously and Azol blend (3.5–4.0) 15 minutes
1–2 times/month, DMAD (0.5–0.62) fed for 30 minutes once a week, and Quat
(0.15–1.5 ppm) fed for 24 hours twice a week. BFNP reported that the results were
as good or better than expected.42

6.6 CONCLUSIONS AND RECOMMENDATIONS

The cooling water system is perhaps among the most poorly engineered systems in a
power or industrial plant. Most employ CS piping and make little or no provision for
controlling corrosion or fouling. Unless a strict chemical regimen is observed at the
outset, CS piping is not an appropriate material for small bore piping and often is not
suitable for large bore piping. As shown in Section 6.1, pitting corrosion rates for CS
piping due to MIC can be as high as 16 MPY or more, leading to through-wall leaks
in as few as 23 years in standard 3/8-inch (0.9525 cm) wall piping for pipe sizes over
12 inches (30.5 cm) and even sooner for smaller line sizes. Further, as shown in
Chapter 7, the flow-passing capability of CS piping is greatly reduced due to the
tuberculation associated with MIC that increases the roughness factor and reduces
the effective inside diameter of the pipe. Studies have shown that in terms of the flow-
passing capability of the piping, the installed cost of SS piping (the preferred material
for hung piping) may be approximately the same as for CS. However, butt-welded SS
piping may be vulnerable to MIC attack at the weld joint, which can lead to leaks and
loss of structural strength over time. Therefore, a biocide treatment is indicated
where SS piping 4-inch (10.2 cm) and larger is employed.
A very large percentage of power and industrial plants are subject to fouling, both
microfouling that initiates MIC and reduces heat transfer in HXs and macrofouling
that obstructs flow. Straining of cooling water at the source is essential to a successful
biocide treatment program. Particles larger than 1/32-inch (0.079 cm) should be
strained out with automatic backwashing strainers such as the S. P. Kinney Model A
strainer with wire baskets is shown in Figure 3.6. Veligers smaller than 1/32-inch
(0.079 cm) have ciliated flaps for swimming and feeding but have not yet developed
hard shells allowing them to close in the presence of a biocide. Therefore, they may
be killed instantly with a low dose of an oxidizing biocide such as chlorine. Failure
to adequately strain cooling water makes controlling mollusks almost impossible, as
they can sense the presence of a biocide and close up their shells.
A biocide system is essential to managing most fouling problems such as mol-
lusks. The cheapest biocide is chlorine, normally sodium hypochlorite (bleach) a
liquid that is highly corrosive and often difficult to handle. Bromine is sometimes
mixed with chlorine to make hyprobromous acid. The advantage of bromine is that it
is less corrosive than chlorine (especially with copper-alloyed tubes), is more
Corrosion and Fouling 91

effective at pH > 7.5, and is less persistent in the environment. However, if the residu-
als are kept low and the cooling water is mixed with the CCW before being dis-
charged, a chlorine-only biocide is not normally offensive. Commercial biocides
such as Nalco’s Actibrome 1338 also contain a proprietary dispersant. Actibrome
1338 has been shown to clean up corroded CS piping over time. However, the result
may be a relocation of the iron and other constituents on the surfaces of HX tubes.
Handling and mixing liquid chlorine and bromine in the right proportions can be
challenging. In some locations where pH < 7.5, one may be able to achieve satisfactory
results with just sodium hypochlorite alone or in conjunction with a polyphosphate to
soften and remove MIC nodules over time and/or a copolymer dispersant to maintain
sequestered small particles in suspension to be removed by the flowing water if required.
By far the greatest impediment to a successful biocide treatment program is the
reliability of the chemical feed system. Adequate engineering must go into the design
of these systems to ensure reliable operation. Consideration should be given to
installing redundant components such as pumps, etc., or even full redundant trains of
equipment to achieve the required 100% availability. This job is often left to the
chemical manufacturer without adequate oversight.
Common practice is to suspend biocide treatment except during the mollusk
spawning season or to rely on periodic shock treatments. Biocide treatment programs
that rely on intermittent operations permit veligers to enter the cooling water system
where they hide out and grow the protective shell. Even if the periodic treatment is of
such duration as to starve the mussel, the result is a dead mussel in the system, which
is hardly better than a live one.
In areas where zebra mussels are a problem, one should control their growth
upstream of the point of biocide introduction. The use of antifouling paints such as
cuprous oxides, copper epoxies, or nontoxic coatings such as silicone rubber impreg-
nated with silicone oil should be considered. The system at WBNP of introducing
BCDMH (commonly used in swimming pools) upstream of the trash racks and/or
traveling screens is of interest in that it holds promise for controlling zebra mussels
as well as Asiatic clams. One might surmise that this arrangement could lead to accel-
erated corrosion of the traveling water screens if they are made of CS.
Unlike oxidizing biocides such as chlorine or bromine where the residual becomes
relatively inert when their demand is satisfied, non-oxidizing biocides such as Clam-
trol or DGH-QUAT are poisons that are long-lasting and very expensive. Their use in
open cooling water systems is highly problematic, generally requiring the system to
be shut down, isolated, and flushed after the mollusks have been killed.
The perfect is often the enemy of the good. In some cases, the cost of the chemical
treatment system has become so high as to become unaffordable, resulting in the
elimination of chemical treatment all together. Care should be taken to implement the
least costly system that is effective. In most instances, this will be determined by site-
specific conditions.

REFERENCES
1. Bain, W. S., et al. Corrosion in Carbon Steel Raw Water Piping, (TVA Ref.
ESS800416204) Tennessee Valley Authority, Knoxvville, TN, 1979.
92 Engineering of Power Plant and Industrial Cooling Water Systems

2. Bowman, C. F. and W. S. Bain. A New Look at Design of Raw Water Piping. Power
Engineering Magazine, Vol. 84, No. 8, pp. 73–77, August 1980.
3. TVA Mechanical Design Guide DG-M5.2.1. Corrosion/Erosion Allowance for
Determination of Minimum Pipe Wall Thickness in Carbon Steel Piping Systems,
Tennessee Valley Authority, Knoxville, TN, 1980.
4. IN 85-30. Microbiologically Induced Corrosion of Containment Service Water System,
Nuclear Regulatory Commission, Washington, DC, 1985.
5. Generic Letter 89-13. Service Water System Problems Affecting Safety-Related
Equipment, Nuclear Regulatory Commission, Washington, DC, 1989.
6. “Microbiologically Influenced Corrosion (MIC)”, Institute of Nuclear Power Operations
Significant Event Report SER 73–84, 1984.
7. Licina, George. Sourcebook for Microbiologically Influenced Corrosion in Nuclear
Power Plants, Electric Power Research Institute, Palo Alto, CA, 1988.
8. Pavinich, W. A. et al. Four-Point Bend Testing of MIC-Damaged Piping, (TVA Ref.
B45890825251) Tennessee Valley Authority, Knoxville, TN, 1989.
9. Deardorff, A. F., et al. Evaluation of Structural Stability and Leakage from Pits Produced
by MIC in Stainless Steel Service Water Lines. Corrosion 89, 1989.
10. Pavinich, W. A. et al. Mechanical Properties of Stainless Steel Weldments Degraded by
MIC, Paper No. 531. Corrosion 90, 1990.
11. TVA Civil Design Standard DS-C1.2.8, Structural Evaluation of Microbiologically Induced
Corrosion Degradation in Piping, Tennessee Valley Authority, Knoxville, TN, 1991.
12. Blackburn, F. E., and Mullin, L. J. Detection and Control of Bacterial Corrosion Using
Internal Corrosion Monitors. Microbially Influenced Corrosion and Biodeterioration,
MIC Consortium, Knoxville, TN, 1990, pp. S1–S24.
13. Bowman, C. F. and P.V. Guthrie, Jr. Corrosion in Carbon Steel Service Water Piping.
1994 ASME Pressure Vessel and Piping Conference, ASME, 1994.
14. Tuovinen, J. E., et al. Bacterial, Chemical, and Mineralogical Characteristics of
Tubercles in Distribution Pipelines. Journal AWWA, American Water Works Association,
Denver, CO, pp. 626–635, November 1980.
15. Metell, H. M. Corrosion from Iron Bacteria in Plant River Water Systems. Proceedings,
Workshop on Microbe Induced Corrosion, Electric Power Research Institute NDE
Center, Charlotte, NC, 1986.
16. ANSI/ANS-59.1-1986. Nuclear Safety Related Cooling Water Systems for Light Water
Reactors. American Nuclear Society, 1986.
17. Generic Letter 90-05. Guidance for Performing Temporary Non-Code Repair of ASME
Code Class 1, 2, and 3 Piping, Nuclear Regulatory Commission, Washington, DC, 1990.
18. ASME B31G-1991. Manual for Determining the Remaining Strength of Corroded
Pipelines, American Society of Mechanical Engineers, New York, NY, 1991.
19. Herro, H. M. Corrosion and Fouling in Nuclear Power Plant Service Water Systems,
Electric Utility Service Water System Reliability Improvement-A Compendiun of
Presentations, Electric Power Research Institute, Palo Alto, VA, 1993.
20. Herro, H. M. and R. D. Port, The NALCO Guide to Cooling Water Systems Failure
Analysis, McGraw Hill, Inc., New York, 1993.
21. Characklis, W. G., et al. Biofouling Control Technology: The Roll of Fouling Monitors.
Condenser Technology Symposium, Electric Power Research Institute, Providence, RI,
1987.
22. Jenner, H. A., et al. Hydroecologie Appliquee, Tome 10, Volume 1–2, Electricite de
France, 1998.
23. Derrworth, J. E., et al. Duke Power Company’s Development of a Biofouling Monitoring
Program at Two Nuclear Power Plants on the Southeastern Reservoirs. International
Macrofouling Symposium, Electric Power Research Institute, Orlando, FL, 1990.
Corrosion and Fouling 93

24. Mussalli, Y. G. Zebra Mussel Monitoring and Control Guide, Electric Power Research
Institute, Palo Alto, CA, 1992.
25. Jenner, H. A. Biomonitoring in Chlorination Anti-Fouling Procedures to Achieve
Discharge Concentrations as Low as Reasonable. International Macrofouling
Symposium, Electric Power Research Institute, Orlando, FL, 1990.
26. Armor, A. Zebra Mussels: The EPRI Response Update. International Macrofouling
Symposium, Electric Power Research Institute, Orlando, FL, 1990.
27. DeMoss, D., et al. Design Features to Mitigate Fish, Zebra Mussel, and Other
Macrofouling of Plant Components. International Macrofouling Symposium, Electric
Power Research Institute, Orlando, FL, 1990.
28. Matisoff, G., et al. Controlling Zebra Mussels at Water Treatment Plant Intakes II -
Velinger Dose/Response Static Tests. International Macrofouling Symposium, Electric
Power Research Institute, Orlando, FL, 1990.
29. Garbarino, L. N. Testing of Chemical Methods for Clam Control under Simulated Field
Conditions. International Macrofouling Symposium, Electric Power Research Institute,
Orlando, FL, 1990.
30. Isom, B. G. et al. Controlling Corbicula (Asiatic Clams) in Complex Power Plant and
Industrial Water Systems. American Malacological Bulletin, Special Edition No. 2. pp.
95–98, 1986.
31. Sommerville, D. C. and F. L. Steinert. Development of Alternative Macrofouling Control
Methods for the Diablo Canyon Power Plant. International Macrofouling Symposium,
Electric Power Research Institute, Orlando, FL, 1990.
32. Gross, A. C. Macrofouling Problems and Control Techniques in Marine water around
Long Island. International Macrofouling Symposium, Electric Power Research Institute,
Orlando, FL, 1990.
33. Kahabka, J. The Zebra Mussel: New York’s Experience. International Macrofouling
Symposium, Electric Power Research Institute, Orlando, FL, 1990.
34. Harwood, D. B. and D. J. Buda. Effective Zebra Mussel Control at Detroit Edison
Harbor Beach Power Plant. American Power Conference, Chicago, IL, 1993.
35. Guthrie, P. V., et al. Sequoyah Nuclear Plant Dipersion Tests on Carbon and Stainless
Steel Pipe Samples, Test 6 (R3), SME-COR-88-001, Singleton Materials Engineering
Laboratory, 1988.
36. Guthrie, P. V., et al. Sequoyah Nuclear Plant Evaluation of the Effects of Cl/Br Biocides
on the Corrosion of Metals within the ERCW, SME-COR-88-011, Singleton Engineering
Laboratory, 1988.
37. Guthrie, P. V., et al. Sequoyah Nuclear Plant Evaluation of the Effects of Cl/Br Biocide
Water Treatment on Buna-N Rubber, SME-COR-88-019, Singleton Engineering
Laboratory, 1988.
38. Van, A. et al. Accelerated Tests of Effects of Bromide Additions to Biocide Treatments on
Microbiologically Induced Corrosion (MIC) on Mild Steel, Institute for Applied
Microbiology, 1988.
39. Bowman, C. F. Solving Raw Water Piping Corrosion Problems. Power Engineering
Magazine, Vol. 98, No. 7, pp. 35–38, July 1994.
40. Hewette D. Utility Experience Report – Tennessee Valley Authority. Service Water Working
Group Meeting Proceedings, Electric Power Research Institute, Palo Alto, CA, 1989.
41. Riggle, K. TVA Watts Bar Nuclear Plant – Service Water Chemical Treatment Program,
Service Water Systems Reliability Improvement Seminar, Electric Power Research
Institute NDE Center, Charlotte, NC, 1993.
42. Garbarino, N., et al. Raw Water Treatment at the TVA Browns Ferry Nuclear Plant.
Service Water Reliability Improvement Seminar, Electric Power Research Institute,
Orlando, FL, 1997.
 7 Pipe Flow

7.1 EQUATIONS FOR HEAD LOSS BY PIPE FRICTION

7.1.1 Darcy-Weisbach Equation
Equation (7.1) is the Darcy-Weisbach1 equation for the friction head loss with turbu-
lent flow in piping that is running full.

2
L Vpipe
hL , f = f (7.1)
di 2 g

where
hL,f = friction head loss
f = friction factor
L = length of pipe
di = inside diameter of pipe
Vpipe = water velocity in a pipe or tube
g = acceleration due to gravity.
Equation (7.2) is the value for the friction factor, f, determined from the Colebrook
equation2 as follows:

1  e 2.51 
  2 log10    (7.2)
f  3.7 D Re f 

where
e = absolute roughness of pipe, ft
Equation (7.1) is more commonly expressed in terms of volumetric flow rate in
English units as follows:3

LQ 2
hL , f = 0.0311 f (7.3)
d5

where
Q = Flow rate, Gal/min
d = diameter of pipe, in.

7.1.2 Hazen-Williams Equation
Whereas the Darcy-Weisbach equation provides a more analytically based mathe-
matical solution to the head loss due to pipe friction, the empirical Hazen-Williams4
equation is more appropriate for power plant cooling water system applications. As
95
96 Engineering of Power Plant and Industrial Cooling Water Systems

shown in Chapter 6, the interior of cooling water piping systems is subject to deterio-
ration with age depending on chemical properties of the water and pipe material in
contact with the water. Scaling, corrosion, or bacterial attack can therefore adversely
affect the flow capacity of the system. Some provision must be made in pressure drop
calculations to account for this deterioration of the pipe interior surface. The most
readily available information for a quantitative expression of pipe surface condition
relative to operating experience and length of service life is the “C” factor used in the
Hazen-Williams equation shown in Equation (7.4).4

 100   Q1.85 
hL , f  0.002083 L    4.8655  (7.4)
 C  d 

where
C = Hazen-Williams “C” factor
The Hazen-Williams formula is based on water at 60°F (15.6°C) having a kine-
matic viscosity of 1.13 centistokes. The kinematic viscosity of water can vary from 1.8
for water at 32°F (0°C) to 0.29 centistokes for water at 212°F (100°C), respectively.
However, this variation in kinematic viscosity for the normal range of cooling water
systems results in negligible changes in the friction head loss and may be ignored.

7.1.3 Selection of Roughness Coefficient

The main element of uncertainty in both the Darcy-Weisbach and Hazen-Williams
methods of calculating the head loss due to pipe friction is the choice of the roughness
coefficient (i.e. “e” Darcy-Weisbach and “C” for Hazen-Williams). Table 7.1 lists the
recommended values to be used for the equivalent length from the indicated source.

TABLE 7.1
Commonly Used Roughness Coefficients for Design Purposes
Type of Pipe C e
Cement asbestos 130, 140
5 4 0.00755
Bitumastic enamel-lined iron or steel centrifugally applied 1404
Smooth glass or plastic 135,5 1401 0.0055
Cement-lined iron or steel centrifugally applied 130,5 1404 0.00755
Cement mortar trawled in place 1255 0.0115
Copper, brass, lead, tin, or glass pipe and tubing 1304
Welded and seamless carbon steel 55-70,6 1001,4,5 0.0605
Welded and seamless stainless steel 1206
Wrought iron, cast iron 1001,4,5 0.0605
Tar-coated cast iron 1004
Galvanized iron 905 0.0125
Concrete, formed 80,5 100,4 1301 0.225
Spiral-riveted steel (flow with lap) 1004
Spiral-riveted steel (flow against lap) 904
Corrugated steel 604
Pipe Flow 97

7.2 PIPE FORM LOSSES

7.2.1 Form Loss Equation
Form losses are caused by changes in flow direction in valves and fittings, etc. in a pip-
ing system as opposed to those caused by friction at the pipe wall. Friction losses are
calculated from the pipe centerline intersections of such components. Separate equa-
tions are necessary since form losses are not affected by changes in pipe wall rough-
ness.6 Form losses are independent of the properties of the fluid and of the pipe to which
the valves and fittings, etc., are connected. Since the loss is a function solely of geometry
and fluid velocity, coefficients (i.e., k factors) can be established that are valid for com-
ponents of similar size and shape. Equation (7.5) is the equation relating the form loss to
the “k” factor and the velocity of the cooling water passing through the component.2

V2
hL , F = k (7.5)
2g

where
hL,F = form head loss
k = resistance coefficient.
Tests have been conducted for most commercially available components of simi-
lar size and shape. The k factors for flanged pipe fittings shown in Figure 7.1 and
valves in Table 7.2 are from Reference 5.
The “k” factor for other components such as strainers and HXs must be obtained
from the manufacturer. (See Section 3.4 regarding S. P. Kenney strainers.)

7.2.2 Equivalent Lengths
When performing head loss calculations by hand, common practice is to express the
head loss through components in the cooling water piping system in terms of the
equivalent length of pipe that would produce the same head loss. Table 7.3 shows the
equivalent lengths of piping for an array of components from Reference 4.
The equivalent lengths were calculated for steel pipe carrying cold water. The
error for water in a steel pipe at 100°F (37.8°C) is about -5%. The error for epoxy-
coated or copper pipe is about 15% for water at 40°F (4.4°C) and 25% for water at
100°F (37.8°C). However, since these errors are in the conservative direction, they
are commonly neglected.2
As seen in Reference 6 by F. P. Carr and C. F. Bowman, a separate form of the
Hazen-Williams equation is required for form losses in piping components, since
they are not affected by changes in wall roughness. Therefore, in order to use the
equivalent length type analysis for form losses, the Hazen-Williams “C” factor asso-
ciated with the published data for such losses must be inserted into the equation as
seen in Equation (7.6).6

 Q1.85 
hL , F  0.001486  4.8655  (7.6)
d 
98 Engineering of Power Plant and Industrial Cooling Water Systems

FIGURE 7.1 Recommended energy loss coefficient, k, for flanged pipe fittings.

Using the equivalent length approach, the head losses computed for form losses in
piping components using Equation (7.6) may be directly added to the head loss due
to the friction in the pipe using Equation (7.4) to determine the total head loss through
a run of cooling water piping.
Pipe Flow 99

TABLE 7.2
Recommended Energy Loss Coefficient, k, for Valves Fully Open
Valve type K
Angle 1.8–2.9
Ball 0.04
Butterfly
25-lb Class 0.16
75-lb Class 0.27
150-lb Class 0.35
Check valves
Center-guided globe style 2.6
Double door
8 in or smaller 2.5
10 to 16 in 1.2
Foot
Hinged disc 1–1.4
Poppet 5–1.4
Rubber flapper
V < 6 ft/s 2.0
V > 6 ft/s 1.1
Slanting discd 0.25–2.0
Swingd 0.6–2.2
Cone 0.04
Diaphragm or pinch 0.2–0.75
Gate
Double disc 0.1–0.2
Resilient seat 0.3
Globe 4.0–6.0
Knife gate
Metal seat 0.2
Resilient seat 0.3
Plug
Lubricated 0.5–1.0
Eccentric
   Rectangular (80%) opening 1.0
   Full bore opening 0.5

7.3 THE BOWMAN-BAIN EQUATIONS FOR FRICTION LOSS

During preoperational testing of the EECW System at the BFNP during the summer
of 1976, certain HXs were found to be receiving inadequate cooling water flow due
to a buildup of foreign materials on the interior of the CS piping servicing the equip-
ment. As discussed in Chapter 6, a study was undertaken by TVA to determine the
pervasiveness of this problem in the TVA system and to develop recommended prac-
tices to mitigate its effects in the design of future power plants.7 The author super-
vised the CS raw water piping sampling program and analysis conducted by W. S.
Bain of TVA that was initiated to determine the extent of the problem in TVA power
plants existing at that time.7,8 The extent of the buildup of corrosion products that
were found in the 50 samples that were removed from nine different TVA power
 100
TABLE 7.3

Engineering of Power Plant and Industrial Cooling Water Systems

Equivalent Lengths of Straight Pipe for Piping Components
Nominal pipe size, in 0.5 0.75 1 1.25 1.5 2 2.5 3 4 5 6 8 10 12 14 16 18 20 24 30 36 42 48
Gate valve 0.4 0.6 0.7 0.9 1.1 1.4 1.7 2 2.7 3.4 4 5.3 6.7 8 8.8 10 11.3 13 15 19 23 27 31
90o elbow 1.6 2.1 2.6 3.5 4 5.2 6.2 7.7 10 13 15 20 25 30 33 38 42.2 47 57 70 85 100 115
90o long radius elbow 0.8 1.1 1.4 1.8 2.2 2.8 3.3 4.1 5.4 6.7 8.1 11 13 16 18 20 22.5 25 30 37 45 53 61
45o std elbow 0.8 1.1 1.4 1.8 2.2 2.8 3.3 4.1 5.4 6.7 8.1 11 13 16 18 20 22.5 25 30 37 45 53 61
Std tee thru flow 1 1.4 1.8 2.3 2.7 3.5 4.1 5.1 6.7 8.4 10 13 17 20 22 25 28.1 31 38 47 57 67 77
Std tee branch flow 3.1 4.1 5.3 6.9 8.1 10 12 15 20 25 30 40 50 60 66 75 84.4 94 113 140 170 200 230
Close return bend 2.6 3.4 4.4 5.8 6.7 8.6 10 13 17 21 25 33 42 50 55 63 70.3 78 94 117 142 167 192
Swing check valve 5.2 6.9 8.7 12 13 17 21 26 34 42 51 33 42 50 55 63 70.3 78 94
Angle valve 7.8 10 13 17 20 26 31 38 50 63 76 100 125 149 164 188 210 235 283
Globe valve 18 23 39 39 46 59 70 87 114 143 172 226 284 338 372 425 478 533 641
Butterfly valve 7.8 9.3 12 15 19 23 30 29 35 38 31 35.2 39 47
90o welding elbow r/d=1 3.5 4.1 5.1 6.7 8.4 10 13 17 20 22 25 28.1 31 38 47 57 67 77
90o welding elbow r/d=2 2.1 2.5 3.1 4 5.1 6.1 8 10 12 13 15 16.9 19 28 28 34 40 46
Miter bend 45o 2.6 3.1 3.8 5 6.3 7.6 10 13 16 16 19 21.1 24 35 35 43 50 58
Miter bend 45o 10 12 15 20 25 30 40 50 66 66 75 84.4 94 14 14 17 20 23
Pipe Flow 101

plants as measured by the average reduction in pipe diameter is illustrated in Figure

6.4. Flow tests were performed at the following three fossil power plants: Widows
Creek, Kingston, and Gallatin. The objective of these flow tests was to evaluate the
effects of corrosion product buildup on pressure drop and to arrive at a means of
predicting the friction loss in CS cooling water piping. The sites were selected to
cover a range of age as well as a variety of water sources. All tests were made on
straight lengths of pipe to avoid consideration of bends. Tees were included in some
of the piping systems tested but the pressure drop across the tee was neglected, since
the run of the tee was always in line with the test flow and the lateral branch was
always closed. An orifice inserted in a length of new piping was installed in each of
the piping systems to measure flow rate. The orifice was installed adjacent to the sec-
tions of piping where pressure drop measurements were taken. Taps were installed in
the lines to allow pressure drop measurements to be made. Mercury manometers
were used to measure the pressure drops across the orifice and each section of piping.
Samples removed from each test line were analyzed to determine the percent volume
reduction of the pipe interior due to the corrosion product buildup.
Samples removed from the 3-inch (7.6 cm) test line at Widows Creek had a sub-
stantial amount of iron oxide and silicon oxide buildup, whereas the samples removed
from the 6-inch (15.2 cm) line at Kingston were found to have only a small amount
of uniform buildup but had very large randomly spaced tubercles [some approaching
2 inches (5.1 cm) in height]. The samples from Gallatin were found to have a more
uniform buildup than the Kingston 6-inch (15.2 cm) line but also had large, randomly
spaced tubercles. Table 7.4 shows the average diameter reduction of each set of
samples.
The corresponding diameter reduction for each test line was then used with the
pressure drop test data to develop appropriate equations for predicting pressure drop.
The Hazen-Williams and Darcy-Weisbach equations for pressure drop were both
considered, with the set of data being treated separately and then analyzed to estab-
lish a correlation to the other sets of data. Predictive equations were then formulated
to predict pressure drop in CS raw water piping after 40 years of service. The analysis
was based on the head loss per 100 feet of pipe so that Equation (7.4) becomes

hL , f  100   Q1.85 
 0.2083 L    4.8655  (7.7)
100  C  d 

TABLE 7.4
Diameter Reduction in Flow Test Samples
Average Measured
Plant Diameter Reduction (in)
Widows Creek 0.405
Kingston 0.133
Gallatin
Section A 0.270
Section B 0.320
102 Engineering of Power Plant and Industrial Cooling Water Systems

where
hL,f /100 = friction head loss in feet per 100 ft of pipe.
Equation (7.8) is a least squares curve fit of the form of Equation (7.7).

hL , f
= a1 Q1.85 (7.8)
100

where a1 is a constant obtained from each set of test data. By setting Equation (7.7)
equal to Equation (7.8), one may solve for d as shown in Equation (7.9).
1

 
  100 
1.85
 4.8655
d   0.2083  (7.9)
 a1  C  

For each set of data with a unique value of a1 a tabulation of values for d that satisfy
Equation (7.8) was created.
Per Figure 6.1, the fact that the flow through a CS cooling water pipe occluded by
corrosion products is a function of more than the roughness of the interior of the pipe
is obvious. Accordingly, a dimensionless parameter, d*, defined in Equation (7.10),
was found to be useful in correlating the values of d calculated from Equation (7.9)
with the measured values of diameter reduction. (See Section 6.2.1.)

d 
 dnom  dCALC 
(7.10)
d MEAS

where
dNOM = nominal inside diameter of new pipe
dCALCC = calculated inside diameter of pipe using Equation (7.9)
ΔdMEAS = diameter reduction corresponding to the percent volume reduction.
Figure 7.2 shows d* plotted as a function of “C” for all of the pressure drop tests.
The smallest variation of d* occurs at a value of “C” of approximately 57 at a
value of d* approximately equal to 2. Equation 7.11 shows the Hazen-Williams
equation modified by Bowman and Bain as discussed herein based on a slightly more
conservative value of “C” equal to 55 and a diameter reduction equal to twice that
measured.

hL , f 0.63 Q1.85
 (7.11)
100
 dNOM  2 d MEAS 
4.8655

The Darcy-Weisbach equation can be written (for a pipe length of 100 ft) in the form
shown in Equation (7.12).

hL , f Q2
= 3.11 f (7.12)
100 d5
Pipe Flow 103

8
Gallatin 8 in. Section 1
Gallatin 8 in. Section 2
Kingston 6 in.
Windows Creek 3 in.
7

d*, calculated diameter reduction/measured diameter reduction

6

0
20 40 60 C 80 100 120

FIGURE 7.2 d* as a function of Hazen-Williams “C” factor.

Equation (7.13) is a least squares curve fit of the form of Equation (7.12).

hL , f
= a2 Q 2 (7.13)
100

where a2 is a constant obtained from each set of test data.

Setting Equation 7.12 equal to Equation 7.13 and solving for the friction factor f
yields Equation (7.14).

a2 d 5
f = (7.14)
3.11

Equation (7.15) is the expression for the friction factor in fully rough flow by Moody.9
Full rough flow is almost certain to exist at design flow in old, corroded piping.

1 3.7
= 2 log10 (7.15)
f e
d

where
“e” is in inches.
104 Engineering of Power Plant and Industrial Cooling Water Systems

10
Widows Creek 3 in.
Gallatin 8 in. Section 2
Gallatin 8 in. Section 1
Kingston 6 in.

, in. (based on reduced diameter

0.1

0.01
0.0 0.5 1.0 2.0 1.5 2.5 3.0
d*, calculated diam. reduction/measured diam. reduction

FIGURE 7.3 d* Darcy-Weisbach “e” factor as a function of d*.

Equation (7.16) is Equation (7.15) rearranged to solve for e/d.

e 3.7
 (7.16)
d 1
10 
2 f

For each pressure drop test with its unique value of a2, values of f were calculated for
different assumed d using Equation (7.14). Equation (7.16) was used to calculate e/d.
Figure 7.3 shows a plot of “e” vs d*.
The smallest variation in “e” appears to occur at a value of d* equal to 1.0 to 1.2
where e = 0.9 to 0.8 inches.
Equation (7.17) shows the Darcy-Weisbach equation as modified by Bowman and
Bain using a value of e = 0.9 inch and a calculated diameter reduction equal to the
measured value of diameter reduction (d* =1).

hL , f 3.11 f Q 2
 (7.17)
100
 dNOM  d MEAS 
5

where
2
f  2 log10  4.1  d NOM  d MEAS   (7.18)

As one may see from Table 7.1, highly regarded sources agree on the adoption of a
Hazen-Williams “C” of 100 as the commonly used value for the design of CS piping
Pipe Flow 105

FIGURE 7.4 Comparison of measured and predicted pressure drops for 6-inch pipe at
Kingston Fossil Power Plant.

FIGURE 7.5 Comparison of measured and predicted pressure drops for 3-inch pipe at
Widows Creek Fossil Power Plant.

in cooling water systems. Such was the case at TVA until it was proven otherwise.
Figures 7.4 through 7.6 show the raw data taken for various pressure drop tests along
with curves representing different methods of calculating pressure drop using the
Bowman–Bain equation.
These curves based upon the empirical evidence contained herein, illustrate the
inadequacy of an assumption of C = 100 for design purposes. The Bowman-Bain
Equations (7.11) and (7.17), account for not only the roughness of the inside surface
of the pipe but also the diameter reduction to predict pressure drop in piping.
At reasonable velocities, the Hazen-Williams equation as modified by Bowman
and Bain consistently predicts values of pressure drop greater than the modified
Darcy equation, perhaps because the large diameter reduction projected after 40 years
is doubled in Equation (7.11) but not doubled in Equation (7.17). As indicated in
106 Engineering of Power Plant and Industrial Cooling Water Systems

FIGURE 7.6 Comparison of measured and predicted pressure drops for 8-inch pipe at
Gallatin Fossil Power.

Figures 7.2 through 7.4, the resulting head loss comparison is based on C = 55 and
Δd = 0.4 inch (1.02 cm), since those are the values that resulted in a good correlation
for the three flow tests performed. Results may vary for other sites.
Like many power and industrial plants located on the interior, the Tennessee River
enjoys relatively soft water having an average pH of 7.3. For those plant sites that
have similar conditions, the results of this study should have a profound impact on
the design of future raw water piping systems, since the pressure drops calculated by
methods recommended herein are significantly greater than those which would be
calculated by presently recognized standard methods. It should be noted that these
results apply only to friction losses and should not be applied to form losses.
In 1986, the results of the TVA flow and pressure drop tests were presented at an
EPRI workshop on MIC.10 The presentation gained little notice.

7.4 FLOW AND PRESSURE DROP TESTS OF LARGE DIAMETER

PIPING
The system friction losses from one point (l) to another point (2) in the system may
be arrived at by first calculating the form losses in the system and subtracting the
form losses from the dynamic head loss calculated from Equation (7.19).

hL   P1  P2    Z1  Z 2   VH1  VH 2  (7.19)

where
hL = total head loss
P1 and P2 = gauge pressure
Z1 and Z2 = elevations of the gauge pressure
VH1 and VH2 = velocity head.
Pipe Flow 107

The velocity head is calculated from Equation (7.20).

2
Vpipe
VH = (7.20)
2g

where
Vpipe = velocity in the pipe
Form losses, hL,F, are calculated based on the equivalent lengths and subtracted
from the total head loss to yield the friction head loss, hL,f . (See Section 7.1.)
At TVA’s SNP ERCW system flow tests were conducted annually from 1980 to
1993.11 These test data yielded information to calculate the apparent “C” factor in
two large headers between the IPS and the auxiliary building. Each of the two 4000-
foot (1,220 m) runs of piping includes 24-inch (61.0 cm), 30-inch (76.2 cm), and
36-inch (91.4 cm) diameter sections of piping in series with varying flow rates in the
different sections of piping.
For a run of piping with multiple line sizes in series, the Hazen-Williams equation
as modified by Bowman and Bain may be expressed as Equation (7.21).

1.85 s    L 
 100  Qn1.85
hL , f
100
 0.2083 
 C 
    d  2 d MEAS 
4.8655
  n   (7.21)
  100  

n 1  NOM , n


where
s = number of pipe sections
Qn = flow in each section of pipe, gal/min
dNOM, n = nominal pipe diameter in each section of pipe, in
L = length of pipe in each section, ft.
Solving for “C” yields

 1.85
 s    L  1
Qn1.85
C
 0.2083 x 100 

 hL , f 

 
  d NOM ,n  2 d MEAS 

4.8655 
n 
  100  
1.85
(7.22)
 100 
1   

At TVA’s Sequoyah plant, flow tests were conducted on the ERCW system by mea-
suring the cooling water pressures at the IPS and at the auxiliary building. The
Sequoyah test data have yielded sufficient information to calculate the apparent “C”
factor in the two large headers between the IPS and the auxiliary building. Each of
the two runs of piping has varying flow rates in the different sections of piping.
Since it is impractical to remove a section of large-diameter piping to take a mea-
surement, a value of ΔdMEAS = 0.2 in. (0.51 cm) was assumed based on Figure 6.4 and
the age of the piping. However, the “C” factor calculated from Equation (7.22) is
relatively insensitive to the value of ΔdMEAS for such large diameter pipe.
Figures 7.7 and 7.8 show the results of these tests for headers 2A and 2B,
respectively.
108 Engineering of Power Plant and Industrial Cooling Water Systems

100

Hazen-Williams "C" Factor

95

90

85

80

75

70
80 82 84 86 88 90 92 94
Test Year

FIGURE 7.7 Sequoyah Nuclear Plant ERCW header 2A Hazen-Williams “C” factor.

95
Hazen-Williams "C" Factor

90

85

80

75

70

65
80 82 84 86 88 90 92 94
Test Year

FIGURE 7.8 Sequoyah Nuclear Plant ERCW header 2B Hazen-Williams “C” factor.

The results are somewhat scattered even though these tests were performed with
calibrated instruments and the data was independently verified. However, this scatter
is to be expected, since the tests involved measuring pressures with separate pressure
gauges approximately one-half mile apart and measuring flows in 30-inch (76.2 cm)
piping with standard flow orifices.
The results are somewhat surprising, because the “C” factors do not appear to be
decreasing with age as expected. However, as discussed in Section 6.5, the piping
was first placed in service in 1977 and 1978. By 1982, the “C” factor had degraded
to approximately 90 for Header A and 70 for Header B when the biocide treatment
was implemented. The fact that Train A is normally operated, while Train B is not,
may be a consideration. (See Figure 6.13.)
Pipe Flow 109

TABLE 7.5
Watts Bar Nuclear Plant ERCW Header Hazen-Williams “C” Factors Before
and After Lining with Cement Mortar In Situ
“C” Factor “C” Factor
Supply Header for CS for CML
1A 74.6 163.7
2A 80.7 184.8
1B 77.1 219.5
2B 75.4 154.1
Average 77.0 181.5

In 1982, TV A conducted similar flow test on the supply headers at the WBNP
before and after the CML was applied as discussed in Section 5.4.5.12 The Hazen-
Williams “C” factor for the piping before and after the CML was applied is shown in
Table 7.5.
The increase in flow-passing capability of the piping is quite dramatic. The result
has been an increase header pressure in the ERCW system inside the auxiliary build-
ing. The average value for CML is even higher than the published Values of “C” =
130-140. (See Table 7.1.) The scatter in the results of the CML tests when compared
to the CS tests is due to the lower head loss measured relative to the test instrument
accuracy. These tests were conducted prior to completion and commercial operation
of WBNP, so each header was operated about the same amount of time and without
biocide treatment.
In 1983, a similar flow and head loss test was conducted on the BFNP EECW
system. The EECW system pressure was measured at the EECW pump discharge
and at the entrance to the emergency diesel generators. The flow to the diesel genera-
tor HXs was measured in the 14-inch (35.6 cm) and 18-inch (45.7 cm) CS pipes from
the pumps. The “C” factor calculated by the procedure described in this section was
C = 88 assuming Δd = 0.2 (0.51 cm) in.

7.5 MULTIFLOW PIPING ANALYSIS SOFTWARE

Realizing that the cooling water piping systems engineered prior to 1980 were inad-
equate because they were designed on the basis of Hazen-Williams C = 100, TVA
initiated a reanalysis of the safety-related cooling water systems in all of their nuclear
plants. Since this analysis was highly labor intensive, software (MULTIFLOW) was
developed for hydraulic analyses of raw water and condensate piping networks in
nuclear power plants by the TVA Norris Engineering Laboratory.13 MULTIFLOW
incorporates TVA’s established hydraulic design standards6 into an intuitive, menu-
based, computer code for balancing steady, incompressible flows and temperatures in
raw water, and condensate pipe networks of arbitrary complexity and provides output
reports.
A network is built by specifying links, nodes, tees, and boundary conditions using
a series of menus and dialog boxes. Extensive built-in tabular data based on TVA
110 Engineering of Power Plant and Industrial Cooling Water Systems

hydraulic design standards provide appropriate default values for internal diameters
and roughnesses of piping, equivalent lengths of various fittings, and loss coefficients
for combining and dividing flow through tees. Available inputs include global or
local specifications of corrosion effects on pipe roughness and inner diameter values.
The network input data are saved in text format to a disk file that may be edited later
using either the editing capabilities available in MULTIFLOW or any text editor.
After all necessary input data are provided, MULTIFLOW solves the system and
provides output reports to a printer or to the monitor.
MULTIFLOW was developed and tested in accordance with the requirements of
TVA Nuclear QA criteria for computer software. The requirements included
validation by comparing results obtained using MULTIFLOW with results obtained
by hand calculations for an extensive set of problems exercising all of the capabilities
of MULTIFLOW. In addition to its adherence to TVA QA criteria and mechanical
design standards, MULTIFLOW has several input, solution, and output capabilities
that are not usually available in general-purpose network codes. The size of a network
solved using MULTIFLOW is limited only by available computer memory.
MULTIFLOW iteratively computes flows through tees using separate loss coefficients
for each flow path that depend on the ratio of flow through the path to total flow
through the tee. MULTIFLOW computes fluid temperatures throughout a network
based on specified supply flow temperatures and specified loads at HXs. For
condensate systems, fluid densities, viscosities, and vapor pressures are treated as
variable with temperature and pressure. MULTIFLOW allows the flow in a link to be
fixed at a user-specified value and computes the valve flow coefficient or pump total
head required to achieve the desired flow. Components handled by MULTIFLOW
include pipes, pumps, check valves, pressure regulating valves, relief valves, tees,
fixed flows, and various sources of energy loss (e.g., valves, elbows, reducers, etc.).
MULTIFLOW also contains built-in design diagnostic tools that flag unacceptable
conditions, such as pressures below the vapor pressure.

7.6 PIPING NETWORK ANALYSIS USING MULTIFLOW

As discussed in Section 3.3 and 10.1, Section 316(b) of the CWA requires power and
industrial plants that draw water from the waters of the United States to implement
the BTA to reduce injury or death of fish and other aquatic life. The first alternative
proposed by the EPA to accomplish the stated goal was to install a closed-cycle recir-
culating heat rejection system.
In the 1970s to comply with the CWA, TVA constructed two NDCT at the SNP to
operate in a “helper” mode to cool the CCW prior to discharging it into the Tennessee
River. Provision was made in the modification to be able to operate in closed-cycle
recirculating mode by means of a channel back to the CCW pumping station.
Provision was also made to add a third NDCT if required.
For many years, compliance with Section 316(b) of the CWA was subject to liti-
gation. Final rules for compliance with this section were promulgated by the EPA in
2014. In 2017, TVA conducted a study at the SNP to determine the best method of
complying with Section 316(b) of the Act.14 Section 3.2.1.3 of Reference 14 describes
Alternative A, which is to operate the existing NDCT in closed-cycle recirculating
mode as follows:
Pipe Flow 111

• Alternative A – Refurbish the existing two NDCTs

  Alternative A would require the refurbishment of the existing two NDCTs
including the removal, disposal, and replacement of the existing fill material,
hangers, and drift eliminators. The fact that no additional heat dissipation capa-
bility is provided for Alternative A results in the maximum temperature enter-
ing the condenser and RCW System being as much as 10°F higher than is
presently the case and approximately 4°F higher than is the case for Alternatives
B and C. To compensate for the higher temperature of the cooling water enter-
ing the RCW System, the flow rate to the HXs served by the system must be
increased, and not only must much of the CS RCW piping be replaced with SS
which exhibits superior flow-passing capability, but also the existing RCW
Pumps would have to be replaced with higher head pumps. Overall, Alternative
A results in the most significant loss at 55 MWe per Unit of SQN’s generation
output or a reduction of 4.5% per unit per year.

The RCW system for each nuclear unit at SNP takes its suction from the CCW supply
conduit upstream of the MC. There are four 20” (50.8 cm) RCW strainers and five
horizontal, centrifugal, dry-pit RCW pumps. One of the RCW pumps is an installed
spare. The RCW system serves innumerable non-safety-related HXs in the SNP such
as the turbine lube oil cooler, stator water cooler, hydrogen cooler, and a host of smaller
HXs serviced by small CS piping. The RCW system discharges back into CCW dis-
charge conduit at various points downstream of the MC. MULTIFLOW was utilized to
determine which CS RCW piping would be required to be replaced with SS.
Measurements of the RCW system pressures were made at the RCW pump suc-
tion and discharge and at 15 points in the system. Table 7.6 show the results of
MULTIFLOW runs for an array of Hazen-Williams “C” factors and values of Δd.
The results of the test implied that the Hazen-Williams “C” factor and the diameter
reduction that best modeled the actual pressures in the SNP RCW system are C = 80
and Δd =0.15 inch (0.381 cm), with an average difference between the measured pres-
sure and the pressure calculated using MULTIFLOW of only 0.3 psi (2.1 kPa).

TABLE 7.6
Sequoyah Nuclear Plant RCW System Pressure Difference* For an Array of
Hazen-Williams “C” Factors and Δd
Hazen-Williams Average
“C” Factor Δd ΔP (psi)
100 0 7.6
90 0.3 −1.0
85 0.2 −1.8
85 0.3 −0.5
80 0.15 0.3
80 0.2 0.5
80 0.3 0.3
70 0.3 2.3

* ΔP = Measured pressure – calculated pressure w/MULTIFLOW.

112 Engineering of Power Plant and Industrial Cooling Water Systems

TABLE 7.7
Sequoyah Nuclear Plant RCW System Piping Replacement
Line size (in) Length (ft) Elbows Gate Globe Reducer Butterfly
0.75 44 8 8 4 1
1.00 994 44 15 13 17
1.50 229 20 5 3 2
2.00 1803 58 14 18 4
2.50 573 68 20 1 12
3.00 471 25 3 6 4
4.00 379 26 2 1 2
6.00 76 7 1
14.00 104 4

Table 7.7 shows which CS RCW piping would need to be replaced with SS. The
RCW Pumps would also need to be replaced with higher head pumps.

REFERENCES
1. Linsley, R. K. et al. Water-Resources Engineering, 4th ed., pp. 347–349, McGraw-Hill,
Inc., 1992.
2. Hansen, E. G. Hydronic System Design and Operation-A Guide to Heating and Cooling
with Water, p. 61, 63, McGraw-Hill, Inc., 1985.
3. Crane Technical Paper No. 410, 24th Printing, Crane Co., King of Prussia, PA, 1988.
4. Cameron Hydraulic Data, 16th ed., pp. 3–7, 3–8, 3–20, Ingersoll-Rand, Woodcliff Lake,
NJ, 1981.
5. Mays, L. W., Water Resource Engineering, 2nd ed., pp. 473, 489–492, John Wiley &
Sons, Inc., 2011.
6. Mechanical Design Standard DS-M3.5.1, Rev. 7, Pressure Drop Calculations for Raw
Water Piping and Fittings, Tennessee Valley Authority, Knoxville, TN, 1999.
7. Bain, W. S., et al. Corrosion in Carbon Steel Raw Water Piping, (TVA Ref.
ESS800416204) Tennessee Valley Authority, Knoxville, TN, 1979.
8. Bowman, C. F. and W. S. Bain. A New Look at Design of Raw Water Piping. Power
Engineering Magazine, pp. 73–77, Vol. 48, No. 8, August 1980.
9. Moody, L. F. Friction Factors for Pipe Flow. Transactions of the ASME, Vol. 66, No. 8,
pp. 671–684, 1966.
10. Bowman, C. F. Microbe – Induced Corrosion of TVA Raw Cooling Water Piping.
Proceedings of the EPRI Workshop on Microbe Induced Corrosion (MIC), Electric
Power Research Institute, 1986.
11. Bowman, C. F. Solving Raw Water Piping Corrosion Problems. Power Engineering
Magazine, Vol. 98, No. 7, pp. 35–38, July 1994.
12. Bowman, C. F. In Situ Cement-Mortar Lining of Safety-Related Service Water Piping
Systems. International Joint Power Conference, 94-JPGC-NE-6, 1994.
13. Schohl, G. A., et al. MULTIFLOW, a Quality Assured Intuitive Computer Code for
Hydraulic Analysis of Pipe Networks. Proceedings of the 1995 International Joint
Power Generation Conference, Vol. 2, pp. 67–79, 1995.
14. Sequoyah Nuclear Plant §316(b) – §122.21(r)(10) – (13) Information, Appendix A to
Transmittal from TVA to Tennessee Department of Environment and Conservation,
June 29, 2018.
 8 Cooling Water System
Engineering Pitfalls

8.1 HYDRAULIC PRESSURE GRADIENT

The hydraulic pressure gradient shown in Figure 8.1 illustrates the importance of
maintaining a positive gauge pressure throughout the cooling water system. The head
loss indicated by the hydraulic gradient is calculated as discussed in Chapter 7.
Consider a cooling water pump with pressure at the discharge of the pump suffi-
cient to pump the water 140 feet (42.7 m) above the lake level, which is at zero gauge
pressure. The hydraulic pressure gradient represents the head loss through the cool-
ing water system. At any point in the system, the gauge pressure is the difference
between the hydraulic gradient and the elevation of the piping or component in the
system where the pressure is measured in lbf/in2 (or kPag). In a cooling water system,
there are likely to be changes in elevation. To avoid serious problems, one should
ensure that the hydraulic gradient does not fall below zero at the highest elevation
points. This is especially true for a HX, which is often located relatively high in a
power or industrial plant as illustrated by Figure 8.2. If the gauge pressure drops at
or below zero lbf/in2 (kPag), depending on the temperature of the cooling water,
vaporization, and/or air liberation can occur as discussed below. One solution to this
problem is to increase the head of the cooling water pump and locate a throttling

150 Hydraulic pressure gradient

100
Head (Ft)

Gauge pressure

Cooling HX
Water
50 Pump

Lake level
Cooling
water piping

FIGURE 8.1 Hydraulic gradient with positive gauge pressure.

113
114 Engineering of Power Plant and Industrial Cooling Water Systems

150 Hydraulic pressure gradient

HX
100
Vacuum
Head (Ft)

Gauge pressure

Cooling
Water
50 Pump

Lake level
Cooling
water piping

FIGURE 8.2 Hydraulic gradient with negative gauge pressure.

Hydraulic pressure gradient

150

Gauge pressure HX
100
Head (Ft)

Cooling
Water
50 Pump

Valve

Lake level
Cooling
water piping

FIGURE 8.3 Hydraulic gradient with throttling.

Cooling Water System Engineering Pitfalls 115

valve downstream of the HX to provide a backpressure in the HX as shown in

Figure 8.3. Another solution would be to locate the HX lower in the plant.
The information in Sections 8.2 through 8.4 is largely taken from Reference 1 by
F. P. Carr.

8.2 AIR LIBERATION
Since the atmospheric gases (primarily nitrogen and oxygen) are soluble in water,
significant quantities of these gases can be absorbed and carried into the piping sys-
tem. These gases can also cause flow disturbances and reduced heat transfer rates in
HXs if they come out of solution inside the system.
Henry’s law relates the important variables affecting gas solubility as shown in
Equation (8.1).
Px
Nx = (8.1)
Kx
where
Nx = mole fraction of dissolved gas, x
Px = partial pressure of dissolved gas x
Kx = Henry’s Law coefficient for gas x.
This equation shows that the number of moles of gas that can be dissolved in a
given quantity of water is directly proportional to its partial pressure and inversely
proportional to its Henry’s law coefficient. For a particular gas, the latter coefficient
is primarily a function of temperature, increasing as the temperature increases. Thus,
water becomes less soluble to air as the pressure decreases or the temperature
increases. Since this is exactly what happens as cooling water passes through the
cooling water system, it is possible for gases to be released within the system.
Air liberation within the system can affect heat transfer of system HXs. For exam-
ple, if a HX were to operate with a sufficiently high temperature rise, or was located
so that the local pressure was sufficiently low, air liberation could occur. This would
not necessarily create any problems. However, if the geometry of the HX or the cool-
ing water piping were such that liberated air could be trapped within the HX, the air
could displace the water and effectively reduce the available heat transfer area. This
could occur in HXs with a bottom outlet or where the cooling water piping turned
down at the outlet nozzle.
Air liberation can also affect fluid flow if conditions are such that the flow is in the
so-called “slug” flow regime. In this regime, large bubbles of liberated gases are
periodically swept through the system, so the cooling water flow is not steady but
pulsing and unacceptable vibration can also result.
The preferred approach in initial design is to avoid air liberation entirely. The first
step to achieve this goal is to calculate the maximum quantity of air dissolved in the
cooling water inlet. Henry’s law may be used in the following combined form, which
includes the effect from all atmospheric gases as shown in Equation (8.2).

Pb
N air = (8.2)
K air
116 Engineering of Power Plant and Industrial Cooling Water Systems

where
Nair = mole fraction of all dissolved gases
Kair = Henry’s Law coefficient from Figure 8.4.
The barometric pressure should be the standard atmospheric pressure correspond-
ing to the elevation of the free surface water level. To maximize Nair, the Henry’s law
coefficient, Kair, should be based on minimum cooling water supply temperature.
The final factor, C, accounts for the degree of saturation of the inlet water. Since
it is common for bodies of water to approach 100 percent saturation in the winter,
“C” should not be less than 1.0. However, it is also known that the water can be
supersaturated with air downstream of dams or where the source of cooling water is
an evaporative cooling device such as a cooling tower or spray pond. (See Chapter 10.)
Evaporative cooling devices may therefore justify a “C” value of 1.5, whereas a cool-
ing lake may dictate a value of 1.0. Hydroelectric turbine operation can cause satura-
tion up to 100 percent, while spillways and evaporative cooling devices can cause an
increase to 150 percent. Having determined the mole fraction of dissolved air in the
inlet cooling water, the next step is to determine the minimum pressure required to
keep this air in solution. Again, use Henry’s law as shown in Equation (8.3).

N air max
Prequired  ’
(8.3)
K air

12

10
Henry's Law Coefficient, K AIR (x104)

0
0 10 20 30 40 50 60 70 80

Tem perature [°C]

FIGURE 8.4 Henry’s law coefficient for air.

Cooling Water System Engineering Pitfalls 117

where
Prequired = minimum gauge pressure to keep air in solution
Nair-max = maximum mole fraction of air in the entering cooling water
K’air = Henry’s Law coefficient based on the HX outlet temperature from Figure 8.4.
The key difference is that K'air is based on the HX outlet temperature. The outlet
temperature may be calculated using Equation (8.4).

Q
Tout  Tin  (8.4)
m cp

where
Tout = HX outlet temperature
Tin = HX inlet temperature
Q = maximum HX heat load
m = cooling water mass flow rate
cp = constant pressure specific heat.
Since Tout is greater than Tin, K’air in Equation (8.3) will be greater than Kair in
Equation (8.2) as evident from Figure 8.4. Tout must be calculated for each HX and
Prequired be determined for each different value.
The next step is to determine the actual minimum gauge pressure, Pactual, on the
discharge side of each HX. This involves simply calculating the hydraulic gradient as
discussed in Section 8.1. If “Pactual” is less than “Prerquired”, the potential exists for air
to be liberated within the HX and/or downstream piping due to the increase in
temperature. The fact that air is released is not necessarily unacceptable. The flow
regime must be determined to make the evaluation of those having the potential for
air liberation by first calculating the Froude number with Equation (8.5) and then
consulting Figure 8.5.

V2
Fr = (8.5)
gD

where
Fr = Froude number
V = local velocity
g = acceleration due to gravity
D = pipe or tube inside diameter.
Next one must calculate the air-to-water volume ratio, Vr, of air liberated at the
pipe or tube section being investigated. To determine this ratio, one must first find the
actual mole fraction of air released. This will be the difference between the mole
fraction of dissolved air in the inlet water, Nair, and the mole fraction that local condi-
tions will permit to remain in solution, “N’air” shown in Equation (8.6)

P’
’
N air = ’
(8.6)
Pb K air
118 Engineering of Power Plant and Industrial Cooling Water Systems

FIGURE 8.5 Bubbly/slug flow regime map.

where
N’air = mole fraction that local conditions will permit to remain in solution
P’ = local static pressure
K’air = Henry’s Law coefficient at local temperature (from Figure 8.4).
The volume of air liberated per unit volume of water, Ve, is then calculated by first
considering the ideal gas law shown in Equation (8.7) and (8.8).

PV = nRT (8.7)

nRT
V = (8.8)
P

where
P = pressure
V = volume
n = number of moles
R = universal gas constant
T = absolute temperature
Cooling Water System Engineering Pitfalls 119

The volume of air liberated per unit volume of water, Vr, may then be calculated
by Equation (8.9).

Vr 
N air 
 N air
’
RT ’
(8.9)
P’

where
Vr = air-to-water volume ratio
T’ = absolute local temperature.
The values of “Fr” and “V” are then plotted on Figure 8.5. Those that fall within
the bubbly flow regime are generally acceptable. Those in the slug flow region may
require modification.
Many possibilities exist to correct HX loops found to be in the slug flow regime.
However, many of the solutions become increasingly difficult as design and
construction progress. Perhaps the simplest solution is to relocate the flow-balancing
valve beyond the region having the potential for air liberation as shown in Figure 8.3.
This will pressurize the critical area by the amount of the differential across the
valve, possibly eliminating the potential for air release or transforming it to bubbly
flow. Another possibility is to reroute the piping to eliminate problem high points. It
may even be possible to relocate a HX to a lower floor as shown in Figure 8.1.
Booster pumps to pressurize problem areas are also effective, particularly on the
supply side of the HXs.

8.3 FLASHING
The term “flashing” refers to the change of state from liquid to gas. It is generally
associated with the formation of large volumes of vapor as a result of heating, expan-
sion into a low pressure region, or both. Flashing might occur, for example, within a
HX, because of the temperature rise of the water. The effective heat transfer area
would thereby be reduced, leading to a further rise in temperature, more flashing, etc.
Eventually the HX could become completely filled with vapor and unable to remove
its design heat load.
Flashing can also occur in regions of low pressure. Such regions might exist
downstream of flow-balancing valves or automatic temperature control valves. For
this reason, such valves should generally be located downstream of their associated
HXs to preclude the possibility of collecting vapor within the HX.
The first step in evaluating the potential for flashing is to determine the HX outlet
temperature as shown in Equation (8.4). The saturation vapor pressure, Psat, corre-
sponding to “Tout” may then be read from the ASME Steam Tables.2,3 Next, the local
HX outlet pressure is calculated as described in Section 8.1. If Plocal < Psat, as illus-
trated in Section 8.2, flashing will occur. The outlet of flow-balancing valves or tem-
perature control valves should also be checked for flashing. Flashing should not be
allowed to occur within system HXs. Possible corrective measures include all of
those mentioned in Section 8.2 for air liberation. It might also be possible to prevent
flashing by increasing the design flow rate of the cooling water, thereby decreasing
120 Engineering of Power Plant and Industrial Cooling Water Systems

the temperature rise. HX performance should be evaluated when determining the

feasibility of the latter option.
Flashing downstream of throttling valves may lead to unacceptable levels of
cavitation, vibration, or surging. Criteria for evaluating this problem are presented in
the next section. If found to be unacceptable, and if the local temperature rather than
local pressure is the main problem, bypassing a portion of the cooling water inlet
flow around the HX to mix with valve inlet flow might effectively prevent flashing.

8.4 CAVITATION
Whereas flashing deals with the formation of vapor, cavitation deals with the damag-
ing effects as the vapor bubbles change back to liquid. The implosion of the vapor
cavities can generate extremely high local pressures. The shock wave itself is suffi-
cient to fracture the grain boundaries of metal surfaces, causing removal of whole
grains and leaving a roughened metal surface. Erosion may lead to structural failure
or leakage. Corrosion accelerates the process.
Cavitation is generally associated with throttling valves, orifice plates, venturis,
and other such devices that create a sudden change in local pressure. The propensity
for and severity of cavitation may be determined by calculating the “cavitation index”
shown in Equation (8.10).

Pd  Psat
’
Ki  (8.10)
Pu  Pd

where
Ki = cavitation index
Pu = local upstream pressure
Pd = local downstream pressure
Psat = saturation vapor pressure at the maximum operating temperature.
The acceptable cavitation level or index, “Ki”, so calculated may then be inter-
preted from Figure 8.6. If unacceptable cavitation is indicated by this procedure, the
corrective actions in Sections 8.2 for air liberation and/or 8.3 for flashing may be
effective. Generally, the low outlet pressure is the major contributor to the cavitation
problem, and steps to increase it will be effective. Adding a multihole restricting
orifice downstream of the cavitating component will increase the backpressure of the
component and reduce or eliminate the cavitation. However, the effect on system
performance at off-design points and over the plant life must be carefully considered.
If other measures are not feasible, the use of a special anti-cavitation valve or multi-
ple orifices might be considered. Initial piping design should attempt to minimize the
required throttling in the high-flow HX loops.
If the cooling water system design uses vertical pumps, provisions should be
made for air and vacuum, release between the pump discharge and the discharge
check valve. (See Figure 3.1.) The vacuum relief valve is required when the difference
in pump discharge elevation and the design low water level is greater than
approximately 33 feet (10.1 m) to prevent column separation when a pump is shut
Cooling Water System Engineering Pitfalls 121

FIGURE 8.6 Cavitation indices for orifices (from Reference 4).

down. When the pump is restarted, the air in the piping must be vented to prevent air
from being forced into the cooling water system, which could degrade the perfor-
mance of HXs served by the system.

8.5 WATER HAMMER
The information in this section is largely taken from Reference 5 by G. C. Dunn and
J. D. Hubble.

8.5.1 Concepts of Water Hammer

Water hammer is a hydraulic pressure transient caused by changes in fluid velocity
and the resulting conversion of fluid kinetic energy into pressure energy, which is
propagated as a pressure wave. Analysis and control of water hammer requires some
understanding of this pressure wave propagation. The fundamental equations
involved are presented and discussed in the remainder of this section.
The equations of continuity and momentum may be applied to obtain the magni-
tude of change in fluid pressure as a function of the magnitude of change in fluid
velocity.6 The instantaneous pressure increase is, in general, related to the velocity
decrease by Equation (8.11).

A
H  Ho 
g
Vo  V  (8.11)
122 Engineering of Power Plant and Industrial Cooling Water Systems

where
H = pressure head
Ho = initial pressure head
A = wave speed
g = gravitational constant
V = fluid velocity
Vo = initial fluid velocity.
Equation (8.11) applies only for a simple wave propagation when the deceleration
in fluid velocity from “Vo” to “V” occurs over a period of time, T, where t < 2L/A,
where “L” is the length of pipe and where the wave speed, A, is much greater than
the initial fluid velocity, Vo. The pressure wave thus generated travels through the
pipe. The travel time required for the return of this wave is defined in Equation (8.12).

2L
T = (8.12)
A

An expression for wave speed in an infinite reservoir in English units as derived in

Reference 6 is shown in Equation (8.13)

Ew
A  (8.13)
w

where
Ew = bulk modulus of elasticity of water
ρw = density of water.
The wave speed in a pipe depends on the material, wall thickness, and type of
construction as well as the fluid properties and temperature. Equation (8.14) (in
English units) shows how to compute the wave speed as a function of the particular
piping system.7

Ew 1
A  x (8.14)
 E D
1  w 
 Ep t 

where
Ep = bulk modulus of elasticity of the pipe material
D = pipe diameter
t = thickness of the pipe wall.
Wave speeds for large steel pipe may be as low as 3,000 ft/sec (914 m/s) for small,
high-pressure steel pipe or as low as 4,500 ft/sec (1,372 m/s) in the absence of
ingested air.
To calculate the wave speed in a homogeneous mixture of air and water, Equations
(8.15) and (8.16) are used along with Equation (8.14).
Cooling Water System Engineering Pitfalls 123

Ew
EM  (8.15)
VM a  Ew 
1   1
VMT  Ea 

 VM a   VM w 
M  a    w  VM  (8.16)
 VMT   T 

where
EM = bulk modulus of elasticity of the mixture of water and air
Ew = bulk modulus of elasticity of water
Ea = bulk modulus of elasticity of air
VMw = volume of water
VMa = volume of air
VMT = total volume
ρM = density of the mixture of water and air
ρw = density of water
ρa = density of air.
Measurements of wave speeds as low as 1000 ft/sec (305 m/s) with 0.4 percent air
(by volume) in water have been published which show good agreement with these
equations.
Therefore, the assumption that the wave speed, A, is much greater than the initial
fluid velocity, Vo, in Equation (8.11) is valid.
When a change in fluid velocity occurs over a time interval greater than the travel
time of the resulting pressure wave as expressed in Equation (8.12), the change is
considered slow, and the governing equations in a frictionless system are shown in
Equations (8.17) and (8.18).8

H  Ho  F  f (8.17)

g
V  Vo  
A
F  f  (8.18)

where
F = the magnitude of the pressure wave generated by a change in water velocity
and whose direction is opposite to original water flow
f = the magnitude of the reflected pressure wave generated by a change in water
velocity and whose direction is that of the original water flow.
where “F” is a function of t-x/A and f is a function of t+x/A and “x” is the distance
along the pipe.
Note that “x” is positive in the direction opposite the initial fluid velocity, since it
measures the distance upstream from the point of initial disturbance in flow.
Equation (8.17) implies that at a time, t, at a point “x” feet from the initial distur-
bance, the head rise is equal to the algebraic sum of the pressure wave, F, and the
reflected wave, f. These waves are propagated in opposite directions in the pipe with
124 Engineering of Power Plant and Industrial Cooling Water Systems

a constant velocity of “A” feet per second. When an “F” wave passes an “f” wave,
neither wave is attenuated or undergoes a change in shape.
Equation (8.18) relates the magnitude of the pressure waves “F” and “f” to the
change in velocity which propagates the waves.
Water hammer is generally caused by changes in the mode of equipment operation.
This, in turn, is most often the result of valve operation, startup, shutdown, or
inadequate venting actions.

8.5.2 Control and Analysis of Water Hammer

Rapid valve operation will cause water hammer as described by Equation (8.11). If
valve operation time is less than “T” as defined in Equation (8.12), the generated
pressure waves will reflect against the closed valve.
By solving Equation (8.17) for “F” and substituting into Equation (8.18), an
expression for the change in pressure head resulting from these conditions may be
written as shown in Equation (8.19).

A
H  Ho  
g
V  Vo   2 f (8.19)

Initially, “f” equals zero at the valve. It may be shown that for a wave reflected from
a reservoir, the first reflected wave at the valve is equal in magnitude but opposite in
sign to the direct wave which left the valve “T” seconds earlier.6
For slow valve operation, the severity of potential water hammer may be deter-
mined by approximating the actual valve closure by a series of successive instanta-
neous stepwise movements at some time intervals, Δt, where Δt < T.
In order to determine the changes in velocity for a chosen time interval, a
relationship is required between the percent valve opening with respect to time
(during valve operation) and the valve coefficient as a function of valve opening.
With this information, the change in flow and corresponding change in velocity may
be calculated for the specified time interval. Equation (8.19) is then used to calculate
H – HO for the selected interval. For each stepwise movement, the initial values of
“H” and “V” are taken as the final value calculated from the previous step. This
procedure is repeated until the valve reaches its final position. Given a specified
design, this procedure allows the effects of valve operation to be evaluated.
Long pipe lengths will require relatively long valve closure times in order to pre-
vent water hammer. Equation (8.2) reveals that this factor is usually negligible unless
there are several thousand feet of pipe. However, consideration must be given to the
transient resulting from an instantaneous valve closure due to a motor operator fail-
ure. If such a failure would result in an unacceptable impact on the system, snubbers
(dampers) should be furnished with the valve operator.
Unless properly designed, a cooling water system is likely to experience water
hammer during startup and shut down due to the acceleration and deceleration of
fluid columns inherent in these operations. By proper design and operation, water
hammer can be avoided by controlling the rate of change of fluid column velocity.
Cooling Water System Engineering Pitfalls 125

During a cooling water pump start, the acceleration of the fluid column may be
controlled by starting the pump against a closed or partially closed discharge valve
and then gradually opening the valve. The reverse procedure may be followed when
shutting down a pump. Spurious pump trips can result in column separation in the
pump discharge, with resulting water hammer. The vacuum generated gradually
overcomes the receding water column with a spring-like action, and then collapses as
the column of water is reunited. A vacuum breaker/air relief valve may be used on the
pump discharge to prevent formation of the temporary vacuum.9 Check valves or
automatic-closing butterfly valves located on pump discharges may also aid in
reducing the detrimental effects of column separation at high points in the system by
keeping the upstream leg of piping full of water following a pump trip.
High points in the system which would be pressurized under normal operating pres-
sures, will experience depressurization and column separation upon system shutdown
if the high point is more than 33 feet (10.1 m) above the hydraulic gradient with the
system depressurized. Upon startup, as the system is re-pressurized, the vapor column
will collapse. This results in a water hammer as the column of water is reunited.
Vacuum breaker/air relief valves should be provided at all such high points to prevent
water hammer as the column is reunited. Carefully located check valves in the system
will often provide protection against column separation at high points in the system by
keeping the upstream leg of piping full of water following a system shutdown.
During startup, the system should be filled carefully to vent all air pockets at high
points so that flow will not be impeded. Care must be exercised in the sizing of vent
valves, since the size of the vent valve establishes the velocity at which the column
of water rises at the high point as the air is displaced by water. If the vent valve is too
large, resulting in the water column rising at a relatively high velocity, the resulting
deceleration as the last of the air is released can be very significant. This problem can
be particularly serious in large tanks and HXs. An alternative to careful sizing of the
vent valve would be to place an orifice in the vent line sized to restrict the rate at
which the air may be vented.
It should be emphasized that the complexity and potential consequences of water
hammer lead to design as the preferred method for control of hydraulic transients.
However, a detailed water hammer analysis may be required if the system design
parameters violate one or more of the guidelines described herein.

REFERENCES
1. TVA Mechanical Design Guide DG-M6.3.3. General Design of Essential Raw Cooling
Water Systems, Tennessee Valley Authority, Knoxville, TN, 1986.
2. Meyer, C. A. et al. ASME Steam Tables, 6th ed., American Society of Mechanical
Engineers, New York, NY, 1993.
3. ASME Steam Tables Compact Edition, American Society of Mechanical Engineers,
New York, NY, 1993, 2006.
4. Tung, P. and M. Mikasinovic. Eliminating Cavitation from Pressure-Reducing Orifices.
Chemical Engineering, December 12, 1983.
5. TVA Mechanical Design Guide DG-M3.5.3. Analysis and Control of Water Hammer in
Large-Diameter Raw Water Piping Systems. Tennessee Valley Authority, Knoxville, TN,
1982.
126 Engineering of Power Plant and Industrial Cooling Water Systems

6. Streeter, V. and E. B. Wylie. Hydraulic Transients, McGraw-Hill, Inc., Newe York, NY,
1967.
7. Linsley, R. K. et al. Water-Resources Engineering, 4th ed., p. 367, McGraw-Hill, Inc.,
1992.
8. Paramakian, J., Water Hammer Analysis, Prentice-Hall, Inc, New York, NY, 1955.
9. TVA Mechanical Standard Drawing SD-M6.1.1, Typical System Air Release
Requirements, Tennessee Valley Authority, Knoxville, TN, 1981.
 9 Heat Exchangers

9.1 SHELL-AND-TUBE HEAT EXCHANGERS

The detailed design of the HX is normally left to the HX manufacturer. However, the
engineer is inevitably faced with ensuring that the HX will meet the needs of the plant
over a wide range of operating conditions. Therefore, it is incumbent on the engineer
once the HX is procured to be able to predict the performance of the HX over the
range of anticipated conditions.
There are many types of HXs in power plants and industrial plants such as con-
densers, feedwater heaters, etc. The most common type of HX served by the cooling
water system is the shell-and-tube heat exchanger (STHX) shown in Figure 9.1 con-
sisting of tubes rolled or welded into tube sheets on both ends of the tubes. Cooling
water enters the tubes through the inlet header, passes through the tubes where it
increases in temperature, and passes out of the HX through the outlet header. The
fluid to be cooled enters the space on the outside of the tubes (referred to as the shell)
where it is directed by baffles to passes over the tubes several times to increase the
heat transfer before exiting the shell.
There are many different configurations of STHX. The one illustrated in Figure 9.1
is a simple counter-flow HX in which both the cooling water and the fluid to be
cooled pass through the HX only once. Other configurations include three and four
passes on the tube side and two passes on the shell side with a divider plate in the
middle of the shell.
From the first law of thermodynamics,

mh hh in  mc hc in  mh hh  out  mc hc  out (9.1)

Q  mh  hh in  hh  out   mc  hc  out  hc in  (9.2)

where
hc-in = cold stream enthalpy in
hc-out = cold stream enthalpy out
hh-in = hot stream enthalpy in
hh-out = hot stream enthalpy out
mc = mass flow rate of cold stream
mh = mass flow rate of hot stream
Q = rate of heat transfer.
Since

h  cp T (9.3)

127
128 Engineering of Power Plant and Industrial Cooling Water Systems

FIGURE 9.1 Shell-and-tube heat exchanger. (Courtesy of Thomas).

then

Q  mh c p,h  Th in  Th  out   mc c p , c  tc  out  tc in  (9.4)

where
tc-in = cold stream temperature in
tc-out = cold stream temperature out
Th-in = hot stream temperature in
Th-out = hot stream temperature out
cp,h = hot stream specific heat
cp,c = cold stream specific heat.
Any one of the six parameters may be calculated if the other five are known. For
example,

mc c p  c  tc  out  tc in 
mh  (9.5)
c p  h  Th in  Th  out 

A commonly used method of HX analysis to determine the rate of heat transfer

through a HX is the log-mean temperature difference method defined in Equation
(9.6).

Q = U Ah F LMTD (9.6)

where
U = overall heat transfer coefficient
Ah = effective hot-side HX surface area
F = LMTD correction factor
LMTD = log mean temperature difference
Heat Exchangers 129

where

t1  t2
LMTD  (9.7)
 t 
Ln  1 
 t2 

and

t1  Th,i  t c,o (9.8)

t2  Th,o  t c,i (9.9)

The accepted practice is to take the hot-side (shell-side) effective surface area, Ah, as
the HX reference surface area where

Ah  N tubes N pass  do leff (9.10)

where
Ntubes = number of tubes per pass
Npass = number of passes
do = outside tube diameter
leff = effective tube length
The LMTD correction factor, F, is a function of the type of HX. The value of “F”
equal to one applies for counter-flow arrangements such as seen in Figure 9.1 and in
other counter-flow HX arrangements. (See Section 9.2 and 9.3.) A formula for the
value of “F” is available for a few other flow arrangements in the open literature.1
Another method of HX analysis is the effectiveness method as defined with refer-
ence to the cold stream as

Q  mc c p  c  tc  out  tc in   mc c p  c P  Th in  tc in  (9.11)

tc  out  tc in
P  (9.12)
Th in  tc in

where effectiveness, P, as defined in the equation above is the cooling range divided
by the cooling potential.
A formula for the value of “P” is available for other flow arrangements in the open
literature.2
The overall heat transfer coefficient, U, may be expressed as the inverse of the
sum of the resistances to heat transfer through the following:

• exterior convection layer, rh

• exterior fouling, rf,h
• tube wall, rw
130 Engineering of Power Plant and Industrial Cooling Water Systems

• interior fouling, rf,c

• interior convection layer, rc

1 A  A  A 
 rh  rf ,h   h  rw   h  rf ,c   h  rc (9.13)
U A
 w A
 c  Ac 

where

1 1
=rh = rc (9.14, 9.15)
hh hc

so that
1
U  (9.16)
1  Ah  A  A  1
 rf , h    rw   h  rf ,c   h 
hh  Aw   Ac   Ac  hc
where
hh = hot-side convection coefficient
hc = cold-side convection coefficient.
Note that all fouling resistances are referenced to the hot side (i.e. the shell side)
of the HX tubes. The effective area of the wall and the wall resistance are shown as
follows:
Ao  Ai
Aw  (9.17)
A 
ln  o 
 Ai 

do  di
rw  (9.18)
2 kt

where do and di are the tube outside and inside diameters and kt is the tube material
thermal conductivity.
The value of the cold-side convection coefficient, hc, is a function of the Nusselt
number, Nu, as follows:

k 
hc  Nu  c  (9.19)
 di 

where
kc = thermal conductivity of water.
Several equations are available for the value of the Nusselt number.1 The com-
monly used classic Colburn3 equation is as follows:

1
Nu = 0.023 Re 0t .8 Prt 3 (9.20)
Heat Exchangers 131

where
Ret = tube-side Reynolds number,

Re t  vt d i /  t (9.21)

and
Prt = tube-side Prandtl number,

Prt  c p,t t / k t (9.22)

where
ρ = density
μt = average tube viscosity
vt = tube velocity.
No such similar direct method for calculating the hot-side convection coefficient,
hh, exists, as it is a complicated function of the HX shell geometry such as the type
and spacing of the baffles, the leakage through the tube support plates, etc.
Manufacturers generally hold this data to be proprietary. If a manufacturer’s HX data
sheet is available, the most direct method for calculating hh is as follows:

1
hh, design  (9.23)
1  Ah  A  A  1
 rf , h    rw   h  rf ,c   h 
U design  Aw   Ac   Ac  hc

where the value of “Udesign” is available or may be calculated from the data sheet.
If the design flow rate, inlet and outlet temperatures, number and size of tubes,
and the value of “F” are known, then “Udesign” may be calculated from Equation
(9.24).

Q
U design = (9.24)
Ah F LMTD

For HXs in cooling water systems, the value “hc” may vary considerably during
operation, as the cooling water flow and temperature often change seasonally as the
inlet water temperature changes. The value of “hh” may vary slightly from the design
value during operation, but the change is normally minor compared with the other
resistances in Equation (9.13), and in particular the tube-side fouling term that often
predominates. Therefore, it is often acceptable to assume that the value for “hh” and
thus “rh” remains constant in many cooling water system applications. However, if
that is not the case, consult Reference 2.

9.2 AIR-TO-WATER HEAT EXCHANGERS

An air-to-water heat exchanger (AWHX) transfers heat between warm air and cool-
ing water through an array of finned tubes as shown in Figure 9.2. AWHX are widely
132 Engineering of Power Plant and Industrial Cooling Water Systems

used to cool spaces inside power plants and other industrial plants and for many other
applications such as to control the impact of an accident in a nuclear plant. A familiar
example of an AWHX is the radiator in an automobile, and like that application, it is
typically attached to a fan which draws air through a series of serpentine coils. As
such it is classified as a cross-flow HX in which the LMTD correction factor “F” is
equal to 1 if there are 4 or more passes through the coils as shown in Figure 9.2.
The convection boundary resistance to heat transfer on the air side of an AWHX
can be as much as 20 to 30 times that on the tube side where there is turbulent flow
of water through the tubes. To compensate for the poor heat transfer on the air side,
fins are provided to increase the surface area on the air side by as much as 20 or more
times that of the water side. As a practical manufacturing consideration, these fins are
normally rectangular in shape. Accordingly, due to the resistance to conduction heat
transfer down the fin to the surface of the water, increasing the length of the fin
reaches a point of diminishing returns.2 The measure of this reduction in heat transfer
is referred to as the fin efficiency, ηfin. Some AWHX employ helical fins that are
brazed to the tubes while others employ plate fins that are typically analyzed as
equivalent fins.
The overall heat transfer coefficient at design conditions, Udesign, is calculated
similar to Equation (9.16) for STHXs as shown in Equation (9.25).

1
 fin
U  (9.25)
1 A  A  A  1
 rf , h   h  rw   h  rf ,c   h 
hh  Aw   Ac   Ac  hc

Note that the only difference between Equation (9.19) and (9.25) is that the numera-
tor contains the term for fin efficiency. In most AWHX designs, the fin efficiency is
very nearly 1.0. If a more accurate analysis is required, consult Reference 2.

FIGURE 9.2 Air-to-water heat exchanger. (Courtesy of ChillX Chillers).

Heat Exchangers 133

The tube-side convection coefficient may be computed as in Equation (9.19).

As was the case with STHXs and as shown in Equation (9.26), the air-side con-
vection heat transfer coefficient for the AWHX may be found from vendor data to
eliminate the need for hot-side heat transfer correlations specific to the geometry and
configuration of the HX.

1 /  fin
hh, design  (9.26)
1 A  A  1
 rf , h  rw   h  rf ,c   h 
U design  Ac   Ac  hc

As with the STHX, it is often acceptable to assume that the value for “hh” and thus
“rh” remains constant in many cooling water system applications.

9.3 PLATE HEAT EXCHANGERS

Figure 9.3 illustrates a typical plate heat exchanger (PHX) consisting of a frame and
a pack of corrugated plates separated from each other with gaskets and clamped
together between two end covers with bolts. The cooling water and the fluid to be
cooled flow through the channels created between the plates. The fluids enter and exit
through ports located in the four corners of the plates, and the gaskets seal the plates
at their outer edges and around the ports except as required to achieve the desired
flow between the plates. A variety of flow patterns is made possible by judicious
design of the gaskets. There are many plate designs. A very common plate design is
the “herringbone” in which the plates are stamped with a chevron pattern on either a
30 or 60° angle with the horizontal and alternate plates inverted so that the plates
form crisscross passages with frequent points of contact between the plates. PHX are
compact and are generally easily cleaned with a wire brush in a single day, and their
surface area can be easily increased by simply adding more plates. When compared
to STHX, PHX weigh less, are less expensive, occupy less floor space (no

FIGURE 9.3 Plate heat exchanger. (Courtesy of Alfa Laval Corp.)

134 Engineering of Power Plant and Industrial Cooling Water Systems

tube-pulling space required), and have higher coefficients of heat transfer. Therefore,
they require less cooling water flow to achieve the same degree of cooling. PHX are
primarily suited for liquid-to-liquid heat transfer, although working fluids involving
single-phase gases are possible.
Alfa Laval, a PHX manufacturer, suggests the following performance limits in
their literature:

Pressure, P (k Pa) 2500

Temperature, t (°C) 150
Total effective area, Aeff (m )
2 2200
Mass flow rate, w (kg/sec) 1000

The analysis of PHX is similar to that of other HX. The overall heat transfer coef-
ficient, U, may be expressed as the inverse of the sum of the resistances to heat trans-
fer as shown in Equation (9.27).

1 Q
U   (9.27)
1
hc
 1
hh
 rw  rfouling A F  LMTD 
where
rfouling = is the sum of the fouling resistance on both sides of the plate.
Flow through adjacent passages in a PHX is countercurrent. Therefore, the LMTD
correction factor, “F” is 1 for most PHX applications. The PHX consists of a single
type of plate in which the geometry of the plate is the same on both sides.
The geometry of a PHX is defined by the following parameters:

Number of PHX in service, NHX

Height of plates, LH
Width of plates, Lw
Compressed height of plates, LCP
Number of plates, Np
Thickness of plate, ΔX
Total effective area, Aeff.

Since the plates are corrugated, the effective area of a plate is greater than the product
of the width times the height by the surface enlargement factor, ϕ, which can be as large
as 1.5, depending on the plate manufacture and can easily be measured with a flexible
tape measure. With this information, one may calculate the number of channel passes,
the effective area, the spacing between plates, and the hydraulic diameter as follows:

N cp   N p 1 / 2 (9.28)

Aeff    LW x LH  (9.29)
Heat Exchangers 135

 
b   Lc   X (9.30)
 Np 

4 Lw b
De  (9.31)
 w  2b 
2 L

The coefficient of heat transfer, h, is as shown in Equation (9.32).4,5

hc  hc  C Re
3
4 Pr
1
3
k D 
c
e
(9.32)

The Reynolds number is

Re  DeG /  (9.33)

where μ is the viscosity and

m
G= (9.34)
N p b Lw

and m is the mass flow rate.

The Prandtl number is

 cp
Pr  (9.35)
k

These terms are equally applicable to both the hot and cold sides of the PHX.
In 1999, Bowman6 pointed out that since the geometry on both sides of the PHX
is the same, the expression for “h” for both the hot and cold side is similar. Therefore,
the value of “C” in Equation (9.26) is the same for both sides, and one may solve for
“C” for a clean PHX as follows:

A  LMTD 
U 1  1  rw  (9.36)
hc hh Q

1 1 A  LMTD 
   rw
k D  k D 
3 1 3 1
(9.37)
C Re c 4 Prc 3 c C Re h 4 Prh 3 h Q
e e

3 1 3 1
Re c 4 Prc 3 kc Re h 4 Prh 3 kh 1
C C  (9.38)
De De A  LMTD 
 rw
Q
136 Engineering of Power Plant and Industrial Cooling Water Systems

De De
3 1
 3 1
Re c 4 Prc 3 kc Re h 4 Prh 3 kh
C  (9.39)
A  LMTD 
 rw
Q

This equation makes it possible to estimate the performance of a PHX at other than
design conditions by simply knowing the physical parameters and the predicted heat
transfer rate for a single set of design operating conditions (i.e. hot-side and cold-side
mass flows and inlet and outlet temperatures). If the design data are not available, the
value of C could be determined by conducting a series of heat transfer tests on a clean
PHX over a range of operating conditions.

9.4 HEAT EXCHANGER FOULING

As discussed in Chapter 6, there are two forms of fouling in cooling water systems,
microfouling and macrofouling. Macrofouling occurs in a HX when sediment or
biological creatures such as Asiatic clams or zebra mussels or aquatic weeds such as
seagrass block the flow through a HX. Macrofouling can have a significant impact on
the ability of a HX to operate properly, as the cooling water flow blockage has the
effect of reducing the effective surface area of the HX. (See Equation (9.6)) When
some HX tubes are blocked, the flow through the remaining tubes tends to increase
thus increasing the value of “U” in those tubes. The net effect is to reduce the overall
heat transfer rate, Q, but not as much as the reduction in the effective area would
indicate.
The importance of eliminating macrofouling from the cooling water system as
discussed in Section 6.3 cannot be over-emphasized. In addition, provision should be
made in the design to measure both the flow and pressure drop through the HX, so
that macrofouling can be monitored in operation.
Although a HX is normally procured with a performance specification specifying
the type of HX and the desired amount of cooling, the one additional parameter that
must be specified by the engineer is the fouling resistance, rf. The HX vendor cannot
be expected to know the fouling conditions of the cooling water, rf,c and process side
fluid, rf,h. An estimate of “rf,h” varies widely depending on the fluid that is being
cooled and is beyond the scope of this book.
A distinction should be made between biological fouling and scaling on the tube
side caused by a deposition of substances such as calcium or manganese dioxide that
may plate out on the HX surface due to cooling water with a high pH and/or high
temperature. The thickness of such a deposit may continue to increase without limit,
whereas biological fouling may reach a point where the thickness of the fouling layer
reaches an asymptotic limit due to the shearing effect of the cooling water on the
surface. Somerscales7 described asymptotic fouling (AF) as fouling where resistance
to heat transfer initially increases rapidly when the tube is first exposed to cooling
water but then decreases steadily until the fouling resistance is constant. AF is nor-
mally present when the fouling is due to biological microfouling but not due to sedi-
ment or scale buildup, which is hard and resistant to sloughing off.
Heat Exchangers 137

1.0

not
recom’d
0.8
(Taborek)
fouling resistance m²K/kW

0.6
increasing levels of
chemical treatment
0.4

0.2

TEMA recommendation
0.0
0.5 1.0 1.5 2.0 2.5 3.0 3.5
flow velocity [m /s]

FIGURE 9.4 Fouling resistance.

For raw water, the Tubular Exchange Manufacturer’s Association10 (TEMA) rec-
ommends a minimum design fouling value of 0.002 to 0.003 hr-ft2-oF/Btu (0.00036
to 0.00054 s-m2-oC/J) for tube-side velocities below 3.0 ft/sec (0.91 m/s) and 0.001
to 0.002 hr-ft2-oF/Btu (0.00018 to 0.00036 s-m2-°C/J) for velocities above 3.0 ft/sec
(0.91 m/s). However, Taborek10 noted that the fouling rate is also a function of tube
material and cooling water quality. Figure 9.4 shows the range of fouling rates rec-
ommended by Taborek, overlaid onto those recommended by TEMA for river water
as well as data reported in References 8 and 9. (Taborek10 stated that rf = f(V−1.75).

REFERENCES
1. ASME PTC 12.5-2000. Single Phase Heat Exchangers, September 2000.
2. Thomas, L. C. Heat Transfer- Professional Version, 2nd ed., Capstone Publishing Corp.,
Tulsa, OK, 1999.
3. Colburn, A. P. A. Method of Correlating Forced Convection Heat Transfer Data and a
Comparison with Fluid Friction. Transactions of AIChE, Vol. 29, pp. 174–219.
4. Rauj, K. S. N. and J. Chand. Consider the Plate Heat Exchanger. Chemical Engineering,
Vol. 87, No. 16, pp. 133–144, August 11, 1980.
5. Arpaci, Vedat S. Microscales of Turbulent Heat and Mass Transfer. In Advances in Heat
Transfer, Academic Press, Cambridge, MA, 1997, pp. 1–91.
6. Bowman, C. F. Plate Heat Exchangers. Electric Power Research Institute Service Water
System Reliability Improvement Seminar, Biloxi, MS, 1999.
7. Somerscales, E. F. C. Fouling of Heat Transfer Surfaces: An Historical Review. Heat
Transfer Engineering, Vol. 11, No. 1, pp. 19–36, 1990.
138 Engineering of Power Plant and Industrial Cooling Water Systems

8. Nolan, C. M. and B. H. Scott. On Line Monitoring of Heat Exchangers Microfouling: An

Alternative to Thermal Performance Testing. EPRI SW Reliability Improvement Seminar,
Electric Power Research Institute, Palo Alto, CA, 1995.
9. Zelver, N. et al. Tube Material, Fluid Velocity, Surface Temperature and Fouling: A Field
Study. CTI Paper TP-84-16, Cooling Tower Institute, Houston, TX, 1984.
10. Taborek, J. Assessment of Fouling Research on the Design of Heat Exchangers. Fouling
Mitigation of Industrial Heat Exchangers Conference, Shell Beach, CA, 1995.
 10 Heat Rejection

10.1 REGULATORY REQUIREMENTS
In 1972, the United States Congress passed the CWA, establishing the basic structure
for regulating discharges of pollutants including heat through pipes or man-made
ditches into the waters of the United States. Under the CWA, the EPA was charged
with implementing these pollution control measures and with developing national
water quality criteria recommendations. The EPA has implemented the program
through the NPDES. Each state was required to establish limits for thermal discharge
in their state that are intended to protect the propagation of the receiving body of
water’s balanced indigenous population of fish, shellfish, and wildlife. These limits
address the following parameters outside a reasonable mixing zone near the point of
discharge from a plant:

• maximum discharge temperature

• maximum rate of change in discharge temperature
• maximum temperature rise by the plant.

Under Section 316(a) of the CWA, the permitting state may impose alternative efflu-
ent limitations for controlling thermal discharges in lieu of the effluent limits that
would otherwise be required if the organization receiving the permit can demonstrate
that the otherwise applicable thermal discharge effluent limit is more stringent than
necessary to assure the protection and propagation of the body’s balanced indigenous
population of fish, shellfish, and wildlife. Under the CWA, “new source performance
standards” set the level of allowable discharges from new industrial facilities. This
provision requires the EPA to determine the best available demonstrated control tech-
nology for a given industry that may be more stringent than the limits for existing
dischargers up to and including a “zero discharge” standard.
As stated in Section 3.3, Section 316(b) of the CWA requires power and industrial
plants drawing water from the waters of the United States to implement the BTA to
reduce injury or death of fish and other aquatic life and requires new plants at existing
facilities to install closed-cycle recirculating systems or to reduce actual intake flow
to a level commensurate with that attained by a closed-cycle recirculating system.
Therefore, future cooling water systems may require some method of rejecting
heat directly to the environment such as cooling towers or a spray pond to either
operate as a closed system or in a “helper” mode as discussed in Section 7.6.

10.2 COOLING TOWERS
10.2.1 Cross-Flow Mechanical Draft Cooling Towers
Figure 10.1 Illustrates a common type of cross-flow mechanical draft cooling tower
(MDCT).
139
140 Engineering of Power Plant and Industrial Cooling Water Systems

Air outlet

Fan stack
Fan
Fan gear
Fan motor
Hot water flume Hot water flume

Louvers
Drift eliminators

Air inlet Air inlet

Splash fill

Cold water basin

Water outlet

FIGURE 10.1 Cross-flow mechanical draft cooling tower.

A cross-flow MDCT consists of one or more cells each of which consists of a

large fan affixed on the top of a structure containing splash-type fill material. Warm
cooling water is pumped up to the top of the structure and distributed in the hot water
flume atop the structure. It then passes through specially designed nozzles embedded
in the floor of the hot water flume. The nozzles are fitted with splash plates to evenly
distribute the cooling water over the splash-type fill below. The motor-driven fan atop
each cell of the MDCT operates to draw air into the side of the structure through
louvers. The air passes across the water as it is continually separated into fine drops
by the splash fill before falling to the cold water basin below. After interacting with
and cooling the falling cooling water droplets, the air passes through drift eliminators
and out through the fan at the top of the MDCT. The drift eliminators are designed to
minimize the number of fine drops of cooling water carried out of the fill section of
the MDCT and passed through the fan. A fan stack surrounds each fan to increase the
efficiency of the fan by minimizing recirculation of the air. The cold cooling water is
collected from all of the MDCT cells in a single cold-water basin located below all
of the MDCT cells and then returned to the IPS.

10.2.2 Cross-Flow Natural Draft Cooling Tower

Figure 10.2 Illustrates a cross-flow natural draft cooling tower (NDCT). Typically a
cooling water system alone would not be sufficiently large to require a NDCT, but
since in the past they have been employed in conjunction with a power plant CCW
system and since in power plants the cooling water is frequently discharged into the
CCW system, the NDCT is discussed here for completeness.
A cross-flow NDCT operates in a manner similar to a cross-flow MDCT except
that the air flow is created by the chimney effect of a tall veil. Even when there is no
CCW flow to the NDCT, air flow is induced by the lower density of the air at the top
Heat Rejection 141

Air outlet

Hot water flume

Louvers Splash fill

Air inlet Drift eliminators Air inlet

Water outlet

Cold water basin

FIGURE 10.2 Cross-flow natural draft cooling tower.

of the veil. When the power plant is not operating but there is still CCW flow to the
NDCT, there is less dense saturated air entering the veil, since the molecular weight
of water vapor (18) is less than that of air (29). Depending on the water loading on
the fill section, this may result in an increase in air flow as the lighter air tends to rise.
When the power plant is operating, the heat added to the CCW results in an even
greater air flow, as warm saturated air is even less dense. A canopy between the
donut-shaped heat transfer section around the veil containing the splash-type fill
material and the veil encloses the interior of the NDCT so that no air can bypass the
fill section.

10.2.3 Splash-Type Cooling Tower Fill Material

Figure 10.3 illustrates splash-type fill that is commonly used with cross-flow MDCT
and NDCT.
Splash-type fill material may be constructed from a variety of materials such as
wood slats but most commonly of plastic in large cross-flow cooling towers serving
large plants. The plastic slats may be oriented either parallel to or perpendicular to
the air flow. (The parallel orientation results in less resistance to air flow.) The slats,
normally supported by a system of plastic or coated wire hangers, are frequently
perforated and come in a variety of shapes for strength.
One of the major drawbacks with cross-flow cooling towers, both MDCT and
NDCT, is the propensity for ice to build up on the fill material in cold weather.
Various schemes have been employed to prevent ice buildup such as diverting a
larger portion of the CCW to the perimeter of a NDCT or shutting off or reversing the
fan flow in MDCT, but all such schemes require diligent observation to identify when
142 Engineering of Power Plant and Industrial Cooling Water Systems

Air in Air out

FIGURE 10.3 Splash-type cooling tower fill material.

ice buildup has begun. As a result, cross-flow cooling towers are in somewhat disfa-
vor where there is the potential for ice formation.

10.2.4 Counter-Flow Mechanical Draft Cooling Tower

Figure 10.4 Illustrates a counter-flow MDCT. As is the case with a cross-flow MDCT,
each cell in a counter-flow MDCT consists of a large fan affixed on the top of a struc-
ture but the fill material is of the thin-film variety. The heated cooling water is pumped
through a piping system located above the fill section and passes through especially
designed nozzles located at the bottom of pipes fitted with splash plates to evenly
distribute the cooling water over the thin film fill material below. As with a cross-flow
MDCT, the motor-driven fan atop each cell operates to draw air into the side of the
structure. The air passes through the fill material where the cooling water is running
down the fill material in thin sheets before it falls to the cold water basin below. After
interacting with and cooling the falling cooling water, the air passes through drift
eliminators and out through the fan at the top of the MDCT. The cold cooling water
is collected from all of the MDCT cells in a single cold water basin located below all
of the MDCT cells and returned to the IPS.

Air outlet

Fan Fan stack

Fan motor Fan gear

Drift eliminators

Water inlet

Thin film fill

Air inlet Air inlet

Cold water basin

Water outlet

FIGURE 10.4 Counter-flow mechanical draft cooling tower.

Heat Rejection 143

10.2.5 Counter-Flow Natural Draft Cooling Tower

Figure 10.5 Illustrates a counter-flow NDCT.
In an NDCT, the heated CCW is pumped up through concrete standpipes that
distribute the CCW through a piping system located above the fill section. It then
passes through especially designed nozzles located at the bottom of the pipes fitted
with splash plates to evenly distribute the CCW over the thin film fill material below.
A counter-flow NDCT operates in a manner similar to a counter-flow MDCT except
that as in the case of a counter-flow NDCT, the air flow is created by the chimney
effect of a tall veil. As with a cross-flow NDCT, even when there is no CCW flow to
the NDCT air flow is induced by the difference in the density of the air at the top of
the veil, which is less than that at ground level. Air flow is also increased when the
power plant is operating since the heat added to the CCW results in warm saturated
air that is even less dense. Counter-flow NDCT have proven to be much more resis-
tant to ice damage due to the much more durable fill material and the fact that dam-
age is normally limited to the outer edges of the HX section.

10.2.6 Thin Film-Type Cooling Tower Fill Material

Figure 10.6 illustrates thin film-type fill commonly used with counter-flow MDCT
and NDCT.
One type of thin film-type fill shown in Figure 10.6 consists of several layers of
very thin vertical sheets hung from structural members such that when the cooling
water lands on the top of the top sheet, it tends to dribble down the sheet in a thin film

Air outlet

Drift eliminators

Thin film fill

Air inlet Air inlet

Water inlet Water outlet

Cold water basin

FIGURE 10.5 Counter-flow natural draft cooling tower.

144 Engineering of Power Plant and Industrial Cooling Water Systems

Air out

Air in

FIGURE 10.6 Thin film-type cooling tower fill material.

FIGURE 10.7 Plastic cooling tower fill material.

before falling to the basin below. Another type of cooling tower fill material shown
in Figure 10.7 is thin corrugated plastic glued together in a herringbone pattern
described as “egg crates” manufactured by Munters and Hitachi. This type of fill
material, though more efficient than that shown in Figure 10.6, may be susceptible to
clogging. If the surface of the plastic is permitted to be covered with pseudomonas
bacteria (i.e. slime), suspended solids in the cooling water tend to build up on the
surface of the plastic, eventually obstructing flow. Chemical treatment systems
employing oxidizing biocides such as chlorine are volatile, and the biocide and may
not control the buildup of pseudomonas on the fill material.
Heat Rejection 145

10.3 SPRAY PONDS
10.3.1 Advantages of Spray Ponds
A spray pond is a system of pipes and spray nozzles that spray water into the air to
cool the water. They are similar in this way to cooling towers in that they dissipate
waste heat to the atmosphere principally through evaporation. Much of the advance
in the understanding of the design of spray ponds has been gained because they play
a critical role in the safe operation of some nuclear plants.1 Unlike MDCT, spray
ponds need no fan power or fill maintenance. Spray ponds are less susceptible to ice
damage than cooling towers and can operate satisfactorily under cold conditions if
provided with a spray bypass to maintain minimum temperatures in the pond.

10.3.2 Conventional Flat Bed Spray Ponds

Figure 10.8 shows the conventional flat bed spray pond (FBSP) that was the ultimate
heat sink for the now defunct Rancho Seco nuclear station.1 Typically, a FBSP con-
sists of a series of trees mounted on straight header pipes in a rectangular pattern with
each tree consisting of a riser pipe and four cross arms at 90 degree angles approxi-
mately five feet (1.5 m) long with a spray nozzle at the end of each cross arm pointed
vertically. The dimensions of each of the two FBSP at Rancho Seco is 165 by 330
feet (50.3 by 100.6 m), with 304 nozzles each delivering 53 gal/min (3.34 l/s) at
7 Lbf/in2 (48.3 kPa) nozzle pressure.
In the FBSP design with all spray nozzles oriented in the vertical direction, the
bulk drag force of the water droplets (vertically downward) resists the natural

FIGURE 10.8 Spray nozzles in a conventional flat bed spray pond.

146 Engineering of Power Plant and Industrial Cooling Water Systems

buoyancy of the warm air (which is of the same order of magnitude and is directed
vertically upward). Since the two forces acting on the air oppose each other, the
result is a reduced air flow rate through the spray region and a large increase in the
local wet-bulb temperature (WBT). Additionally with a large FBSP, the falling water
tends to block air flow into the central parts of the array, requiring an interference
factor to be applied to determine the local WBT as a function of the ambient WBT.1

10.3.3 Oriented Spray Cooling System

The oriented spray cooling system (OSCS) employs a radically different spray pond
design that overcomes these problems. Figure 10.9 shows the OSCS that was con-
structed, operated, and thoroughly tested at the Columbia Generating Station (CGS).
The outstanding feature of the OSCS design is the circular arrangement of the
spray nozzles on spray trees. The spray trees are spaced approximately 13’-9” (4.2 m)
on center with Spraying Systems Co. Whirljet Type 11/2 CX SS 27 spray nozzles
arranged in a helical pattern spaced approximately 4 feet (1.2 m) from the mast of the
tree and approximately 2’-8” (0.81 m) apart vertically and oriented at a 35° angle
from the vertical toward the center of the circle.1 In this design, both the bulk drag
force of the water droplets on the air and the buoyant force promote ventilation of the
spray region, and ambient air flow to the spray nozzles is not obstructed. The result
is a reduction on the local WBT in the spray region and improved cooling of the
droplets as they fall through the spray region to the pond surface below. However, the
required nozzle pressure is higher than with a FBSP.

FIGURE 10.9 Oriented spray cooling system at the Columbia Generating Station.
Heat Rejection 147

10.3.4 Analysis of Spray Ponds

The simplest measure of thermal performance of a spray pond is the efficiency, η, of
the device as defined by the following standard formula:

t1  t2
  (10.1)
t1  tWB

where
η = spray efficiency
t1 = temperature of the water entering the spray nozzle
t2 = temperature of the sprayed water
tWB = ambient WBT.
Assuming a constant heat load and WBT one may solve for t2 as follows:

t2  t1    t1  tWB local  (10.2)

where
tWB-local = wet-bulb temperature at the spray nozzle.

10.3.5 Spray Pond Test Results

Only a few quality full-scale spray pond efficiency tests are available in the open
literature.1 Figures 10.10 and 10.11 show the comparison among tests conducted on
the conventional FBSP at Rancho-Seco and Okeelanta Florida complex and the

70%

60%

50%
Efficiency

40% Rancho Seco

Okeelanta
30% Columbia

20%

10%

0%
40 50 60 70 80

Wet-bulb Temperature (F)

FIGURE 10.10 Flat bed spray ponds and oriented spray cooling system efficiencies vs.
WBT.
148 Engineering of Power Plant and Industrial Cooling Water Systems

70%

60%

50%

40%
Efficiency

Rancho Seco
Okeelanta
30% Columbia

20%

10%

0%
0 5 10 15 20
Ambient Air Velocity (Mi/Hr)

FIGURE 10.11 Flat bed spray ponds and oriented spray cooling system efficiencies vs.
ambient air velocity.

OSCS at the CGS. One may see from Figure 10.10 that the spray efficiency increases
with increasing WBT. This is the case for any type of evaporative cooling device such
as a cooling tower.
Figure 10.11 shows the significant impact that ambient air velocity has on spray
ponds, as the spray efficiency increases with increased ambient air velocity. At zero
air velocity, the OSCS is seen to have a very significant advantage in efficiency,
because the OSCS creates its own wind.

REFERENCE
1. Bowman, C. F. and S. N. Bowman Thermal Engineering of Nuclear Power Stations:
Balance of plant Systems, CRC Press, Boca Raton, FL, 2021.
Nomenclature
Symbol Definition English Units SI Units
A corrosion allowance in m
A wave speed ft/sec m/s
a1 constant Dimensionless Dimensionless
a2 constant Dimensionless Dimensionless
A cross-sectional area ft2 m2
Aeff effective surface area ft2 m2
Ah effective hot-side heat exchanger surface area ft2 m2
BHP brake horsepower Hp Hp
C saturation factor Dimensionless Dimensionless
C Hazen-Williams “C” factor Dimensionless Dimensionless
cp specific heat Btu/lbm-oF kJ/kg-oC
cp,c tube-side specific heat Btu/lbm-oF kJ/kg-oC
cp,h hot-side specific heat Btu/lbm-oF kJ/kg-oC
D pump bell diameter ft m
d* dimensionless parameter Dimensionless Dimensionless
dNOM, n nominal pipe diameter in each section of pipe in m
dNOM nominal inside diameter of new pipe in m
dCALC calculated inside diameter of pipe in m
ΔdMEAS measured diameter reduction in m
D pipe diameter ft m
De hydraulic diameter ft m
Do outside diameter min
d diameter of pipe in
di inside diameter of a pipe or tube ft m
do tube outside diameter ft m
d1 orifice diameter in
EM bulk modulus of elasticity of the mixture of lbf/in2 Pa
water and air
Ep bulk modulus of elasticity of the pipe lbf/in2 Pa
Ew bulk modulus of elasticity of water lbf/in2 Pa
e absolute roughness of pipe in
f friction factor. Dimensionless Dimensionless
f magnitude of the reflected pressure wave lbf/in2 kPa
F LMTD correction factor Dimensionless Dimensionless
F magnitude of the pressure wave lbf/in2 kPa
Fd Froude number Dimensionless Dimensionless
g gravitational constant ft/sec2 m/s2
G flow rate gal/min l/s
hL total head loss ft m
hL,f friction head loss ft m
hc cold-side convection coefficient Btu/hr-ft2-oF J/s-m2-oC
hc-in cold stream enthalpy in Btu/lbm kJ/kg
hc-out cold stream enthalpy out Btu/lbm kJ/kg
hh hot-side convection coefficient Btu/hr-ft2-oF J/s-m2-oC
hh,design design hot-side convection coefficient Btu/hr-ft2-oF J/s-m2-oC
hh-in hot stream enthalpy in Btu/lbm kJ/kg
hh-out hot stream enthalpy out Btu/lbm kJ/kg
H pressure head ft kPa
(Continued)

149
150Nomenclature

Symbol Definition English Units SI Units

H total head ft m
hL,F form head loss per equivalent foot of pipe ft m
hL,f /100 friction head loss in feet per 100 ft of pipe ft m
Ho initial pressure head ft kPa
Kair Henry’s Law coefficient Dimensionless Dimensionless
K’air Henry’s Law coefficient based on the HX Dimensionless Dimensionless
outlet temperature
Ki cavitation index Dimensionless Dimensionless
Kx Henry’s Law coefficient for gas x Dimensionless Dimensionless
k resistance coefficient Dimensionless Dimensionless
k thermal conductivity of water Btu/hr-ft-oF kJ/s-m-oC
kc thermal conductivity of water Btu/hr-ft-oF kJ/s-m-oC
kt thermal conductivity of the tube material Btu/hr-ft-oF kJ/s-m-oC
L length of pipe ft m
LCP compressed height of plate in m
leff effective tube length ft m
LH height of plate in m
LMTD log mean temperature difference oF oC

Lw width of plate in m
m mass lbm kg
m cooling water mass flow rate lbm/hr kg/s
MHP Motor horsepower hp Hp
m mass flow rate lbm/hr kg/s
mc mass flow rate of cold stream lbm/hr kg/s
mh mass flow rate of hot stream lbm/hr kg/s
N rotational speed rev/min
Nair mole fraction of all dissolved gases Dimensionless Dimensionless
N’air = mole fraction that local conditions will permit Dimensionless Dimensionless
to remain in solution
Nair-max maximum mole fraction of air in the entering Dimensionless Dimensionless
cooling water
Nx mole fraction of dissolved gas, x Dimensionless Dimensionless
n specific speed Dimensionless Dimensionless
n number of moles Dimensionless Dimensionless
NHX number of PHX in service Dimensionless Dimensionless
Np number of plates Dimensionless Dimensionless
Npass number of passes Dimensionless Dimensionless
NPSHA net positive suction head available ft m
NPSHi cavitation inception ft m
NPSHR net positive suction head required ft m
Ntubes number of tubes per pass Dimensionless Dimensionless
Nu Nusselt number Dimensionless Dimensionless
P effectiveness Dimensionless Dimensionless
P pressure lbf/in2 kPa
Prequired minimum gauge pressure to keep air in lbf/in2 kPa
solution
P’ local static pressure lbf/in2 kPa
P1 gauge pressure 1 lbf/in2 kPa
P2 gauge pressure 2 lbf/in2 kPa
P internal pipe pressure lbf/in2 kPa
Pb barometric pressure lbf/in2A kPaa
Pd local downstream pressure lbf/in2 kPa
(Continued)
Nomenclature 151

Symbol Definition English Units SI Units

Psat = saturation vapor pressure at the maximum lbf/in A
2 kPaa
operating temperature
Pu local upstream pressure lbf/in2 kPa
Pv vapor pressure at the cooling water ft M
temperature
Px partial pressure of dissolved gas x lbf/in2A kPaa
Pr Prandtl number Dimensionless Dimensionless
Q Flow rate gal/min
Qn flow in each section of pipe gal/min
Q maximum HX heat load Btu/hr kJ/s
Q rate of heat transfer Btu/hr kJ/s
R Universal gas constant ft-lbf/lbm-mole oR cal/gm-mole oK
Re Reynolds number Dimensionless Dimensionless
Ret tube-side Reynolds number Dimensionless Dimensionless
rfouling fouling resistance on both sides of the plate hr-ft2/Btu m2-oC/W
S maximum allowable stress lbf/in2
Smin minimum submergence ft m
s suction specific speed Dimensionless Dimensionless
s number of pipe sections Dimensionless Dimensionless
S submergence ft m
T absolute temperature oR oK

T’ absolute local temperature oR oK

T time sec s
t thickness of pipe wall in m
t temperature oF oC

tm minimum required wall in

Tout HX outlet temperature oF oC
Tin HX inlet temperature oF oC
T1 temperature of water entering the spray nozzle oF oC

T2 temperature of sprayed water oF oC

tc-in cold stream temperature in oF oC

tc-out cold stream temperature out oF oC

Th-in hot stream temperature in oF oC

Th-out hot stream temperature out oF oC

tWB ambient wet-bulb temperature oF oC

tWB-local wet-bulb temperature at the spray nozzle oF oC

U overall heat transfer coefficient Btu/hr-ft2-oF W/m2-oC

Udesign design overall heat transfer coefficient Btu/hr-ft2-oF W/m2-oC
V volume ft3 m3
Vpipe water velocity in a pipe or tube ft/sec m/s
Vr air-to-water volume ratio Dimensionless Dimensionless
VS velocity at suction inlet ft/sec m/s
VH1 velocity head 1 ft m
VH2 velocity head 2 ft m
VMw volume of water ft3 m3
VMa volume of air ft3 m3
VMT total volume ft3 m3
V velocity ft/sec m/s
Vo initial fluid velocity ft/sec m/s
vt tube velocity ft/sec m/s
w mass flow rate lbm/sec
y temperature coefficient Dimensionless Dimensionless
Z1 elevation of the gauge 1 ft m
(Continued)
152Nomenclature

Symbol Definition English Units SI Units

Z2 elevation of the gauge 2 ft m
ΔP pressure drop across the orifice lbf/in2 kPa
ΔX thickness of plate ft m
ηm motor efficiency Dimensionless Dimensionless
η spray efficiency Dimensionless Dimensionless
ηfin fin efficiency Dimensionless Dimensionless
ηp pump efficiency Dimensionless Dimensionless
μ dynamic viscosity of the water lbm/ft-hr kg/m-s
μt tube-side viscosity lbm/ft-hr kg/m-s
Δ density lbm/ft3 kg/m3
ρa density of air lbm/ft3 kg/m3
ρw density of water lbm/ft3 kg/m3
ρM density of the mixture of water and air lbm/ft3 kg/m3
σ Thoma’s cavitation constant Dimensionless Dimensionless
Index
Page numbers in bold indicate tables, page numbers in italic indicate figures.

A Bowman-­Bain Equation, 99, 105

Bowman, C. F., 21, 47, 66, 74, 97, 99, 102,
316(a), 139 104–105, 107, 135
absolute pressure, 9–10, 30 brake horsepower, 33, 35
Actibrom 1338, 84–85 bromine, 84–85, 87, 89–91
Advisory Committee on Reactor Safeguards, 56 bromochlordimethylhydantoin, 88–89, 91
aerobic, 77 Browns Ferry Nuclear Plant, 61–62, 68–72,
air liberation, 113, 115, 117, 119–120 89–90, 99, 109
air release valve, 16, 23, 35 Brunswick, 4, 21
air venting 35 bulk drag force, 145–146
air-­to-­water heat exchanger, 131–133 Bulletin 80–24, 3
algae, 2, 22, 51 Bulletin 81–03, 3, 6
Alley, J. R., 67 buna-­N rubber, 85
allowable stress, 51–52, 59 buoyancy, 146
alloy cladded pipe, 51 buried pipe, 41, 47, 52–53
AL-­6XN, 85 butterfly valve, 3–4, 7, 16, 23, 57–58, 100, 125
American Society of Mechanical Engineers, 37, butt weld, 68–69, 73, 75, 79, 82, 89–90
49, 52, 55, 58–59, 69–70, 75, 77 byssal thread, 80–81
American Water Works Association, 42–44, 48, 61
anaerobic, 77–78
analysis of spray ponds, 147 C
analysis of water hammer, 124
Calvert Cliffs, 5
anti-­foulant paint, 25
carbon steel, 2–3, 12, 19, 22, 37–38, 41–42,
approach velocity, 15
50–51, 55–56, 61, 65, 67–70, 73–79,
aquatic animals, 2, 22
83–86, 88–91, 96, 99, 101–102, 104,
aquatic life, 19, 22, 110, 139
109, 111–112
Arkansas Nuclear One, 2, 4
Carolina Power & Light, 68–69
Asiatic clam, 4, 6, 68, 80–81, 83, 87, 89, 91, 136
Carr, F. P., 97, 115
ASME Section XI, 37, 70
Catawba, 4, 51, 55
automatic backwashing strainers, 25, 90
cathodic protection, 42, 75
axial flow, 27, 30, 30
cavitation, 7, 30–31, 33–34, 50, 56, 120–121
azol blend, 90
cavitation index, 120
CCW pump, 1–2, 17–18, 22, 27–29, 31, 33, 36,
B 110
Celsius, 9
316(b), 19, 110, 139 cement-­mortar lining, 45, 47–49, 52–56, 109
Bain, W. S., 66, 99, 102, 104–105, 107 centrifugal pump, 1, 27, 30, 33
ball valve, 58 “C” factor, 96–97, 103, 107, 108, 108, 109, 109,
Beaver Valley, 3 111, 111
best technology available, 19, 21–22, 110, 139 chemical discharge limits, 82
Big Bend, 21 chemical injection, 82
biocide, 51, 76, 79–81, 84–87, 91, 108–109, 144 chemical treatment, 78, 80–82, 84, 87–91, 137,
biocide testing, 84 144
biofilm, 77, 79, 86–87 chlorination, 81–82, 83, 87
biofouling, 4, 25, 79–81 chlorine residual, 83
bituminous coating, 42, 45 Clean Water Act, 19, 110, 139
bivalves, 6 Clinton, 5–6
Blackburn F. E., 70 closed impeller, 28, 37
bolting material, 2 closed recirculating, 1

153
154Index

coatings for zebra mussel control, 82 counter-­flow natural draft cooling tower, 143
Colbert, 61 cross-­flow mechanical draft cooling tower,
Colburn equation, 130 139–140
cold-­water basin, 140 cross-­flow natural draft cooling tower, 140–141
Colebrook equation, 95 Cumberland, 62
colony-­forming units, 86 cuprous oxide, 82, 91
Columbia, 147–148 cylindrical wedgewire screens, 20–21
Columbia Generating Station, 146, 148
component cooling water, 5
concrete pressure pipe, 3, 41–42, 50 D
condenser circulating water, 1–3, 17–19, 21–22, Darcy-­Weisbach Equation, 95, 101–102, 104
27–29, 31, 33, 33, 36, 82, 90, 110, 111, Deardorff, A. F., 69
140–141, 143 debris, 2, 4, 18, 21
condenser discharge valve, 36 density, 3, 10–11, 13, 21, 35, 52, 75, 122–123,
condenser vacuum, 36 131, 140, 143
condenser vacuum pump, 36 design deficiencies, 18
condenser waterbox, 70 design error, 5
containment, 3, 5, 68 design margin, 69, 77
containment air cooler, 5 Detroit Edison, 4, 82
containment spray, 5, 87 DGH-­QUAT, 89, 91
control and analysis of water hammer, 124 Diablo Canyon, 5, 82
control of water hammer, 121 diesel generator, 3, 5, 7, 109
control room chiller, 5 differential aeration cell, 76–77
control valve, 4, 31, 58, 119 dispersant, 84–85, 88–91
conventional flat bed spray ponds, 145 dry-­pit pumps, 27
convection coefficient, 130–131, 133 Dunn, G. C., 121
cooling coil, 5 dynamic viscosity, 11–12
cooling tower, 1–2, 19, 22, 27, 48, 116, 139–145,
148
cooling tower fill material, 2, 141–144 E
cooling tower makeup pump, 2
cooling water pump, 1–2, 5, 16–18, 16, 23, 24–25, earthquake, 52–53
27–28, 30–31, 33, 35–38, 57, 89, 113, effectiveness, 84–85, 129
113–114, 125 efficiency, 27, 29–31, 33–36, 34, 59, 132, 140,
cooling water system, 1–7, 9–10, 15, 18–20, 147, 147, 148, 148
22–25, 27, 31, 35–38, 41, 55, 57–58, emergency equipment cooling water, 61, 68–69,
68, 74–75, 77, 79–84, 89–91, 95–96, 72, 99, 109
105, 109, 113, 115, 120–121, 124, 127, engineering, 1, 5, 7, 9, 15
131, 133, 136, 139–140 Environmental Protection Agency, 19–21, 110,
copolymer, 89–91 139
copper, 3, 82, 85, 90–91, 96–97 epoxy coating, 3, 45, 50
copper epoxies, 82, 91 equivalent lengths, 97, 100, 107, 110
corbicula fluminae, 80 erosion, 6, 45, 56, 77–78, 120
corrosion, 2–6, 38–39, 41–45, 50–51, 56, 59, essential raw cooling water, 47–48, 56, 69, 73, 78,
61–62, 64–65, 65, 66–68, 70, 70, 71, 82, 84–85, 87–89, 107–108, 109, 109
72, 72, 73–79, 82–91, 96, 99, 101–102, evaporative cooling, 116, 148
110, 120 extensive properties, 11
corrosion allowance, 51, 59, 66, 74, 77 exterior corrosion, 42–43, 64, 66
corrosion in cooling water systems, 74, 77
corrosion inhibitor, 75, 82–84, 89 F
corrosion inhibitor testing, 82
corrosion of shaft, 2 Fahrenheit, 9
corrosion product, 3–4, 27, 47, 50, 62, 67, 73, 77, fan, 132, 140, 140–142, 142, 145
79, 83–86, 88, 99, 101–102 Farley, 3–4
corrosion rate, 42, 47, 65, 65, 68, 70, 70, 71, 72, fiberglass-­reinforced plastic, 3, 41, 45, 55
72–74, 77–78, 85, 87–90 fill material, 2, 111, 140–144
counter-­flow mechanical draft cooling tower, 142 fill section, 140–143
Index 155

fine mesh screens, 20–21 heat exchanger, 1, 3–7, 12, 18, 20, 31, 45, 56, 61,
fire protection piping, 72–73 79, 81, 87–91, 97, 99, 109–111, 113,
fire protection pumps, 2, 17 113–114, 115, 117, 119–121, 125,
fish, 2, 4, 18–24, 79–80, 110, 139 127–134, 136, 143
fish eggs, 21, 24 heat exchanger fouling, 5, 136
Fitzpatrick, 2–4 heat transfer, 6, 12, 18, 79, 88, 90, 115, 119,
flashing, 119–120 127–129, 132–136, 141
flat bed spray pond, 145–148 heat treatment, 5, 39, 81
floating boom, 21–22 heat treatments for zebra mussel control, 82
flow, 1–6, 11–12, 15–16, 17, 17, 19, 21–24, 27, Henry’s law, 115–116
29–31, 33, 35, 37–38, 41–42, 44, 47, Henry’s law coefficient, 115–118, 116
50, 57–58, 61–62, 63–64, 64, 66–68, Herro, H. M., 77
70, 73, 75, 77–80, 87–91, 95–97, 96, high density polyethylene, 3, 41, 44, 51, 55
99, 100, 101, 101, 102–103, 106–111, Hope Creek, 2
115, 117–120, 123–125, 127, 129, Horizontal centrifugal pump, 1
131–136, 137, 139–144, 146 Hubble, J. D., 121
flow and pressure drop tests, 5, 88, 106 hung piping, 50–51, 79, 90
flow test, 4, 47, 101, 101, 106–107, 109 hydraulic gradient, 36, 113–114, 117, 125
flow velocity, 77, 137 hydraulic pressure gradient, 113, 113–114
foreign objects, 2 hydrogen cooler, 111
form loss, 97–98, 106–107 hyprobromous acid, 88–90
form loss equation, 97
fouling, 4–6, 25, 61, 79–81, 90–91, 129–131, 134,
I
136, 137, 137
fouling resistance, 130, 134, 136, 137, 137 IN 2007-­06, 7
frazil ice, 2, 18, 24 IN 2007-­28, 5
friction loss, 15, 97, 99, 101, 106 IN 81-­21, 4
Froude number, 15, 117 IN 83-­46, 3
IN 85-­24, 3
IN 85-­30, 3
IN 86-­96, 4
G
IN 88-­37, 4
Gallatin, 61, 101, 101, 103–104, 106 IN 89-­76, 4
galvanic pitting corrosion, 3 IN 90-­26, 5
galvanized, 64–65, 67, 96 IN 90-­39, 4
gate valve, 3, 57–58, 100 IN 92-­49, 2
gauge pressure, 9–10, 10, 106, 108, 113, 113–114, IN 94-­61, 3
117 IN 94-­79, 3
general corrosion, 47, 68, 76, 86 Indian Point, 3
Generic Letter 89–13, 2, 6 Institute for Applied Microbiology, 85–86
Generic Letter 90–05, 3 Institute of Nuclear Power Operations, 18, 37,
Generic Letter 96–06, 5 68, 76
globe valve, 58, 100 intake, 1–2, 5, 15–16, 18–23, 82, 139
Grand Gulf, 2 intake blockage, 2
Guthrie, P. V., 47, 74 intake channel, 15, 23
intake pumping station, 1–3, 16–19, 22–25, 47–48,
73, 79, 81, 88–89, 107, 140, 142
H
intake structure, 2, 15, 18, 20–21, 23
halogen, 86–89 intensive properties, 11
Harbor Beach, 82 iron oxide, 61, 67, 74, 88, 101
Hazen-­Williams Equation, 95–97, 102, 105, 107 iron-­reducing bacteria, 77
head, 10, 22, 27, 29–31, 33, 33–34, 34–35, 37–38,
38, 44, 47–48, 56–57, 73, 78–79, J
87, 95–98, 101–102, 106–108, 108,
109–111, 113, 113–114, 122–124 John Sevier, 62–63
head loss, 27, 95–98, 101–102, 106–107, 109, 113 joint efficiency, 59
156Index

K N
Kelvin, 9 NaBr/NaOC1, 86
kinematic viscosity, 11, 96 National Pollution Discharge Elimination System,
Kingston, 62, 101, 101, 103, 104, 105 19–20, 139
natural draft cooling tower, 22, 110–111,
140–141, 143
L
net positive suction head, 30–31, 34
larvae, 21, 24, 80–81 nodules, 79, 85, 89, 91
leak, 3, 7, 20, 37, 43, 66, 68–69, 72–76, 78, 90, nondestructive examination, 69–70
120, 131 NRC oversight, 6
leaves, 2 Nuclear Regulatory Commission, 1–7, 55–56, 68,
Licina, G., 68 75, 80
Licensee Event Report 97-­037, 3 nuclear safety, 5, 7, 18, 37, 47, 52, 75
Licensing Event Report 96-­001-­01, 5 NUREG 1275, 6
line shaft, 1, 5, 27–28, 36–38 Nusselt number, 130
line shaft bearing, 36
LMTD correction factor, 128–129, 132, 134 O
log-­mean temperature difference method, 128
louver, 20, 140, 140, 141 oil spill, 18
lube oil cooler, 111 Okeelanta Florida Complex, 147
Ontario Ministry of the Environment, 4
open once-­through, 1, 27
M
open recirculating, 1
macrofouling, 79–80, 90, 136 oriented spray cooling system, 146–148
main condenser, 1, 22, 27, 36, 71–72, 111 orifice, 16, 35, 101, 108, 120–121, 125
makeup, 1–2 orifice plate, 120
manganese, 67, 75 overall heat transfer coefficient, 128–129, 132,
manganese dioxide, 136 134
marsh grass, 2, 22 oxidizing biocide, 51, 76, 89–91, 144
mass, 22, 35, 75, 117, 127, 134–136 oxygen scavenger, 82
McGuire, 5 Oyster Creek, 2–3
Metell, H. M., 75, 77–78 oyster shell, 4
Metro Water District of Southern California, 52
microbiologically induced corrosion, 3, 7, 12, 25, P
47, 68–70, 73, 75–79, 82, 85–86, 88–91
microfouling, 5, 79–81, 90, 136 Palo Verde, 3, 5
microorganism, 61, 74, 76, 85 Pavinich, W. A., 69
MIC tubercles, 76 Peach Bottom, 4
Miller, W. T., 8, 56 Pickering, 2
Millstone, 2–3 pipe coating, 3, 6, 41–45, 50–52, 65, 82, 91
mils per year, 47, 63, 65, 65, 70, 71, 72, 72, pipe, concrete pressure, 3, 41–42, 50
73–74, 78, 90 pipe fitting, 45, 48–51, 56, 58–59, 97–98, 110
minimum wall thickness, 73 pipe flow, 95
mixed flow, 1, 15, 27, 30, 37–38 pipe form losses, 97–98, 106–107
modulus of elasticity, 52, 122–123 pipe friction, 95–96
mole fraction, 115–118 pipe hangers, 41, 50–51
molluscicide, 81 pipe, hung, 50
mollusk, 4, 58, 79–82, 90–91 pipe, in situ lined, 45
momentum diffusivity, 12 pipe lining, 41, 45–46
Moody diagram, 11–12 pipe, plastic, 44–45
Moody, L. F., 103 pipe, pre-­lined ductile iron, 43
motor, 1-­3, 16, 22, 23, 24, 27–28, 35–38, 57, 124, pipe wraps, 41
140, 140, 142, 142 piping, buried, 41, 47, 52–53
Mullin, L. J., 70 piping codes, 37, 47, 49, 52, 55, 58, 75, 77
MULTIFLOW, 109–111, 111 piping leak, 3, 7, 66, 68–69, 72–75, 78, 90
multi-­pressure condenser, 22 piping materials, 3, 41–42, 45, 75, 84
Index 157

piping network analysis, 110 raw cooling water, 4, 38, 111–112, 112
piping, pre-­lined ductile iron, 41, 43 reactor, 3–5, 37, 56, 75
piping standards, 44–45, 47, 55 reduction in pipe diameter, 66, 101
piping wall thickness, 58–59, 62, 65–67, 73, 75, regulatory requirements, 139
77–79, 122 relief valves, 110, 125
pitting, 3, 47, 51, 61–62, 63–64, 65, 65, 68, resistance coefficient, 97
72–74, 76–79, 86, 90 Reynolds number, 11, 131, 135
pitting corrosion rate, 65, 65, 72–74, 77–78, 90 Ristroph traveling screen, 20
planktonic, 80, 85 Robinson, 3, 68
plate heat exchanger, 133 room cooler, 5–6
Point Beach, 5 rototional speed, 29
polyphosphate, 82–84, 83, 89, 91 roughness coefficient, 96, 96
polyvinyl chloride, 3
potential improvements, 22, 37, 77
S
Prandtl number, 11–12, 131, 135
pressure, 1–5, 9, 9–10, 10–12, 19–20, 22, 24–25, Safety Evaluation Report, 56
30–31, 35, 37, 41–45, 49–51, 55, Salem, 4–5
57–59, 67, 77, 79, 88–89, 96, 101–102, San Onofre, 4
104–111, 111, 113, 113–114, 115–125, screen wash pump, 2, 16, 23–24
134, 136, 145–146 seagrass, 79–80, 136
pressure drop, 5, 11–12, 22, 25, 31, 35, 41, 57–58, seal water cooler, 4
67, 79, 88, 96, 101–102, 104–106, seaweed, 2, 18
113, 136 Section 316(a), 139
pressure wave, 121–124 Section 316(b), 19, 110, 139
properties of water, 9–10 sedimentation, 50–51, 78
psudemonas, 77 seismic qualification, 52, 56
pump capacity, 15 semi-­open impeller, 27–28, 37–38
pump efficiency, 30, 33, 34, 35 Sequoyah Nuclear Plant, 25, 62, 63, 69, 73–74,
pump head, 31, 33, 33–34, 37 78, 82, 84–85, 87–88, 107–108,
pump head capacity curve, 31, 33 110–111, 111–112
pump head curve, 31 service factor, 36
pump impeller, 1–2, 27, 29–30, 34 Service Water Assistance Program, 68
pump impeller profile, 30 service water pump, 2, 17
pump in-­service testing, 37 sessile, 85–86
pump line shaft, 1, 5, 27–28, 36–38 shell-­and-­tube heat exchanger, 127–128, 132–133
pump line shaft seal, 38 silica, 67, 75
pump margin, 37 silicone oil, 82, 91
pump minimum flow rate, 35 silicon oxide, 101
pump motor, 16–23, 35–36, 38 Singleton Materials & Engineering Laboratory,
pump run-­out, 31 52, 70, 71, 73, 84, 88
pump shaft coupling, 2, 36 skimmer wall, 23
pump shut-­off head, 31, 35 Slime-­forming bacteria, 77
pump size, 15, 30, 33 snubber, 124
pump speed, 5, 29, 33, 36–37 socket weld, 56, 73, 79
pump suction, 1–2, 4, 15, 27, 29–31, 34, 39, 111 sodium bromide, 84–85
sodium hypochlorite, 81, 83–85, 88–91
Q sodium sulfide, 82
Somerscales, E. F. C., 136
qualification testing, 82 South Texas Project, 7
quality assurance, 7, 55, 110 spawning, 81, 89, 91
specific gravity, 13, 33
specific heat, 12, 117, 128
R
specific speed, 29–31, 33
radiographic testing, 69, 73 splash-­type fill material, 140–141
rake, 2 spray nozzle, 145–147
Rancho-­Seco, 147, 147 spray pond, 1, 116, 139, 145–148
Rankine, 9 spray pond test results, 147
158Index

spray tree, 146 tubercle, 61, 65, 67–68, 73–77, 83–84, 86, 101
stagnant flow, 62, 64 Tubular Exchange Manufacturer’s Association,
stainless steel, 3, 69, 96 137, 137
standard international, 9–10 Tuovinen, J. E., 74
stator water cooler, 111 turbine, 1, 3–4, 9, 15, 22, 27, 35, 111, 116
steam generator, 4
stop log, 2, 16, 23
U
strainer, 2–3, 23, 24–25, 36, 38, 81–82, 87–90,
97, 111 ultrasonic testing, 68, 73
submergence, 15, 24, 31 U.S. Fish and Wildlife Service, 4
suction specific speed, 30–31
sulfur-­reducing bacteria, 67, 77–78, 86
sump, 2, 16, 17, 17, 39 V
Surry, 3
vacuum pressure, 9, 10
Suspended solids, 56, 79–80, 144
vacuum relief valve, 120
Swing check valve, 57, 100
valve, 3–6, 16–23, 31, 35–36, 41, 45, 48,
system head curve, 31
56–58, 80, 97, 99–100, 110, 114, 115,
119–120, 124–125
T vapor column, 125
veil, 140–141, 143
Taborek, J., 137, 137 veligers, 24, 80–82, 89–91
temperature, 5, 9–11, 13, 19, 22, 28, 31, 35, velocity cap, 19–22
44–45, 59, 70–72, 77, 80–82, 88–89, vertical runs, 66
109–111, 113, 115–116, 116, 117–120, vertical wet-­pit, 1–2, 27, 29–31, 36, 38
122, 127–128, 131, 134, 136, 139, viscosity, 11–12, 96, 131, 135
145–147, 147 volume, 10–11, 35, 49, 62, 75, 82–83, 95,
temperature coefficient, 59 101–102, 117–119, 123
Temporary Instruction 2515/118, 7
Tennessee River, 56, 70, 106, 110
Tennessee Valley Authority, 47–50, 52–53, 55–56, W
61–62, 66–68, 70, 72–74, 78, 80–82,
84, 86–88, 99, 105–107, 109–110 wall resistance, 130
test, 3–7, 25, 37–38, 47–49, 52–56, 61, 67–70, 70, wall thinning, 61–62, 64–67
71, 74–75, 81–86, 83, 88, 91, 97, 99, water hammer, 6, 35–36, 121, 124–125
101–110, 108, 136, 146–147 waters of the United States, 19, 110, 139
thermal conductivity, 12, 130 Watts Bar, 62
thermal diffusivity, 12 Watts Bar Nuclear Plant, 47–48, 55–56, 68–69,
thermal discharge effluent limit, 139 72–73, 88–89, 91, 109, 109
thermoplastic, 44, 50 wave speed, 122–123
thermosetting, 44–45 wearing ring, 37
thermosetting epoxy resin, 46 wedgewire screens, 20–21
thin film-­type fill, 143 wet-­bulb temperature, 146–147, 147, 148
Thoma’s cavitation constant, 34 Widows Creek, 62, 101, 101, 104, 105
through-­wall leak, 7, 66, 68, 72, 78, 90
through-­wall pit, 68–69, 72 Z
total residual chloride, 81, 89
trash rack, 2, 15, 16, 18, 23, 25, 81, 91 zebra mussel, 4, 25, 80–82, 89, 91, 136
trash sluice, 16–23 zebra mussel control, 82
traveling screen, 2, 15, 16, 20–21, 23, 24–25, 89, 91 zero discharge, 139
traveling water screen, 2, 18–21, 23, 25, 91 zinc polyphosphate, 82–84, 83