Editor:

Ch. Daublebsky
von Eichhain HRSG
Heat Recovery Steam Generators
Design and Operation
2nd edition

RR

2
Titel HRSG_2.indd 5
PP PUBLICO Publications

29.07.15 13:26
 The FINEST way of water treatment!

POWER PLANTS HEATING SYSTEMS COOLING SYSTEMS WASTE WATER

ABWASSER

FINEAMIN® – inhibitors for water treatment against corrosion

and scale in your systems

FINEAMIN® is effective and protects the environment

Customer service: fast and competent service

Your partner in
water chemistry

CWB Wasserbehandlung GmbH h2o facilities sa

Glienicker Weg 95 8 av. Grandes Communes
D -12489 Berlin CH -1213 Petit Lancy | Switzerland
Tel.: +49 (0)30 67 89- 37 51 Tel.: +41 22 879 95 00
Fax: +49 (0)30 67 89-38 21 Fax: +41 22 879 95 09
info @cwb-berlin.de info @h2o-f.ch
www.fineamin.de www.fineamin.com

Titel HRSG_2.indd 4 29.07.15 13:26

U4 Heat 06.02.2003 17:39 Uhr Seite 2

Prof. Dr.-Ing. Jovan Mitrovic (Editor)

Heat Exchanger and

Condenser Tubes
Tube Types – Materials – Attributes – Machining

2004. 311 pages with numerous figures

and tables.
ISBN 3-934736-08-4.

Soft cover E 38,--

This book is the english version of the handbook

„Wärmeübertrager-Rohre“. It gives a practical
oriented and comprehensive overview concerning
the different materials and their specifics
especially refering to their applications, about the
different marks and their advantages.
Furthermore the different techniques in manufactu-
ring, surface conditioning and damage removal are describben.

Contents:
0. Introduction
1. Tube Types
1.1 Materials
1.2 Optimization with Special Forms
2. Manufacturing of Heat Exchanger Tubes
2.1 Construction/Prefabrication/Machining
2.2 Welding
2.3 Welding/Rolled Tube Joint/Expanding
3. Surface Treatment
3.1 Cathodic Protection
3.2 Pickling/Electrochemical and Chemical Polishing
3.3 Inlet Tube Lining
4. Damages/Damage Removal/Maintenance

Bestellungen an:
PP PUBLICO Publications
Witteringstr. 10 + D – 45130 Essen/Germany
Tel.: ++49(0)201/79 12 12
Fax: ++49(0)201/79 88 278
e-mail: [email protected]
www.pp-publico.de
II

HRSG
Heat Recovery Steam Generators
Design and Operations

2nd edition

Copyright© 2015 PP PUBLICO Publications. All rights reserved.

Expected as permitted under the Urheberrecht der Bundesrepublik Deutschland, no part of this
publication may be reproduced or distributed in any form or by any means, or stored in a data
base or retrieval system, without the prior written permission of the publisher.

Direct all inquiries to:

PP PUBLICO Publications,
Witteringstr. 10, • D - 45130 Essen/ Germany
Phone: +49(0)201/79 12 12
e-mail: [email protected]
www.pp-publico.de

ISBN: 3-934736-32-7
ISBN-13: 978-3-934-736-32-0
EAN: 978 3 934 736 320

Cover photos:
KED, D-München
Annotation of the publisher:
the quality of figures and tables generally depends on the material made available from the
authors. Place of jurisdiction for all matters concerning this book is Essen/Germany.
I
 III

HRSG
Heat Recovery
Steam Generators
Design and Operations

Editor: Christian Daublebsky von Eichhain

PP PUBLICO Publications
IV

Compact Heat Exchangers

Designs - Materials - Applications

2010. 288 Pages with

numerous tables and figures
ISBN 3-934736-16-5

Hard cover € 44,-

This handbook presents innovative

knowledge concerning designs, nate-
rials and applications of current and
future orientated kinds of compact heat
exchangers.

All authors are recruted from leading

scientifical institutions or apparatus
producers.

Content:

I. Foreword IV.3 Graphite

II. Apparatus Designs IV.4 Ceramics
II.1 Plate Heat Exchangers IV.6 Plastics
II.2 Plate & Shell Heat Exchangers V. Surface Technology
II.3 Spiral Heat Exchangers VI. Preventive Measures for Mitigation of
II.4 Block Heat Exchangers Fouling
II.5 Microstructure Heat Exchanger VI.1 Inspection
III. Plate structurization VI.2 Filtration/Mirco Filtration
IV. Material Technology VI.3 Chemical Conditioning
IV.1 Copper VI.4 Cleaning and Reconditioning
IV.2 Tantalum VII. Applications

PP PUBLICO Publications
Witteringstr. 10 + D – 45130 Essen/Germany

e-mail: [email protected]
www.pp-publico.de
Content V

1. Introduction 1
1.1. Abstract 2
1.1.1. Design 2
1.1.2. Operation 2
1.2. Overview 2
1.2.1. Gas turbine cycle 3
1.2.2. Rankine- Cycle 4
1.2.3. Steam turbine 5
1.2.4. Heat Recovery Steam Boiler 6
1.2.5. Combined Cycle II 7
1.2.6. Market of Heat Recovery Steam Generator 8
1.2.7. History 10
1.3. Conversion of heat to electrical power 11
1.3.1. Thermal efficiency 11
1.3.2. Electrical efficiency 13

2. Design of a HRSG 15
2.1. Over all design of a HRSG 16
2.1.1. Pressure levels 16
2.1.2. Drum type boiler vs. once through boiler 19
2.1.3. Pinch Point method 19
2.2. How to design a boiler 24
2.2.1. Design of the duct 24
2.2.2. Tube diameter, fin dimensions and tube pitches 24
2.2.3. Scaling of fins 26
2.2.4. Corrosion 26
2.2.5. Fouling 28
2.2.6. Fin efficiency and fin material 32
2.2.7. Pipe wall thickness 34
2.2.8. Header wall thickness 35
2.2.9. Drum wall thickness 35
2.2.10. Gas Side Pressure Drop 35
2.2.11. Pressure drop on water side 36
2.2.12. Natural circulation 37
2.2.13. Forced through circulation 38
2.2.14. Fin tube heat transfer 38
2.2.15. Pipe turbulent heat transfer 38
2.2.16. Pipe evaporation heat transfer 38
2.2.17. Heat conductivity of steel 38
2.2.18. Overall heat transfer 39
2.2.19. Logarithmic mean temperature 39
2.2.20. Designing of heating surfaces 40
2.2.21. Noise and vibration problems at heat exchanger 40
2.2.22. Regenerative feed water preheating vs. condensate preheating 43
2.2.23. General Remarks 44
2.2.24. Duct burner 46
VI Content

2.2.25. Ductwork and casing 48

2.2.26. Environmental considerations 48
2.2.27. Site conditions 48
2.2.28. Steaming in economizers 49
2.2.29. Important notes 50

3. Operation of steam boiler 51

3.1. Example of a start up 52
3.1.1. Gas turbine mass flow 52
3.1.2. Gas turbine temperature 52
3.1.3. HP Steam mass flow 53
3.1.4. HP Steam pressure 53
3.1.5. Gradients 54
3.2. Start up 54
3.2.1. Deaeration of economizers 54
3.2.2. Purging 54
3.3. Drain 55
3.4. Drum water level 56
3.5. Water running through the economizer 57
3.6. Start up of the gas turbine 57
3.7. Life Cycle Fatigue 58
3.8. Temperature gradients drums and headers 59
3.9. How to start up faster 62
3.10. Control system 62
3.10.1. Drum water level control 63
3.10.2. Level measurement 63
3.10.3. Swell and shrink 63
3.10.4. Single element control 64
3.10.5. Two element control 64
3.10.6. Three element control 64
3.10.7. Four element control 66
3.10.8. Pressure control 66
3.10.9. Spray cooler control 66
3.10.10. Control Methods 66
3.10.11. Ziegler-Nichols Methods Facilitate Loop Tuning 67
3.10.12. Load change 68
3.10.13. Sliding Pressure 68
3.10.14. Example of a load change with duct burner 69
3.10.15. Load change of the gas turbine 70
3.10.16. Duct burner 71
3.10.17. Shut down 72
3.10.18. Run out of turbine 72
4. Appendix I Converting factors 72
5. Appendix II Disclaimer 73
6. Literature 76
7. Contact 77
 VII

Prof. Dr.-Ing. H. Müller-Steinhagen

Dr.-Ing. H. U. Zettler (Editors)

Heat Exchanger Fouling

Mitigation and Cleaning Technologies

2nd revised and enlarged edition

2011. 470 Pages with
numerous tables and figures
ISBN 3-934736-20-3

Soft cover € 58,-

This handbook presents innovative knowledge

concerning designs, preventive measures,
maintenance services and monitoring.

All authors are recruted from leading scientifical

institutions, apparatus builders or leading main-
tenance offeres.

Content:
3.3 Chemical Fouling Mitigation and
1. Introduction Cleaning
3.4 Physical and Energetical Water
2. Heat Exchangers for Fouling Duties Conditioning
2.1 Constructional Disposion
2.2 Conditioning Disposion 4. Off-Line Cleaning Methods
4.1 Introductional remarks
3. On-Line Mitigation and Cleaning 4.2 Chemical Cleaning
Methods 4.3 Mechanical Cleaning
3.1 Introductional remarks
3.2 Mechanical Fouling Mitigation 5. Fouling Monitoring
and Cleaning

PP PUBLIC
Witterings O Publica
tion
– 45130 E s
tr. 10 + D
e-mail: pp ssen
-publico@
online.de
www.p p-publico
.de
 Editor:
Ch. Daublebsky
von Eichhain HRSG
Heat Recovery Steam Generators
Design and Operation
2nd edition

2
Titel HRSG_2.indd 1
PP PUBLICO Publications

16.07.15 10:53
 1. Introduction 1

1. Introduction

001-014 HRSG pages.indd 1 13.07.15 11:50

 2 1. Introduction

1. INTRODUCTION

1. Introduction

1.1. Abstract

This book is about the design and operating of a Heat Recovery Steam Generator
(HRSG)

1.1.1. Design

• How many pressure stages are taken and why

• How to determine the pressure of each pressure stage
• How to design the superheater, evaporator, economizer
• Tube dimension of the heating surfaces
• Fin dimension
• Tube arrangement
• Velocities of flue gas side, watersteam side and piping
• Pressure drop
• How to design a natural circulation system

1.1.2. Operation

• Start up with purging, drain, considering the temperature gradients of drum and
headers
• Start the duct burners
• Load change

1.2. Overview

Combined Cycle
The combined cycle is the combination between a gas turbine thermodynamic
cycle (Brayton- Cycle) and a steam cycle (Rankine- Cycle). The Brayton Cycle has high
source temperature and rejects heat at a temperature that is conveniently used as the
energy source for the Rankine Cycle. The most commonly used working fluids for com-
bined cycles are air and steam. Other working fluids (organic fluids, potassium vapour,
mercury vapour, and others) have been applied on a limited scale.

001-014 HRSG pages.indd 2 13.07.15 11:50

 1. Introduction 3

Fig. 1: Flow diagram of a modern HRSG

1.2.1. Gas turbine cycle

Fig. 2: Gas turbine cycle

1 - 2 : Isentropic Compression
2 - 3 : Reversible Constant Pressure Heat Addition
3 - 4 : Isentropic Expansion
4 - 1 : Reversible Constant Pressure Heat Rejection (Exhaust and Intake in the open cycle)

001-014 HRSG pages.indd 3 13.07.15 11:50

 4 1. Introduction

Fig. 3:
Enthalpy – Entropy (h-s) diagram
of a gas turbine cycle

1.2.2. Rankine- Cycle

Fig. 4: Flow diagram and Temperature – Entropy (T-s) diagram of a Rankine Cycle

1- 2 Feed Water Pump

2- 3 Economizer – Evaporator – Superheater
3- 4 High pressure turbine
4- 5 Reheater
5- 6 Low pressure turbine

001-014 HRSG pages.indd 4 13.07.15 11:50

 1. Introduction 5

6- 1 Condensor

Fig. 5: Pressure – Volume diagram of a Rankine Cycle

4-1 Feed Water Pump

1-2 Economizer – Evaporator – Superheater
2-3 High pressure turbine
3-4 Condenser

1.2.3. Steam turbine

In the steam turbine the transferred heat from flue gas of gas turbine to the water –
steam of the HRSG is converted to mechanical power.

001-014 HRSG pages.indd 5 13.07.15 11:50

 6 1. Introduction

Fig. 6:
Steam Turbine

1.2.4. Heat Recovery Steam Boiler

Fig. 7: Cross
section of a
modern triple
pressure HRSG

1. Inlet with inside insulation covered by stainless steel liner panels.

2. High pressure superheater (HP).
3. Reheater section (RH).
4. Gas or distillate oil fueled duct burner
5. High pressure boiler section and required downcomer piping.
6. High pressure steam drum with internals to meet steam purity requirements.
7. Carbon monoxide (CO) converter and selective catalytic reduction (SCR) System.
8. Intermediate pressure (IP) superheater section.

001-014 HRSG pages.indd 6 13.07.15 11:50

 1. Introduction 7

9. High pressure economizer section.

10. Intermediate pressure boiler section and required downcomer piping.
11. Intermediate pressure economizer section.
12. Low pressure boiler section with downcomer piping.
13. Carbon steel or stainless steel condensate preheater section.
14. Intermediate pressure steam drum.
15. Low pressure (LP) steam drum with internals adapted for integral deaerator
arrangement.
16. Deaerator tank with required pegging steam and equalizer lines.
17. Outlet stack with required environmental monitoring connections and test Ports.
18. Access platforms, ladders and stairway

1.2.5. Combined Cycle II

Fig. 8: Flow diagram and Temperature- Entropy (T-s) Diagram of a combined cycle

001-014 HRSG pages.indd 7 13.07.15 11:50

 8 1. Introduction

1.2.6. Market of Heat Recovery Steam Generator

1.2.6.1. Market survey GT Europe

According Gas Turbine World 2003 Handbook

Foster Wheeler is not mentioned in this survey.

Fig. 9: Market shares of gas turbine OEMs

1.2.6.2. Prices of CCPP economics of scale

With increasing capacity the prices per kW drops significantly.

Until ca. 450MW installed capacity the size of gas turbine and steam turbine is increas-
ing then the economics of scale is much lower because then there are more gas tur-
bines and HRSGs required, the size of steam turbines can get bigger.

001-014 HRSG pages.indd 8 13.07.15 11:50

 1. Introduction 9

Fig. 10: Prices of Combined Cycle power plants

1.2.6.3. Market CCPP

The market for Combined Cycle Power Plants (CCPP) has experienced rapid growth in
the last years. This growth has been driven by different reasons:
• Deregulation in the U.S. and Europe.
• Low prizes
• Fast to build up (in some cases the gas turbine is installed very quickly and the
HRSG is installed later)
• Rather low fuel cost of natural gas
• Less problem with environmental requirements
• Very good cycling behaviour

Due to this, e.g. independent power producers (IPP) rose and have induced both, a
growth in new power production and a shift from coal and solid-fuel-fired conventional
steam plants to gas turbine (GT) plants and CCPP leading to economically interesting
returns of investment (ROI).
In the U.S. alone, while gas turbine and combined cycle plants represent only 10% of
the existing base of 860 GW, they currently provide well over 90% of all new capacity
[Got1].
In Europe the markets seem to hesitate. Until now deregulation has taken place in some
countries only, e.g. the U.K., but is on the way for the rest of the EU.

001-014 HRSG pages.indd 9 13.07.15 11:50

 10 1. Introduction

Expectations in Europe are rather for a consistent growth, than a boost like that in the
U.S., which is unlikely, due to the fact that governmental responsibilities for sufficient
and reliable power generation in the past led to capacities above the actual needs. Even
though these plants, mostly fossil fired, need replacement in the coming one or two
decades. In addition coal fired plants, in several European countries, serve great public
economic benefits as a result of large own resources. Same applies to hydro power, e.g.
Norway, Austria and Switzerland. Nevertheless growth expected in Europe – selecting
France, Germany, Spain, Sweden, UK and Finland – as an average number, is 70 GW
for new capacity until 2005 [FTE1]

The deregulation-driven growth is expected to fall off in North America, while at the
same time, combined cycle power plants will support continued HRSG growth in the
recovering Asia market. Another key driver is the aggressive technical development of
large frame combustion turbines (170 to 250 MW, even 370 MW in a test stage) tar-
geted for the utility power generation market. Over the last decade, large combustion
turbines have been developed with higher efficiency and dramatically improved emis-
sions profiles. More efficient water/steam cycles have been developed to take advan-
tage of higher exhaust temperatures from advanced combustion turbines installed in
combined cycles. Capital costs of gas fired combined cycle are about 40% of coal fired
steam plants [Got1]. Gas price and availability support a life cycle cost advantage in
many regions of North America and Europe. The net efficiency of the combined cycle
power plant (up to 60% – expected in the near future, at the time being 58% for high end
CCPP’s) is much higher than – with conventional steam plants (typically 35% to 40%,
up to 50% for high end plants). Combined cycle plants also continue to offer improve-
ments in permitting and Installation time thereby reducing the capital cost and risk to
plant developers. Combined cycle plants are able to provide lowest levels of NOX and
CO emissions per kWh of electricity produced, especially if low NOX burners and SCR,
CO catalysts are considered.

This all results in a necessary development in HRSG technology, as well as a new un-
derstanding of the HRSG supplier delivering a less priced, though key component of a
plant gaining more and more shares in power generation and economic success of the
owner.

1.2.7. History

To efficiently mate the Rankine steam cycle with high-temperature gas turbines, new
HRSGs had to be developed that could operate at substantially higher flue- gas temper-
atures. New HRSG designs also were required to match each incremental jump in gas-
turbine size as combined cycle units grew larger and larger. Perhaps the most important
development in HRSG design was the move from single- to dual- pressure steam pro-
duction. This change, which enabled lower stack temperatures and thus greater recov-
ery of thermal energy from the gas-turbine ex-haust, increased thermal efficiency of a
combined-cycle plant by nearly four percentage points. Later designs went one step

001-014 HRSG pages.indd 10 13.07.15 11:50

 1. Introduction 11

further, from dual- to triple-pressure steam production, and yielded approximately one
more percentage point gain for the overall cycle. Today, virtually all HRSG manufactur-
ers offer triple pressure reheat steam systems to maximize efficiency [Swa1].

1.3. Conversion of heat to electrical power

The main purpose of a HRSG is to convert the hot flue gas of the gas turbine to electri-
cal power. In some cases the HRSG converts a part of the input energy in district heat-
ing.

The thermal efficiency of the HRSG is rather low, according EN 12952- 15 based on
higher heating value (HHV) or ASME PTC it is about 70% - 77%.

According EN 12952- 15 based on lower heating value (LHV) or DIN 1942 it is 80%
-88%. Direct fired steam generators has efficiencies up to 95% based on LHV. The low-
er thermal efficiency of the HRSG is caused by the rather low input flue gas temperature
and the big flue gas mass flows causes high stack losses.

Fig. 11:
Sankey energy
diagram of a HRSG

1.3.1. Thermal efficiency

There are different methods to calculate the thermal efficiency.

The thermal efficiency is defined:

001-014 HRSG pages.indd 11 13.07.15 11:50

 12 1. Introduction

1.3.1.1. Input – Output Method

η = Useful_heat
Input_heat

1.3.1.2. Heat loss Method

η = 1 – Heat_Losses
Input_heat

Because of

Input_heat = Useful_heat + Heat_Losses

both methods must lead to the same results.
Which method is used for testing the efficiency depends on the meas-urement data. For
example it is not so easy to get the radiation and con-vection losses, so it is better to
calculate the efficiency according the In-put- Output – Method.
One hint: The blow down is not a loss, it is a part of the useful heat.

The thermal efficiency of the HRSG doesn’t give an answer how much electrical power
the steam can produce.
It is possible to have a boiler with a very high thermal efficiency and the electrical effi-
ciency is very low.
For converting heat in electrical power very often hot steam with high pressure is used.
A turbine converts the hot steam with high pressure in mechanical power according
Newton’s second law:

W mech = m• (ν Steam_in – ν Steam_out ) ν Turbine_Blade

W mech = m• (ν Steam_in – ν Steam_out ) •S Axel-Turbine_Blade

• rev Turbine

S Axel-Turbine_Blade distance turbine axle to middle of turbine blade [m]

revolution of turbine per second (normally US: 60 1/s [Hertz]
rev Turbine Europe 50 1/s [Hertz])

001-014 HRSG pages.indd 12 13.07.15 11:50

 1. Introduction 13

Fig. 12: Velocity triangles of a steam turbine

In a nozzle the hot steam with high pressure will be expanded and accelerated to the
velocity v Steam_in (turbine Inlet).

In the turbine the steam will be decelerated to v Steam_out and the turbine produces the
mechanical power out of the velocity differences.
Enthalpy and velocity has a close connection:

h in – h out = v out – v in
2 2

2 2
The mechanical power of the turbine is converted in the generator in electrical energy.
The efficiency of converting mechanical power in elec-trical power in a generator is
rather high (about 98%). But even if the losses are rather low, the generators must be
cooled (a 1000 MW gen-erator has losses of about 20 MW!) by hydrogen or water. In
some cases the generator is cooled by air.

1.3.2. Electrical efficiency

The electrical efficiency is much lower.

Electrical efficiency is defined

ηel = electrical_Power
Input_heat

001-014 HRSG pages.indd 13 13.07.15 11:50

 14 1. Introduction

001-014 HRSG pages.indd 14 13.07.15 11:50

 2. Design of a HRSG 15

2. Design of a HRSG

015-050 HRSG pages.indd 15 16.07.15 10:51

 16 2. Design of a HRSG

2. Design of a HRSG

2.1. Over all design of a HRSG

2.1.1. Pressure levels

Why to make different pressure levels ?

The steam turbine works with high velocities (about 985ft/s [300 m/s]). The steam tur-
bine “takes” the energy of the steam, so if the steam transferred too much energy, the
steam starts to condense to water. Thus the local stress (= compressive stress = pres-
sure) on a turbine blade increases dramatically and may destroy it.
The stress is:

• •
F m (νin – νout ) Vρ (νin – νout ) νin Aρ (νin – νout )
σ lokal = = = =
A A A A

σ lokal = νin ρ (νin – νout )

The difference of density of water and steam is the difference of local compressive
stress. The density of water is more than 1,000. times higher than of steam (in low pres-
sure stages up to 50,000. times higher)

The “reheating” of the steam can prevent, that there is too much water in the steam. So
the reheating can avoid erosion of the turbine blades and of course increases the per-
formance. If the steam has a high enough pressure, nearly all the energy transferred to
the reheat steam can be recovered by the turbine (multiplied with the turbine efficiency
i.e. ca. 85%).
So another very important advantage of the reheating is, that the efficiency of the ther-
modynamic process is increasing dramatically. So introducing multiple pressure stages
minimize the “exergy” losses. The exergy it this part of the input energy that can’t be
transformed to mechanical engergy.
The minimum of the exergy losses in the HRSG is, if the heating of the working fluid (in
this case Water) has a minimum temperature difference to the cooling of the other (hot)
fluid (flue gas of the gas turbine).

Increasing efficiency

There are 3 main ways to decrease the temperature differences between flue gas and
water:
1. Multiple pressure stages
2. Once through boiler
3. Binary fluids (e.g. H2O – NH3 Kalina process)

015-050 HRSG pages.indd 16 16.07.15 10:51

 2. Design of a HRSG 17

You can see the temperature difference in the QT diagram

Fig. 13: Temperature – Transferred Heat (T-Q) Diagram triple pressure diagram

Fig. 14: Temperature – Transferred Heat (T-Q) Diagram dual pressure diagram

015-050 HRSG pages.indd 17 16.07.15 10:51

 18 2. Design of a HRSG

Fig. 15: Temperature – Transferred Heat (T-Q) Diagram single

pressure diagram

Fig. 16: T-Q Diagram diagram theoretical ideal steam generator

015-050 HRSG pages.indd 18 16.07.15 10:51

 2. Design of a HRSG 19

2.1.2. Drum type boiler vs. once through boiler

Advantages drum type boiler

• Easy to control
• More safety and longer possible feed water stop (i.e. switch feed water pump)
because of the much higher water mass in drum and circulation system

Advantages of once through boiler

• Faster reaction of load change (less water mass and steel mass)
• Faster to start up, no drum preheating

Disadvantage of once through boiler

• Maldistribution of water in the pipe
• Can cause gas side temperature streams
• Very expensive water treatment necessary
• Very fast reaction to the changing of heat input, because of this the control system
must be very fast, reliable and sophisticated.

Fig. 17: Once through HRSG

[Fran1]

2.1.3. Pinch Point method

2.1.3.1. Pinch Point

The pinch point is defined as the difference between the gas temperature exiting the
last evaporator section and the saturation temperature in that drum. That means with a
lower pinch point more steam is produced at that pressure stage.

015-050 HRSG pages.indd 19 16.07.15 10:51

 20 2. Design of a HRSG

2.1.3.2. Approach Point

The approach point is defined as the temperature difference between saturation tem-
perature in the drum and economizer outlet temperature. If the approach point is de-
creased, less steam must be condensed to preheat the economizer outlet water to
saturated temperature.

Fig. 18: Pinch Point and Approach Point in a T- Q Diagram

The pinch point and the approach point have a big influence to the steam flow, if it is as-
sumed that the other parameter are fixed e.g. gas turbine flue gas flow and temperature,
superheater steam temperature and pressure, feed water temperature etc.
To decrease the pinch point it is normally necessary to increase the transferred heating
power. That means often to increase the heating surface or the gas side pressure drop.

So there is a search for the optimum with higher efficiency and lower costs.

After the decision how many pressure stages there should be, the pressures of each
pressure stages can be determined:
First of all: The temperature of the HP Steam an RH Steam must be defined. Some
small gas turbines don’t produce flue gas with high temperatures (lower than 930 F
[500°C]), so the HP Steam temperature is determined as flue gas temperature minus
ca. 18 F [10°C] (There must be always a temperature difference to transfer heating pow-
er. The lower the temperature difference the bigger must be the heating surface area) If
the gas turbine produces higher temperatures the superheating temperature is a ques-
tion of the pressure and tube material. The higher the temperature the lower should be
the pressure and the more expensive is the material.
The key components, whose performance is critical, are high-pressure steam piping,
headers, and super heater tubing. All these components have to meet creep strength
requirements, but thermal fatigue resistance and weldability are important, too. Ferritic-
martensitic steels are preferred because of their lower coefficient of thermal expansion
and higher thermal conductivity compared to austenitic steels.

015-050 HRSG pages.indd 20 16.07.15 10:51

 2. Design of a HRSG 21

Among the 9% Cr steels fully commercialised, the P91 steel has the highest allow-
able stress and has been extensively used all over the world as a material for head-
ers, steam pipes and superheater tubes operating at steam temperatures up to 1103 F
[595°C] - nominal, 1139F [615°C] as a maximum for HRSG applications according to
the German TRD Code or up to 1202F [650°C] tube metal temperature according to
ASME. The steel P-92, developed by substituting part of the Mo in P-91 by W, has even
higher allowable stress values and can be operated up to steam temperatures of 1175F
[635°C]. P-92 is already approved by the ASME boiler code, but no approval according
to the German rules is available for the time being. Further developments are E-911,
which is already approved in Germany (material number 1.4905) and P122, which was
developed in Japan and has been approved by ASME. The allowable creep strength of
these new steels at 1112 F [600°C] is about 25% higher than that of P-91 [Vis1]. As an
example for application, a super heater made of E-911 and steam loops made of E-911
and P-92 are operating at steam temperatures of 1202 F [650°C] in the conventional
fired power station of RWE in Germany. Therefore it must be remarked that the limit-
ing factor for efficiency increasing high steam temperatures is the high end steam tur-
bine, which is commercially available for steam temperatures at a maximum of 1049 F
[565°C], only [Nes1].

With the material of the superheaters, reheaters and headers respectively the live steam
a reheat steam temperature is fixed.
The condensate pressure should also be known (e.g. an air cooled condenser has an
higher pressure than an sea cooled or river cooled condenser (ca. 0.75 PSI [0,05171
bar]))

Then there must be the maximum water content in steam (ca. 5% - 10% mass fraction
water in the steam (= 95% - 90% steam content)) defined and the efficiency of the tur-
bine (The data is normally received of the turbine manufactory).
So the end- point of the graph in the h- s (enthalpy – entropy) can be determinate (see
end point 1 in picture). In a computer calculation the enthalpy (h) and entropy (s) of
steam water mix is a function of the pressure and water content h(p,x) s (p,x).

Then determinate the enthalpy differences between this point and the point with the
same entropy and the superheating or reheating temperature respectively. Divide the
enthalpy difference with the efficiency of the turbine and search for points with the same
entropy with the condenser pressure, the SH or RH temperature and the enthalpy dif-
ference (see example). So the start point for expansion is fixed too.

015-050 HRSG pages.indd 21 16.07.15 10:51

 22 2. Design of a HRSG

Fig. 19: Turbine Expansion in a h- s diagram

015-050 HRSG pages.indd 22 16.07.15 10:51

 2. Design of a HRSG 23

Fig. 20: Triple pressure turbine expansion in a h- s diagram

015-050 HRSG pages.indd 23 16.07.15 10:51

 24 2. Design of a HRSG

2.2. How to design a boiler

2.2.1. Design of the duct

The flue gas duct of the boiler should be longer along the pipes than across the pipes,
because with a smaller width of the boiler, less pipes must be welded in the headers.
The ratio can be 3 to 4 times along the tubes to the width.

Fig. 21: Cross section of a HRSG

The first dimension of the duct must be guessed and during an iterative calculation
adapted. With the length of the duct the fine tuning of gas side pressure drop and the
heating surface area can be made very easy, e.g. 10% length of the duct means 10%
more heating surface and 17% decrease of pressure drop.

2.2.2. Tube diameter, fin dimensions and tube pitches

It must be decided which outer diameter and fin height should be used.
The geometry effect is the apparent anomaly in heat transfer surface between various
vendors for the same performance. As an example, a vendor with 2.0" [51mm]OD tubes
may propose 25 % more surface than the competitor who uses 1.5" [38mm] OD tube.
This does not mean that the lower surface is the result of high technology heat transfer
equipment design. This happens simply because of the nature of heat transfer itself.
Lower diameter tubes give the same amount of heat absorption with less surface. Simi-
lar anomalies exist for other geometry parameters such as fin type, fin geometry, tube
length etc. For this reason it is not prudent to eliminate designs which may have too
much or too little surfaces.

015-050 HRSG pages.indd 24 16.07.15 10:51

 2. Design of a HRSG 25

Transversal pitch is recommended, cause higher pressure drop but higher heat transfer,
lower number of rows in flue gas flow direction necessary.

Typical values are

Parameter Traditional What is used

value today
Tube OD inch [mm] 2.0 [50.8] 1.25 [31.8]
Fin type Solid Serrated
Fins/inch [Fin/m] 5 [200] 0,1 – 8 [10-315]
Arrangement Inline staggered
Tube pitches inch [mm] 3 – 6 [76-152] 2.5 [63]
Fin height inch [mm] 0.75 [19.1] 1.00 [25.4]

Tab. 1: Typical design values of HRSGs

Out of the outer diameter and fin height the transversal and longitudinal tube spacing
can be calculated. It is recommended to have distances between fin tips of about 0.5
inch – 0.25 inch [12.7mm – 6.4mm].

Fig. 22: Serrated fins

The fin density can be chosen between 0.5 fins/in [20 fins / m] and 7.5 fins/in [300 fins
/ m] [Brü1] depended on the needed heat transfer, maximum flue gas velocities and
pressure drop.
There are different methods to manufacture the finings on the tubes. A very dearly
(close) mounting with a continuous welding (very seldom soldering) is recommended.
There should be no spot welding. During the whole live there shouldn’t be any mechani-
cal or pitting corrosion dismantling. If there is only a tiny gap between fins and tube, the
fins don’t transfer any heat and can start scaling.

015-050 HRSG pages.indd 25 16.07.15 10:51

 26 2. Design of a HRSG

2.2.3. Scaling of fins

The fin tips has a much higher temperature than the pipe wall. So there is a danger of
scaling of fins. The fin tip temperature can be calculated by hand or computer program.

Fig. 23: Fin efficiency diagram

T Fin-Tip = T Wall +θ (T FlueGas – T Wall )

Carbon steel fins can have fin tip temperatures up to 1112 F [600°C] [Berg1], if there is
no chlorine, vanadium, very less sulphur and sodium in the flue gas.

2.2.4. Corrosion
Potential problem areas as a result of load cycling or on-off cycling include: gas turbine
exhaust dew point corrosion, corrosion fatigue, and consequences of not maintaining
proper steam cycle chemistry (i.e., on-line, off-line storage and return to service). Cor-
rosion and fatigue damage are cumulative and can not be reversed. Using HRSG ini-
tially designed for base load operation in cycling operation defines the need to carefully
evaluate several occurrences with regard to HRSGs. Special attention has to be paid to
three of them at least:

2.2.4.1. Stress Corrosion Fatigue

Since cycling means temperature and pressure gradients from ambient to operational
level and air ingress during longer outages, stress corrosion fatigue as a result of these
influences will occur. A proper chemistry regime, i.e. maintaining low dissolved oxygen,
pH within the required range and proper feed water quality (VGB, O2 < 0,1 mg/kg), is a
must. From the HRSG operating side, the boiler should be kept under pressure as long
as possible, e.g. no forced cooling and closing of the stack damper to prevent rapid
natural draft cooling.

2.2.4.2. Flow Accelerated Corrosion

First, the HRSG designer has to consider flow velocities lower than the known limits to
dissolve protective Magnetite layers in water and/or lines carrying two phases, water

015-050 HRSG pages.indd 26 16.07.15 10:51

 2. Design of a HRSG 27

and steam. Second, the chemistry regime has to be maintained in a way that the Oxy-
gen content is not too low to prevent a proper magnetite layer from forming - erosion
corrosion is increasing - and on the other hand not too high to accelerate Stress Corro-
sion Fatigue. In Europe this has been taken into account by the increased maximum O2
content (VGB, TRD, etc.) for boiler feed water (from 0,02 mg/kg to 0,1 mg/kg for pH >
9). Best choose is not to fall below 0,05 mg/kg (VGB minimum for pH neutral feed wa-
ter) considering the above.

2.2.4.3. Gas Side Corrosion

Cold end corrosion is a well known phenomenon. It can be prevented by increasing the
water inlet temperature, e.g. condensate recirculation, above the dew point of the flue
gases. Cycling leads to a situation at each start up, when the inlet temperature can not
be properly increased - deposits on the cold end of the HRSG surfaces are the conse-
quence. This results in decrease of thermal efficiency and increase of draft losses at the
long term, fin and tube corrosion, if the deposits are moistened - by air humidity or wash-
ing. To prevent or limit the effect of cold end corrosion during cycling Operation, regular
inspections and cleaning of the boiler surfaces is recommended. This is usually done
by air blasting (little deposits), dry ice blasting (up to 6 layers affected) and washing with
large amounts of low pressure water (entire surfaces). The water washing is the most
effective, although special considerations have to be made and actions set to prevent
corrosion of the casing (horizontal type HRSG) or poisoning a catalyst (vertical type
HRSG). Start up after performing water washing is recommended to prevent corrosion
of other HRSG parts. The ultimate solution to cold end corrosion is the use of corrosion
resistant materials - the only reliable and lasting but expensive solution.

Fig. 24: Sulphur acid dew point diagram

015-050 HRSG pages.indd 27 16.07.15 10:51

 28 2. Design of a HRSG

Dew point of sulphuric acid as a function of partial pressure of sulphuric trioxide and
water vapour

T_S = (A – B • ln ( p H2O) – C • ln ( p H2SO4 ) + D • ln ( p H2SO4) • ln ( p H2O))-1

A = 2.988 • 10-3 • K-1

B = 5.97 • 10-5 • K-1
C = 1.161 • 10-4 • K-1
D = 6.2 • 10-6 • K-1

P H2O = x H2O p
P H2SO4 = x H2SO4 p

[Ver1]

Unfortunately the sulphuric trioxide content in flue gas is not known. Normally it is as-
sumed that the converting rate form SO2 to SO3 is up to 5% [Gan2] [Ras1] but other
articles say, it can be up to 50% [Wic1].

2.2.5. Fouling

Outside Fouling Factors Minimum Fin Spacings

Fuel hr ft² F/Btu [m²K/kW] in [mm]

Dry Air 0.000 - 0.001[0.000-0.176] 0.05 [1.27]

Natural Gas 0.001 - 0.003[0.176-0.528] 0.07 [1.78]
Propane 0.001 - 0.003[0.176-0.528] 0.07 [1.78]
Butane 0.001 - 0.003[0.176-0.528] 0.07 [1.78]
No. 2 Fuel Oil 0.002 - 0.004[0.352-0.704] 0.12 [3.05]
No. 6 Fuel Oil 0.003 - 0.007[0.528-1.233] 0.18 [4.57]
Crude Oil 0.008 - 0.015[1.409-2.642] 0.20 [5.08]
Residual Oil 0.010 - 0.030[1.761-5.283] 0.20 [5.08]
Coal 0.010 - 0.050[1.761-8.805] 0.34 [8.64]
Wood Wastes 0.010 - 0.050[1.761-8.805] 0.34 [8.64]

Tab. 2: Fouling and fin spacing as a function of fuel

015-050 HRSG pages.indd 28 16.07.15 10:51

 2. Design of a HRSG 29

TEMA fouling resistances for cooling water (hr ft² F/Btu [m²K/kW])

Type of cooling water Fouling resistance

Seawater (Tout < 113 F [45°C]) 0.001- 0.002 [0.18 – 0.35]

Brackish water (Tout < 113 F [45°C]) 0.002-0.003 [0.35 – 0.53]
Treated cooling tower water 0.002-0,003 [0.18 – 0.53]
(Tout < 122 F [50°C])
Treated recirculated water 0.002 [0.18]
Fluvial water 0.002-0.003 [0.35 – 0.53]
Engine cooling water 0.001 [0.18]
Distilled water or condensate 0.0005 – 0.001 [0.09 – 0.18]
Treated boiler feedwater 0.0005 [0.09]
Boiler blowdown 0.002-0.003 [0.35 – 0.53]

Tab. 3: Fouling as a function of water type

Fouling resistance in heat transfer from gaseous combustion products to finned heat
transfer surfaces(Wei[1])

Fuel Fouling resistance Flow velocity ft/s

hr ft² F/Btu [m²K/kW] [m/s]
Natural gas 0.0005-0.003 [0.09-0.53] 98 –131 [30 – 40]
Propane 0.001-0.003 [0.18-0.53]
Butane 0.001-0.003 [0.18-0.53]
Clean turbine gas 0.001 [0.18]
Moderately clean turbine 0.0015-0.003 [0.27-0.5] 82 – 98 [25 – 30]
gas 0.002-0.004 [0.36-0.7]
Light fuel oil 0.003 [0.53]
Diesel 0.003-0.007 [0.53-1.24] 59 – 79 [18 – 24]
Heavy fuel oil 0.004-0.015 [0.7-2.7]
Crude oil 0.005-0.050 [0.89-8.85] 49 – 69 [15 – 21]
Coal

Tab. 4: Fouling and maximum flue gas velocity as a function of fuel

015-050 HRSG pages.indd 29 16.07.15 10:51

 30 2. Design of a HRSG

Fouling through evaporating liquids

Fouling problems in evaporators are caused by corrosion or local concentration or pre-
cipitation of components with a lower vapour pressure than that of the carrier liquid. In
situ corrosion of heated surfaces presents much less problems than the deposition of
products of corrosion formed upstream [Som1], [Goo1].
Since fouling is furthered by bubble formation and severely affects the high hest transfer
coefficients normally encountered in evaporation, very strict codes apply the purity of
boiler feed water. The values recommended in 1975 by the ASME Research Committee
for Water in Thermal Power Stations for operating cycles of one year [Sim1] are listed in
tab. 5. For Germany the values are given in tab. 6 [VGB1].
Guide values for boiler feed water

Pressure Iron Copper SiO2 Hardness Alkalinity Conductivity

PSI [bar] ppm ppm ppm ppm CaCO3 1/(μΩ in)[μS/cm]
0-290 [0–20] 0.100 0.050 150 0.300 700 1.78 [0.7]
290-435 [20–30] 0.050 0.025 90 0.300 600 1.52 [0.6]
435-580[30–40] 0.030 0.020 40 0.200 500 1.27 [0.5]
580-725 [40–50] 0.025 0.020 30 0.200 400 1.02 [0.4]
725-870 [50 – 60] 0.020 0.015 20 0.100 300 0.76 [0.3]
870-1015 [60–70] 0.020 0.015 8 0.050 200 0.51 [0.2]
1015-1450 [70–100] 0.010 0.010 2 0.000 0 0.038 [0.015]
1450-2031 [100–140] 0.010 0.010 1 0.000 0 0.025 [0.01]

Tab. 5: Feed water requirements

015-050 HRSG pages.indd 30 16.07.15 10:51

 2. Design of a HRSG 31

a. Boiler feed water

Natural-circulation Forced circulation boilers
boilers
<20 bar < 40 bar < 65 bar >80 bar1
General demands Clear and colourless
Oxygen 0.03 ppm max; in continous operation < 0.02 ppm
PH at 20°C 7 – 9.5
SiO2 < 0.02 ppm2
Hardness3 n.d.4 < 1 ppm < 0.5 ppm n.n.
Total iron < 0.02 ppm If possible < 0.05 < 0.03
Copper < 0.003 ppm < 0.01 < 0.005
Total CO2 < 1 ppm if possible < 20 <1
Conductivity < 0.2 μS/cm < 0.3
Permanganate if poss. < 5 ppm if possible < 10 <5
Oil < 0.3 ppm if possible < 1 < 0.5
b. Boiler water
Pressure5 bar 20 40 65 80 125 160
p-Value6 ppm < 500 < 300 < 150 < 50 < 15 <5
SiO2 ppm < 70 + 7 p < 30 + 3p < 10 < 4 < 1.2 < 0.4
Phosphates ppm < 25
7 < 10 < 10 < 3 < 3 <3
Conductivity μS/cm < 8000 < 5000 < 2500 < 1500 < 250 < 50
Density Bc < 0.4 < 0.25
Tab. 6: Feed water requirements

German feed water specifications for water-tube boilers

1 If the local heat flux > 230000 W/m2 the guide values for pressures > 80 bar must be taken.
2 This value applies only if there is no blowdown. Otherwise, the only values to observe are those
for boiler feedwater.
3 mg CaCO2/l
4 n.d. = not detectable
5 If the local flux > 230000 W/m2 the guide values for 160 bar are recommended for all pressure
stages
6 The alkalinity is obtained from the cm3 of N / 10 hydrochloric acid consumed in titration With
phenol-phthalein as indicator. If the pressure is higher than 60 bar, alkali hydroxides should be
added.
7 Can be left out completely if sudden changes in hardness can be reliably avoided

015-050 HRSG pages.indd 31 16.07.15 10:51

 32 2. Design of a HRSG

Fig. 25: Rise in temperature of heated surfaces in evaporators due to depositions of Fe2O3

The rise in temperature of an evaporator wall that results from magnetite (Fe3O4) scale
on the heated surfaces was measured by MacBeth [Mac1], [Mac2], [Mac3] and results
are shown in fig. 25.
Since the heat fluxes transferred in conventional steam generators do not significantly
exceed 317,000. Btu/(hr ft²) [1000 kW/m²], rises in temperature higher than 18 F [10 K]
ought to occur. MacBeth also reported [Mac1] that magnetite deposits reduce the critical
heat flux by 5% - 10% and increased the frictional pressure drop by as much as 50%.

2.2.6. Fin efficiency and fin material

The fin height should be defined in such a way that it makes sense to use fins in the first
place. The efficiency of the fins drops with growing fin heights, because an ever larger
temperature difference is needed which consequently leads to higher fin tip tempera-
tures (for a fin efficiency of 0%, the temperature of the fin tip equals the medium on the
gas side).

The fin efficiency is calculated as follows :

tf
l f - fin height [ft (m)]
b=lf +
2
t f - fin thickness [ft (m)]

015-050 HRSG pages.indd 32 16.07.15 10:51

 2. Design of a HRSG 33

√ 2ho ( t f + w s ) ho - outside heat transfer coefficient

m=
kf•tf ws
[Btu/(hr ft² F) (W/m² K)]

X = tanh (mb) ws - serration (segment) width[ft (m)]

mb
E = X(0,9 + 0,1X) k f - fin thermal conductivity [Btu/(hr ft F) (W/m K)]

Obviously, the coefficient of thermal conduction greatly influences the heat transfer ef-
ficiency and changing to austenitic steel grades should be considered very carefully.

Fig. 26: Fin efficiency as a function of fin height

A very efficient way to increase the heating surface is to increase the number of fins
per meter. Until today, because of fabrication and technical constraints, the maximum
number of fins for a fin thickness of 0.039 in [1 mm] was limited to approx. 88 fins/ft [290
fins/m] .

015-050 HRSG pages.indd 33 16.07.15 10:51

 34 2. Design of a HRSG

Fig. 27: Serrated fins

2.2.7. Pipe wall thickness

See ASME UG 27

Tab. 7: Stress calculation according ASME

015-050 HRSG pages.indd 34 16.07.15 10:51

 2. Design of a HRSG 35

Pipe wall thickness

d
P•
tP = 2
SE1 – 0.6P

2.2.8. Header wall thickness

See ASME UG 27 or EN 12952

2.2.9. Drum wall thickness

See ASME UG 27 or EN 12953

2.2.10. Gas Side Pressure Drop

Pressure drop flue gas side is a direct loss in electrical power

P = • ∆ p VηGT exp • 0.1 [kW]

∆ p [mbar]

V [m³/s]

P = • ∆ p VηGT exp • 0.000117346 [kW]

∆ p [in of Water]

V [ft³/min]

Example
1,080,898. [ft³/min] GT volume flow
15.781 [inch of Water] HRSG pressure Drop
70% [-] efficiency of GT
P loss = 1,401. kW

Gas side pressure drop has a role in determining the surface area but its significance
is limited. Previously it was normal to have at least 0.5 decrease in efficiency of gas
turbine for every inch of gas side pressure drop in the HRSG. Today’s advanced gas
turbines have reduced this by about 25%. The Optimum design seems to be between
10- 14 inwc [25-35 mbar], depending of the numbers of pressure stages and the kind
of boiler. Lower pressure drops increase the area requirements rapidly but at higher al-
lowable pressure drops the area decrease is not very large. The gas velocity changes
with the square root of the pressure drop. Hence high pressure drop results in moderate
velocity and heat transfer increase, because the heat transfer is a bit lower than linear
with the velocity. Consequently, the surface area reductions are small. So the reason

015-050 HRSG pages.indd 35 16.07.15 10:51

 36 2. Design of a HRSG

for keeping the pressure drop in the 10-14 inwc [25 - 35 mbar] region is that this velocity
region is the most economical. On the other hand, very high pressure drop would lead to
higher gas velocities, which may be detrimental to the integrity of the outside surfaces.

2.2.11. Pressure drop on water side

The pressure drop on the water side should be not too low because of bad mass flow
distribution. For example, if the final superheater has a too low pressure drop, it is pos-
sible, that due to an small additional resistance in one single pipe (e.g. sharp edge in a
hole of the header because of not complete drilling) there is too low steam flow in that
pipe, so the pipe is not cooled enough and could cause damages. If the pressure drop
in the pipes is high enough, the influence of an additional resistance is not as big. An-
other reason is, that high pressure drop means higher heat transfer coefficient in the
pipe. Inside heat transfer has much bigger influence to the overall heat transfer at heat-
ing surfaces with fined tubes than heating surfaces with bare tubes.
Also there should be not too high pressure drops on the water / steam side: Too high
velocities can cause damages:
• Erosions corrosion
• Flow accelerated corrosion (FAC)

Some hints for velocities:

Unit Value Reason
HP Superheated steam velocities ft/s 230 Sound,
[m/s] [70] economics
HP Saturated steam velocities ft/s 66 Erosion
[m/s] [20] Corrosion,

Two Phase velocities ft/s 33 FAC, EC,

[m/s] [10] economics

Water Velocities ft/s 6-13 FAC, EC,

[m/s] [2-4] economics
Tab. 8: Recommended velocities in tubes
Like for the pressure drop on the flue gas side, the pressure drop of the water side must
be compensated from the feed water pump. This also causes directly loss of electrical
power:

Example:
• 357,149. lb/hr [162. t/hr]
• 100 PSI [6.89 bar]pressure drop pumped by the feed water pump
• 219 F [104°C]
• 70% efficiency
need 46 kW electrical power.

015-050 HRSG pages.indd 36 16.07.15 10:51

 2. Design of a HRSG 37

The higher the pressure drop in the superheater, the higher is the pressure in the drum
and the lower is the steam flow.
An other reason not to have too high pressure drops is, that the drum pressure would
increase too much and that would mean higher wall thickness lower cycling gradients;
The friction pressure drop in natural circulation system must be very low to have a high-
er natural circulation ratio.

2.2.12. Natural circulation

The natural circulation works with the difference of density of the water in the downcom-
er and the water steam mixture in the evaporators and risers.
The pressure difference is the same as the pressure drop of friction and acceleration.
The natural circulation calculation must fulfil two conditions:
• Pressure at outlet of riser must be the same as at the inlet of downcomer.
• At each junction must be the same pressure

So the natural circulation calculation is a pressure drop calculation and a mass flow dis-
tribution calculation.

The friction of the two phase flow in the evaporator and risers is much higher than the
friction of the water in the downcomers.

The natural circulation ratio (NCR) is defined:

NCR = 1 / mass steam content in riser
Or more simply:
NCR = Mass flow downcomer / steam mass flow out of the drum

The NCR should be bigger than 5 !

The velocity in the downcomer (one phase flow) should not be bigger than
13.1 ft/s [4m/s] lower than 8 ft/s is recommended.
The velocity in the evaporator and riser should not be bigger than 32.8 ft/s [10m/s].

In some cases (mostly with horizontal evaporator pipes) it is recommended to install a

“siphon” at the downcomer, so the natural circulation can’t start in the wrong direction
at the start up. The wrong side start up can be happen because at the first steam bub-
ble production the steam water mixture is pressed in both directions: to the risers and
to the downcomers if there are too much steam bubbles in the downcomer, the down-
comer can become a riser and the circulation goes in the wrong direction, at increasing
load the direction of circulation can change, that can cause very high drum water level
changes. In this case it can happen that the drum water level goes over the maximum
or under the minimum, that triggers a boiler trip.

015-050 HRSG pages.indd 37 16.07.15 10:51

 38 2. Design of a HRSG

2.2.13. Forced through circulation

A modern HRSG has normally no forced through circulation. There are three advan-
tages of having none:
• Saving invest cost because the circulation pump is rather expensive. The pump
must run with the rather high evaporation temperature, the feed water pump runs
with a much lower water temperature.
• Saving electrical power for running the pump.
• Decreasing of the reliability of the boiler because most designs of the circulation
system is so, that it doesn’t run without the circulation pump, if the circulation pump
is damaged, the boiler must trip.

2.2.14. Fin tube heat transfer

There are different heat transfer methods for finned tubes. The methods according
[Kri1] and [Esc1] are recommended.

2.2.15. Pipe turbulent heat transfer

The formulas according Gnielinski are recommended [VDI1] chapter Gb.

2.2.16. Pipe evaporation heat transfer

The formulas according VDI Heat Atlas [VDI1] chapter H.

2.2.17. Heat conductivity of steel

Temp °F 32 68 212 392 572 752 932 1112 1292 1472 1652 1832
ASTM DIN ca.
CS St 35.8 32.9 32.9 32.9 31.2 28.9 26.0 24.3 21.4 21.4 21.4 21.4 21.4
T9 15 Mo 3 29.5 29.5 29.5 28.3 26.0 24.3 22.5 20.8 20.8 20.8 20.8 20.8
T11 13 CrMo 4 4 26.6 26.6 26.6 26.6 24.8 23.7 22.0 20.8 20.8 20.8 20.8 20.8
T22 10 CrMo 9 10 20.2 20.2 21.4 22.0 22.0 21.4 20.2 19.1 19.1 19.1 19.1 19.1
T91 X 10 CrMoVNb 9 1 15.0 15.0 15.6 16.2 16.2 16.8 17.3 17.3 17.3 17.3 17.3 17.3
T304 X 5 CrNi 18 10 8.7 8.7 9.2 10.4 11.0 12.1 12.7 13.9 14.4 15.0 16.2 16.8
T321 X 6 CrNiTi 18 10 8.7 8.7 9.2 10.4 11.6 12.1 12.7 13.9 14.4 15.6 16.2 16.8
T309 1.4833 7.2 7.3 8.1 9.0 10.0 10.9 11.8 12.8 13.7 14.7 15.6 16.5
T310 1.4841 7.2 7.3 8.1 9.0 10.0 10.9 11.8 12.8 13.7 14.7 15.6 16.5
T409 10 CrMo 9 10 13.9 14.0 14.3 14.8 15.3 15.7 16.2 16.6 17.0 17.5 18.0 18.4

Tab. 9: heat conductivity of steel Btu / hr ft F

015-050 HRSG pages.indd 38 16.07.15 10:51

 2. Design of a HRSG 39

Temp °C 0 20 100 200 300 400 500 600 700 800

ASTM DIN ca.
CS St 35.8 57 57 57 54 50 45 42 37 37 37
T9 15 Mo 3 51 51 51 49 45 42 39 36 36 36
T11 13 CrMo 4 4 46 46 46 46 43 41 38 36 36 36
T22 10 CrMo 9 10 35 35 37 38 38 37 35 33 33 33
T91 X 10 CrMoVNb 9 1 26 26 27 28 28 29 30 30 30 30
T304 X 5 CrNi 18 10 15 15 16 18 19 21 22 24 25 26
T321 X 6 CrNiTi 18 10 15 15 16 18 20 21 22 24 25 27
T309 1.4833 12.4 12.7 14 15.6 17.3 18.9 20.5 22.1 23.7 25.4
T310 1.4841 12.4 12.7 14 15.6 17.3 18.9 20.5 22.1 23.7 25.4
T409 10 CrMo 9 10 24.1 24.2 24.8 25.6 26.4 27.2 28.0 28.7 29.5 30.3

Tab. 10: heat conductivity of steel W / m K

2.2.18. Overall heat transfer

1
h=
1
h
S
+ R fo + W +
λ
Ao
Ai
R fi +
1
hi ( )
2.2.19. Logarithmic mean temperature
The logarithmic mean temperature in cross flow:
(v1i – v2i )
v1O = v1i –
1 NTU2 1
+ –
1–e –NTU 1 NTU1 • (1– e –NTU 2) NTU1

(v1i – v1o )
∆tlog Crossflow =
NTU2

Outlet H2O temperature

•
m • cp
ϑOutH2O = ϑinH2O +
Q

Outlet flue gas temperature

•
mFG • cpFG
ϑOutFG = ϑinFG –
Q

015-050 HRSG pages.indd 39 16.07.15 10:51

 40 2. Design of a HRSG

2.2.20. Designing of heating surfaces

The purpose is clear: transfer of heat from one medium to another. In most cases, the
medium flow is continuous so that we will consider this particular case only.

The actual heat transfer can be obtained by the following equation :

Q = h A ∆ tlog Q - Heat flow rate [W]

h - heat transfer coefficient [W/m²K]
A - Area [m²]
∆ tlog - log mean temperature difference [K]

To change the heat transfer capacity, following variations in heating surface geometry
are available:
1. Length of the tubes (a function of duct height)
2. Number of pipes in flue gas direction
3. Number of tubes in transversal direction (is a function of the duct width and
transversal spacing)
4. Tube diameter and tube wall thickness (is a function of stress calculation)
5. Fin height
6. Fin pitch
7. Fin thickness
8. Velocity of the gas medium (is a function of item 1, 3, 4, 5, 6, 7)
9. Velocity of the medium in the tube (is a function of item 3, 4 and how many rows
carrying flows)
10. Changing the difference in temperature

To meet the requirements, these days computer simulations are used and the different
possibilities in a ranges are tried out.

Outlet H2O temperature

m• • cp
ϑOutH2O = ϑinH2O +
Q

Outlet flue gas temperature
•
m • cpFG
ϑOutFG = ϑinFG – FG
Q

2.2.21. Noise and vibration problems at heat exchanger

At serrated fins it may happen, that due to a rather high flue gas velocity a high-pitched
whistle can occurs, that can increase to a threshold noise of pain.
The velocities in the graph is the velocity before the bundle. This velocity was much
easier to obtain in the test facility

015-050 HRSG pages.indd 40 16.07.15 10:51

 2. Design of a HRSG 41

Fig. 28: Noise – Bulk flue gas velocity diagram

Also tube bundles in cross flow are often subject to vibration and noise problems. Vibra-
tion can lead to wear and consequent tube failures. Noise problems can be a nuisance
to operating personnel.

Fig. 29:
Amplitude as a function
of the fluid velocity

Flue gas flows over a tube bundle in inline or staggered arrangement, vortices are
formed and shed beyond the wake of the tubes, resulting in harmonically varying forces
perpendicular to the flow direction. It is a self excited vibration and the frequency of vi-
bration is called vortex shedding frequency. If the frequency of vibration of the von Kár-
man vortices, as they are called, coincide with the natural frequency of vibration of the
tubes, resonance occurs leading to bundle vibration.

015-050 HRSG pages.indd 41 16.07.15 10:51

 42 2. Design of a HRSG

Fig. 30:
Vortexes after
a tube

Fig: 31: Forces

due to vortex
a- negative pressure area shedding
b- origin of vortex

Another phenomenon that occurs with vortex shedding is acoustic vibration, leading
to noise and high gas pressure drop. Standing waves are formed inside the duct. The
acoustic pressure fluctuations are a maximum where the fluid motion is zero; hence the
walls of the enclosure are subject to pressure pulsations and may distort outwardly. The
duct or the bundle enclosure vibrates when the vortex shedding frequency coincides
with the acoustic frequency.

There are five rules to reduce the danger of noise and vibration:
• Not too high velocities of flue gas between the pipes
• Install support sheets in vertical HRSG or support (pipe fixing) construction in
horizontal HRSG (to double the (eigen-) frequency)
• Install the support sheets or support construction not symmetrically to have different
(eigen-) frequencies of one pipe
• To have possibilities to install acoustic baffles to eliminate noise concerns. When a
baffle is inserted in the tube bank, reducing the width by half or a third etc.
Here again, not a symmetric of the baffles.
• Vary the fin density of the tube rows a bit to alter the frequency

015-050 HRSG pages.indd 42 16.07.15 10:51

 2. Design of a HRSG 43

Fig. 32: Eigenfrequency =

Resonance Frequency

There are some possibilities to calculate the vibration for bare tube bundles [Gan1],
but for finned tubes bundle there are existing very few papers with different sources
[Chen1].

2.2.22. Regenerative feed water preheating vs. condensate

preheating
At a big direct fired boiler the regenerative feed water preheating increases the efficien-
cy of the power plant significantly. Despite of taking steam of the turbine for preheating,
the effect is, that the most transferred heat from the steam of the turbine to the feed wa-
ter would be condensate in the condenser. So the heating power for feed water preheat-
ing can be saved and has not to be taken from the flue gas. The flue gas can be used for
air preheating. This is the reason for the increasing of the efficiency. A thermodynamic
explanation would be, that the average input heating power temperature is increasing.
At a combined cycle process this is not the case, because there doesn’t exist an air
preheater, so if the flue gas would be not used for feed water preheating, the flue gas
heating power would be lost through the stack. Therefore decrease of the steam flow
through the turbine due to the steam extraction at the turbine would only decrease the
performance of the turbine instead of increasing the live steam flow.

015-050 HRSG pages.indd 43 16.07.15 10:51

 44 2. Design of a HRSG

2.2.23. General Remarks

Typically, for large GT CCPP the clients and HRSG manufacturers request for the fol-
lowing:

• Triple Pressure Single Reheat HRSG’s - the present existing economic optimum
• High pressure (HP) level - the existing economic optimum is 1885.5 psi [130 bar],
although the thermal optimum lies well above (2610.6 [180 bar] [Eis1]) for triple pres-
sure reheat HRSG.
• Steam Temperatures - economic optimum, defined by the steam turbine, is 1049 F
[565°C].
• Steam Output - defined by the economic determination of the Pinch Point (10.8 to
14.4 F[6 to 8 K]) at the HP Evaporator and the Approach Point at the Economiser
(3.6 to 7.2 F[2 to 4 K]), typically 163 lb/s [74 kg/s] without supplementary firing (SF).
265 lb/s [120 kg/s] using SF.
• Feed water/Condensate Inlet temperature with respect to the type of fuel used,
above 122 F [50°C] for natural gas, at no sulphur content, and above 230 F [110°C]
for light distillate oil to ensure Operation above the acid or the water dew point.
• Stack temperature minimum 176 F [80°C]
• Steam Purity - entering the Super Heater at 99,9%, especially important if the client
requests for solid alkalisation in addition to the all volatile treatment (AVT), being
state of the art for HRSG design in Europe.
• HRSG flue gas draft losses - approx. 0.36 psi [25 mbar], 0.51 psi [35 mbar] if cata-
lysts are required.
• The spray cooler never should spray so much water, that the steam purity goes
under 100% (i.e. saturated steam) because the water droplets in the steam will be
separated in the next heater. Some pipes can get thermo shocks. Be careful at some
supplementary firing cases. Sothere must be a right location of spray cooler

HRSG manufacturers also offer a choice between a horizontal or vertical fluegas path.
Vertical designs - which have originally been developed in Europe where the major sup-
pliers of this kind still are located - offer a smaller footprint and are less vulnerable to
thermal cycling problems than the horizontal designs commonly applied and originated
in North America. Since the vertical HRSG no longer require forced-circulation pumps,
not even for Start ups, due to design improvements of the evaporator systems both
HRSG types offer the Same overall efficiency, although the decision may be directed to
one type of HRSG:

015-050 HRSG pages.indd 44 16.07.15 10:51

 2. Design of a HRSG 45

Horizontal HRSG Vertical HRSG

Output and Efficiency Equal Equal
Surface Area for equal Output Similar, except the reheater and Base
superheater Section, which might
require slightly more heating surface
area mainly due to less advantageous
flue gas flow distribution with regard to
temperatures and mass flow
Purging Worse to purge because unburned light Base
hydro carbons could not be purged out
as consequence of maldistribution of air
due to high opening angle of inlet duct
and low air velocoity
Plot Plan Area for equal Output Up to 30% more, mainly due to the Base
opening angle of the inlet duct and the
stack. Also if supplementary firing systems,
SCR’s, CO Catalysts, etc. are required
Emisson control Requires more HRSG length Requires more HRSG height,
cleaning of downstream fouled
surfaces has to be carried out
carefully, not to poison the catalyst.
Supplementary Firing Readily installed in the HRSG inlet Readily installed in the HRSG inlet
duct or within the boiler surface area duct, difficult to install within the
boiler surface area
HRSG enclosure / boiler house Free Standing, self supporting Attached to and supported by the
enclosure HRSG structure, light enclosure
Natural Circulation State of the art Special design considerations,
though state of the art
Modularized/Standard concepts Better modularising possible Base
Support sheets No tube support sheets needed Support sheets needed; There is a
limit of fin temperatures because
the finned tubes are supported by
sheets, at too high fin temperatures
the fins are bending
Erection Area, prefabrication Equal, though more crane area is Equal, though heavy transportation
on site required for pressure part (harps) 265,000.lb[120 ton] may be
mounting which typically lasts 5 weeks required at site, typical time needed
for large GT CCPP for boiler surface mounting:
3 weeks for large GT CCPP.
Cycling State of the art design experiences Less vulnerable if properly designed
severe cycling problems at superheater designed e.g. because of less
and reheater stages, design considera- headers
tions cost effective
HRSG cost Equal Equal
(ready to run)
O&M cost Higher number of and larger textile Replacement and blocking of tubes
expansion joints, boiler surface replace- possible
ments not possible, repair by blocking
of tubes, cost effective

Tab. 11: Horizontal HRSG vs Vertical HRSG

015-050 HRSG pages.indd 45 16.07.15 10:51

 46 2. Design of a HRSG

Fig. 33: Vertical HRSG

during erection

2.2.24. Duct burner

In the case that the electrical power output should be increased or if there is a need for
an occasional power peak, supplementary fuel is fired in the HRSG by utilizing the duct
burners. Normally the heat input through the burner is fully recovered in the HRSG. But
in addition, more heat from the gas turbine exhaust gas will also be recovered. So the
net effect of firing is to make the HRSG more efficient than the unfired case. This is the
reason for the apparent burner efficiency of greater than 100% or more heat extraction
than the amount put in through the burner. For example if 107,85 MW of heat (fuel LHV)
is input through the burner the steam turbine output should increase by about 34,1 MW
(31,66% efficiency). The steam turbine power increase is about 52,0 MW. The extra
17,9 MW are obtained from the gas turbine exhaust because under fired operation, the
stack temperature decreases causing more heat recovery.

015-050 HRSG pages.indd 46 16.07.15 10:51

 2. Design of a HRSG 47

Fig. 34:
T-Q diagram
with duct burner
in operation

This increase may not be possible if there is an LP boiler operating at a very low steam
pressure at the tail end. For example, if the gas turbine input is increased by about 8%,
the steam turbine power would also go up by the same percentage. If the same amount
of fuel is burned in the HRSG via a duct burner, the ST power will increase by about
9,5%, from the gas turbine exhaust has been recovered over the unfired operation. The
fired Operation generally can be classified in three distinct groups:

1. HRSGs designed for fired Operation only

2. HRSGs designed for occasional firing for peak loads and
3. HRSGs designed for fired and unfired Operation equally.

Of these the first two are relatively easy to design because they will be designed for one
conditions only. The continuous fired Operation boiler, because of higher temperatures,
needs more consideration in the selection of metallurgy to withstand higher tempera-
tures. Today it is feasible to design to a firing temperature of about 1500 F [820 °C] with
convection sections only. Higher temperatures up to 2000 F [1100 °C] can be designed
with a waterwall furnace section [Pas1].

An unfired HRSG is also more easy because design temperatures may not be very high
and can be accommodated with normal materials with normal thicknesses. It should be
noted that today’s advanced gas turbines have about 1200 F [650 °C] at the gas turbine
exhaust with a superheat steam temperature of 1055 F [570 °C]. Care is needed in ma-
terial selections.

015-050 HRSG pages.indd 47 16.07.15 10:51

 48 2. Design of a HRSG

When the HRSG needs to be designed for both fired and unfired operation, the design
becomes difficult. Because of the wide range of operating condition, the steam flows in
fired case may be three times that of the unfired case. The HRSG needs to be designed
with particular attention to the critical areas. Critical areas consist of attemperator sizing,
superheater and reheater velocities, drum internals, non-steaming economizer design,
valve sizing and circulation connections. Detailed specifications of all operating condi-
tions is necessary for a dual, fired and unfired design as they have a significant impact
on the design and operation of the HRSG. Interstage firing, double or triple attempera-
tors, bypassing of the part or full economizer, dual valves are some of the means em-
ployed to optimize the fired and unfired designs.

There are some cases where the HRSG is designed to have the same capacity with out
the gas turbine exhaust flow. This is done by providing a fan and burner System to du-
plicate the gas turbine exhaust conditions. Since it will be very costly to reproduce the
100% gas turbine flow and temperature, the ‘Fresh Air fired’ units are designed to oper-
ate at reduced capacity. In any event this alternative is a very costly one and should be
used only when it is very critical to have uninterrupted steam flow and when other alter-
natives outside the HRSG are not available.

2.2.25. Ductwork and casing

The flue gas ductwork uses an internally insulated, cold casing design. In this design a
combination of ceramic fibre and mineral wool insulation is sandwiched between a du-
rable alloy or carbon steel internal liner and external casing. HRSGs designed for high
rates of supplemental firing may have firing temperatures which are unsuitable for use
with alloy internal liners. In these cases a special rigidised ceramic fibre liner is em-
ployed [Pas1].

2.2.26. Environmental considerations

Environmental considerations, such as emissions have considerable influence an the
HRSG design and operation. Generally speaking, controlling the NOx and CO emis-
sions are of highest mportance. For NOx reduction, a Selective Catalytic Reduction
(SCR) is the applicable technology today. In addition to the capital cost for the hardware
and recurring cost for the injected ammonia, SCRs also increase gas turbine back pres-
sure. This results in the lower gas turbine output and increased HRSG cost since the
HRSG must be designed for a lower pressure drop. The cost of the optional duct burner
for fired cases, has to be increased to provide for a Low NOx burner.

2.2.27. Site conditions

The site conditions have a influence to the gas turbine performance and also the HRSG
output is affected. In cold areas the gas turbines and HRSG produce more power when
compared with the units operating in hot environments. Conversely at higher altitudes,
the capacities are reduced, because at higher altitudes the HRSG pressure drop will be
higher for the same amount of gas flow. This reduces the overall capacity.

015-050 HRSG pages.indd 48 16.07.15 10:51

 2. Design of a HRSG 49

For example a HRSG in Mexico City 7,350. feet [2240 meter] over see level would have
about 33% more volume flow due to the reduced pressure at this altitude.
Environment air born chemicals also have an effect an HRSG performance. HRSG near
oceans or chemical factories may experience higher level of chlorine. In such places,
series 300 stainless steel for feedwater heaters can not be used due to the corrosive
effects of chlorine. For these cases the design is either made low efficient by exhaust-
ing at a higher stack temperature or stainless steel is replaced by other higher order
material.

2.2.28. Steaming in economizers

One of the problems

often encountered in
HRSGs is economizer
steaming or steam for-
mation in economiz-
ers, particularly at low
loads or low steam
generation levels. This
may result in vibration,
noise problems, de-
posit formation inside
tubes and consequent
fouling and poor per-
formance.

Steaming in econo-
Fig 35 Pressure drop – Mass flux diagram with instability
mizers normally oc-
curs in HRSG, at
direct fired boilers the
economizer outlet wa-
ter temperature de-
creases at part load.

Fig. 36: Pressure drop – Mass flux diagram for stabilization

015-050 HRSG pages.indd 49 16.07.15 10:51

 50 2. Design of a HRSG

2.2.28.1. Methods of minimizing Steaming Concerns in a HRSG

• Lower the pinch point

• To avoid instability increase the pressure drop of economizer either to use less par-
allel pipes or insert orifices at the inlet of the pipes. Because of the steaming the
volume and the velocity of the water steam mixture is increasing rapidly, so the pres-
sure drop increases to the square of the velocity, in consequence it can happens
that some pipes of the economizers have not a proper water mass flow, because
the water goes through this pipes with lower pressure drop. This effect is called “in-
stability” [Hel1]

• Last leg of economizer coils should have vertical flow upstream in order to ensure
that the steam bubbles flow smoothly up. Downward motion of steam bubbles can
cause flow stagnation and flow instability problems. The last legs of the economizer
may be designed with multipasses to accomplish this.

• If steaming occurs for a very short duration only, the situation can be handled by
increasing the continuous blow down, though it is not recommended for continuous
operation as treated water is wasted.

• The steaming problem is associated with low steam flows in the HRSG. Hence if
you have auxiliary firing capability, use it to increase the steam flow when steaming

2.2.29. Important notes

• Reduces bypasses of flue gas at the heating surfaces with baffles etc.

• Fired HRSGs are in most cases more efficient than unfired units

• Higher the fin density and surface respectively, lower the overall heat transfer coef-
ficient.

• With a lower the tube side heat transfer coefficient, there should be a smaller exter-
nal fin surface area

• HRSG can be optimised using HRSG simulation methods

• Water temperature affects economizer tube wall temperature much more than
the gas temperature and hence for corrosion prevention consider raising water
temperature

• Fouling inside tubes is more serious in finned tube surfaces than in bare tube sur-
faces
• Surface areas should not be the basis for selecting HRSGs

• Understand the difference in efficiency based on higher and lower heating values

015-050 HRSG pages.indd 50 16.07.15 10:51

 3. Operation of steam boiler 51

3. Operation of steam boiler

051-080 HRSG pages .indd 51 13.07.15 11:54

 52 3. Operation of steam boiler

3. Operation of steam boilers

3.1. Example of a start up

3.1.1. Gas turbine mass flow

Fig. 37: Gas turbine mass flow – time diagram during start up

3.1.2. Gas turbine temperature

Fig. 38: Gas turbine temperature – time diagram during start up

051-080 HRSG pages .indd 52 13.07.15 11:54

 3. Operation of steam boiler 53

3.1.3. HP Steam mass flow

Fig. 39: High pressure steam mass flow- time diagram during start up

3.1.4. HP Steam pressure

Fig. 40: High pressure steam pressure- time diagram during start up

051-080 HRSG pages .indd 53 13.07.15 11:54

 54 3. Operation of steam boiler

3.1.5. Gradients

Fig. 41:
High pressure
temperature and
pressure gradi-
ent- time diagram
during start up

3.2. Start up
Start-up and shutdown operations have little impact to units designed for base load op-
erations. For units with cyclic operation, number of start-ups and shutdowns, the condi-
tion of the HRSG at start-up, the nature of shutdown, all influence the life of the boiler
and hence need to be taken into consideration while designing the HRSG. Generally
thicker components such as HP drum are considered for life time studies. If the HRSG is
exposed to extreme conditions and the frequency of changes is high, dynamic analysis
and life time study is necessary for the HRSG.

3.2.1. Deaeration of economizers

It is very important to take care and attention for a complete deaeration of the economiz-
ers. If there are too big air bubbles after filling the economizers with water the perform-
ance of the economizers can be very low.

3.2.2. Purging
The Purging required as a precondition to start the GT through the boiler, is a common
requirement of all boiler codes to ensure safety operation of the plant.
Germany:
• TRD 411 and TRD 412 [TRD1].
• Europe: EN 12952
• UK : British Gas
• USA : NFPA 8506, NFPA 8606
This rules are historically evolved, since in the beginning of boiler Operation severe ac-
cidents occurred. Today purging a hot HRSG strains all involved boiler parts to a high
extent, special considerations and design features have to be taken into account (see
above) to cope with the requirement of daily start ups for a lifetime of 25 years.

051-080 HRSG pages .indd 54 13.07.15 11:54

 3. Operation of steam boiler 55

The stratification of air flow in horizontal boilers during prestart purge of the HRSG also
suggests that the purge is not accomplishing its intended purpose--to remove combus-
tible gases from the HRSG before turbine ignition. Because natural gas and most vola-
tiles released from distillate fuel oil are lighter than air, it is particularly important for the
purge to flush out any combustibles that have accumulated in the dead spaces at the
top of the duct.

The industry needs

a thorough review
of the purpose of
HRSG purging and
of the circumstanc-
es in which there
is risk of ignition of
combustibles in the
upper duct. So it is
recommended to in-
crease the transition
angles of the inlet
duct of a horizontal
boiler.

1. Extensive temperature 2. Condensate began to clear

3.3. Drain monitoring confirmed that a from superheater tubes once
substantial quantity of condensate steam flow commenced
formed in superheater tubes
Condensation oc- during gasturbine purging,
curs in superheater even in large-bore headers
tubes during every
purge of the HRSG
at warm or hot start
prior to gastur-
bine ignition. This
is because turbine
exhaustgas tempera-
ture falls below satu-
ration temperature.
Quantities of con-
densate are substan-
tial during hot and
warm starts. A repeat
purge can actually fill
the front panel tubes 3. A single, small-bore drain, 4. Condensate clears first from
of the superheater. opened during purging, reduces the tubes closest to the end-pipe
the quanity of condensate, but connections, creating temperature
does not completely eliminate it differences between individual
tubes along the headers
Fig. 42: Condensate forming during warm and hot start

051-080 HRSG pages .indd 55 13.07.15 11:54

 56 3. Operation of steam boiler

Condensate should be removed from lower headers at the peak rate at which it forms
to prevent pooling and flooding. But this requires adequately sized and correctly oper-
ated drains elements that have been overlooked at many large combined-cycle installa-
tions. Many units have no blowdown vessel for high-pressure/temperature drains from
the superheater. Others have a blowdown vessel inadequately rated for flow, pressure,
and temperature of drains from the superheater during hot restart purges. The drain
installed an most superheater designs was sized for maintenance purposes and is too
small for clearing condensate at the rate it collects. In [Eis2] it is reported how to esti-
mate the drain water flow and taking the right nozzles.

Even where superheater drains are installed and connected to a blowdown tank, no
guidance has been given by HRSG manufacturers or EPC contractors as to when and
how they are to be used. Not surprisingly, they often are incorrectly used or not used at
all during hot starts.

To remove condensate from lower headers of vertically tubed HRSGs, the lower head-
ers must have adequate bore in relation to their length and number of attached tubes to
ensure that tubes cannot flood [Pea1].

Fig. 43:
Drainage arrangement

Drain arrangement [Als1]

3.4. Drum water level

Before start up it is recommended to decrease the drum water level, because at the first
evaporation there will be water displaced by steam in the evaporators, causing a “water
swell” in the drum. So a lower drum water level can prevent a too fast increasing over
the high water level mark, that trigger a boiler trip.
The drum water level before start up can be controlled by the blow down valve.

051-080 HRSG pages .indd 56 13.07.15 11:54

 3. Operation of steam boiler 57

3.5. Water running through the economizer

Also it is recommended to run water through the economizers before and during start
up. The water comes out through the blow down valve. The point of it is to have lower
and constant temperature and to pour steam and air bubbles out of the economizer. If
for example there are too much steam bubbles in the economizer, then it can happens
that a fast increasing pressure can cause the collapsing of the bubbles and then the
feed water has to fill first the eco and in worse case the eco “sucks” water out of the
drum. For that reason it is recommended that the connection of the eco to the drum is
above the drum water level.

3.6. Start up of the gas turbine

Fig. 44: Typical Start up of a gas turbine

051-080 HRSG pages .indd 57 13.07.15 11:54

 58 3. Operation of steam boiler

3.7. Life Cycle Fatigue

Often HRSGs operate under cyclic conditions during startup, shutdown, and load
changes. These conditions will induce thermal stress particularly where transitions oc-
cur in thick walled pressure part components. The thermal stress events lead to fatigue
damage. The approach in minimizing the impact and prolonging the useful life of the
HRSG’s was twofold.

Fig. 45: Temperature differences in a drum shell

The design should be in this way to minimize fatigue damage. These features include
full penetration welds throughout the design to eliminate crack initiation points and utili-
zation of the highest grade material practical to reduce pressure vessel wall thickness.
The mechanical design details of tube offsets, tube harps, and supports/hangers should
be arranged to minimize the stresses caused due to thermal expansion.

An examination of the fatigue life based upon ASME Section VIII Div 2 (Europe EN
12952- 3 and Germany also TRD 301 Appendix I possible) criteria should be done and
in some cases finite element stress analysis. The objective of this analysis is to provide
the impact of combustion turbine start-up scenarios and to provide parametric curves
showing the life cycle life for different ramp rates and temperature differentials [Bri1].

051-080 HRSG pages .indd 58 13.07.15 11:54

 3. Operation of steam boiler 59

Fig. 46: Cycling temperature changes

3.8. Temperature gradients drums and headers

One result of life cycle fatigue calculation is a possible temperature gradient to have
an acceptable life time consumption of this part (i.e. drum or header). For the drum it is
possible to calculate out of the temperature gradient a pressure gradient, because the
temperature and pressure at saturation has a fixed relation. The start up, load changes
and shut down should be done according these gradients.

Fig. 47: Stress at tubes in a header Fig. 48: Possible tube-

header arrangement

Most horizontal HRSGs use multiple row harp designs, consisting of one horizontal up-
per header and one horizontal lower header joined by two, three or more rows of vertical
tubes. The temperature of exhaust gas passing across multiple tube rows is succes-

051-080 HRSG pages .indd 59 13.07.15 11:54

 60 3. Operation of steam boiler

ively reduced so that individual tube rows may operate at different temperatures, lead-
ing to differential thermal stress at the weld joints. These stresses can be reduced by
having only one row of tubes between the headers. This allows the use of smaller di-
ameter headers according ASME UG 27 (Wall thickness proportionally to the diameter)
and also minimises circumferential temperature gradients in the headers. Small diam-
eter headers reduce thermal stresses by as much as 60 per cent when compared with
headers having multiple tube rows.

Fig. 49: Drum displacement during preheating at start up

One comparative simulation for hot start has examined the effects of a thermal quench
during shutdown and provides an example of the benefits of the smaller header with a
single row of tubes over a bigger header with two tubes. In the simulated thermal quench,
condensate forms in the tube, the
tube cools to saturation tempera-
ture (545 °F [285 °C]) while the
bottom of the header remains at
nominal steam temperature (800
°F [427 °C]). A simulation shows
that the predicted number of cy-
cles to crack initiation are 18,272.
for the header with a single tube
and 6,156. for the header with two
tubes, which experiences much
higher stresses than the smaller
header [fig. 49].

Fig. 50: Drum shell

stress during start
up with different
stress in steam
and water section

051-080 HRSG pages .indd 60 13.07.15 11:54

 3. Operation of steam boiler 61

Use of finned tubes without bends and radial penetration of the header are also ben-
eficial as they minimise stresses in the tube-to-header welds. Up flow and down flow in
the same harp should also be avoided, so headers need to be designed without division
walls to avoid unacceptable temperature differentials in tubes adjacent to the division
wall, which may occur particularly during transients [Als1].

Fig. 51: Drum shell stress during start up with different stress in steam and water section

Fig. 52: Model of drum with nozzles

051-080 HRSG pages .indd 61 13.07.15 11:54

 62 3. Operation of steam boiler

Fig. 53: Stub with thermo sleeve

3.9. How to start up faster

The start up time depends on maximal allowable temperature changes (temperature

gradient). This comes from the life time fatigue calculation and is a function of cycling
numbers and initial temperature.
The HRSG can be started up faster if the drum metal temperature is higher than the
ambient temperature. There are different possibilities to increase drum metal tempera-
ture before start up:
• Preheating of the drum with auxiliary steam
• Electrical preheating of the drum metal
• Cool down the boiler slowly e.g. close steam turbine valve very fast after shut
down
• Thick insulation of the boiler
• Insulated boiler house
• Stack damper to prevent that cold air pour into the flue gas duct.

3.10. Control system

A HRSG has three main control systems

1. Drum water level control (i.e. feed water mass flow is affected as a function of the
drum water level)

2. Pressure of live steam (i.e. the turbine valve or start up valve is affected as a func-
tion of live steam pressure)

3. Temperature (i.e. a spray cooler valve is affected as a function of the temperature at

a test point)

051-080 HRSG pages .indd 62 13.07.15 11:54

 3. Operation of steam boiler 63

3.10.1. Drum water level control

Steam drum level control is necessary to add makeup water as steam is delivered into
the header and to the associated process equipment. The system should control the
drum level at a specific set point while compensating for varying steam demands and
drum pressures. For a given volume of steam and blowdown leaving the steam drum,
an equal amount of water should replace that inventory. Discussions here include the
primary sensing devices and the various strategies used to control steam drum level
and feedwater rate.

Fig. 54: Three element

drum water level control

3.10.2. Level measurement

The drum level is measured using a differential pressure transmitter. Refer to the draw-
ing on the right. The output of the instrument increases as the differential pressure de-
creases. A typical range is 30 inches of water column. A condensing reservoir is installed
to allow the high side of the D/P to measure the steam pressure plus the hydraulic pres-
sure in the reference leg. The low pressure side senses the boiler drum pressure, the
weight of the water above the low pressure tap and the weight of a column of saturated
steam from the high pressure tap to the water level. Being a differential pressure device,
the boiler drum pressure is cancelled out of the measurement, leaving only the water
column pressure difference. The level measurement is accurate only at a single drum
pressure. If needed, a pressure measurement can compensate for varying drum pres-
sures by applying a gain and bias to the drum level signal.

3.10.3. Swell and shrink

Under steady-state conditions, both water and steam bubbles reside below the water
surface. The average mixture density is constant. Should steam demand increase, the
steam bubbles expand under the water surface, increasing the average mixture density.
This causes an increase in steam drum level without the addition of feed water. This
increase in level proportional to an increased steaming rate and decreased drum pres-
sure is called swell. Inversely, as the steam load decreases, the steam bubbles in the
steam/water mixture decrease in size and volume. This causes a decrease in drum lev-
el, although the mass of water and steam has not changed. This phenomenon is called
shrink. These apparent changes in drum level can be compensated for by implementing
a pressure transmitter and drum pressure correction factors as a part of the feed water
control strategy.

051-080 HRSG pages .indd 63 13.07.15 11:54

 64 3. Operation of steam boiler

3.10.4. Single element control

A single element control strategy uses the drum level signal as the process variable. Af-
ter performing a P+I computation on the level deviation from set point, the output of the
controller is sent to the feed water flow control valve. This configuration is used on small
boilers with a relatively large water volume and steady loads.

3.10.5. Two element control

In addition to the drum level measurement, a two element design uses a steam flow
transmitter as a feed forward input to the feed water controller. This feed forward plus
feedback configuration allows the controller to anticipate a need for additional feed wa-
ter flow before this is sensed as a lowered drum level. The steam flow signal is char-
acterized for gain and applied to a summer modifying the level controller output. The
diagram shows the two element configuration with an optional drum pressure input. This
design is typically used on small to medium boilers with moderate changes in steam
demand. The effects of shrinking and swelling are minimized in this design. However, a
disadvantage is the lack of compensation for varying feed water pressure. This problem
is common in multiboiler installations with a common feed water system. These pres-
sure variations change the material balance between steam flow and feed water flow.
If variable feed water pressure is a problem, a three element control configuration can
correct the difficulty.

3.10.6.
Three element control
In addition to adding a feed water flow
input, two P+I controls are used in a
cascade configuration. Refer to the
SAMA drawing at the right. The com-
puted output of the level controller is
software linked to the remote set point
of the feed water controller. The slow
reacting level control is usually tuned
with a moderate proportional band
setting and a long integral time. As
with two element control, the steam
flow measurement is used as a feed
forward input to anticipate steam de-
mand changes. Adding the fast acting
feed water controller allow quicker re-
actions to load changes. Feed water
pressure variations do not adversely
Fig. 55: affect the control action, since the feed
Three element drum water level control scheme water flow rate is metered.

051-080 HRSG pages .indd 64 13.07.15 11:54

 3. Operation of steam boiler 65

Fig. 56: Advanced three element

drum water level control scheme

It is recommended that the maximum inlet velocities applied to control valves

should be as shown in the tables below.

Globe Valves Size Liquid Steam or Gas

mm inch m/s ft/s m/s ft/s
15 - 25 1/2 - 1 9 30 120 400
40 - 50 1 1/2 - 2 7.5 25 90 300
65 - 100 2 1/2 - 4 6 20 75 250
150 - 200 6 - 8 6 20 70 225
250 - 400 10 - 16 4.5 15 55 175
Angle Valves Size Liquid Steam or Gas
mm inch m/s ft/s m/s ft/s
15 - 25 1/2 - 1 13.5 45 135 450
40 - 50 1 1/2 - 2 12 40 105 350
65 - 100 2 1/2 - 4 10.5 35 90 300
150 - 200 6 - 8 9 30 85 275
250 - 400 10 - 16 7.5 25 70 225

Tab. 12: Recommended maximum inlet velocities

051-080 HRSG pages .indd 65 13.07.15 11:54

 66 3. Operation of steam boiler

3.10.7. Four element control

Require additionally a blow down mass flow measurement and the logic for changing of
the blow down mass flow.

3.10.8. Pressure control

The pressure is controlled by the turbine valve or start up valve respectively and the
input heat. The control loops are rather easy. The turbine valve can influence the pres-
sure of the boiler very fast. The reaction of the pressure to the change of input heat is
also rather fast and direct.

3.10.9. Spray cooler control

The spray cooler control logic can be rather simple too.

Fig. 57:
Spray cooler control scheme

3.10.10. Control Methods

3.10.10.1. PID
Nowadays the PID controls are emulated by micro control logic (i.e. microprocessor).

3.10.10.2. Digital control logic

There is an alternative to the PID control logic, for example:
• Digital logic
• Dead beat control
• Smith predictor [Smi1]

051-080 HRSG pages .indd 66 13.07.15 11:54

 3. Operation of steam boiler 67

3.10.11. Ziegler-Nichols Methods Facilitate Loop Tuning

Tuning a proportional-integral-derivative (PID) controller is a matter of selecting the right

mix of P, I, and D action to achieve a desired closed performance.
The ISA standard form of the PID algorithm is:

[ 1
( ) (
CO (t) =P e(t) + • ∫e(t) dt – TD • d PV (t)
TI dt )]
The variable CO(t) represents the controller output applied to the process at time t,
PV(t) is the process variable coming from the process, and e(t) is the error between the
setpoint and the process variable. Proportional action is weighted by a factor of P, the
integral action is weighted by P/TI, and the derivative action is weighted by PTD where
P is the controller gain, TI is the integral time, and TD is the derivative time.
In 1942, John G. Ziegler and Nathaniel B. Nichols of Taylor Instruments (now part of
ABB Instrumentation in Rochester, N.Y.) published two techniques for setting P, TI, and
TD to achieve a fast, though not excessively oscillatory, closed-loop step response.
Their ‘open loop’ technique is illustrated by the reaction curve in the figure. This is a strip
chart of the process variable after a unit step has been applied to the process while the
controller is in manual mode (i.e., without feedback).

A line drawn tangent to the reaction curve at its steepest point shows how fast the proc-
ess reacted to the step input. The inverse of this line’s slope is the process time constant
T. The reaction curve also shows how long the process waited before reacting to the
step (the deadtime d) and how much the process variable increased relative to the size
of the step (the process gain K). Ziegler and Nichols determined that the best settings
for the tuning parameters could be computed from T, d, and K as follows:

1.5 • T
P=
K•d

TI = 2.5 • d

TD = 0.4 • d

Once these parameter values are loaded into the PID algorithm and the controller is
returned to automatic mode, subsequent changes in the setpoint should produce the
desired ‘not-too-oscillatory’ closed-loop response. A controller thus tuned should also
be able to reject load disturbances quickly with only a few oscillations in the process
variable.

051-080 HRSG pages .indd 67 13.07.15 11:54

 68 3. Operation of steam boiler

Ziegler and Nichols also described a ‘closed loop’ tuning technique that is conducted
with the controller in automatic mode (i.e., with feedback), but with the integral and de-
rivative actions shut off. The controller gain is increased until any disturbance causes
a sustained oscillation in the process variable. The smallest controller gain that can
cause such an oscillation is called the ultimate gain PU. The period of those oscillations
is called the ultimate period TU. The appropriate tuning parameters can be computed
from these two values according to these rules:

P = PU
TI = 0.625 • TU
TD = 0.1 • TU

Fig. 58: Ziegler Nichols Reaction Curve

3.10.12. Load change

See item 3.10.14

3.10.13. Sliding Pressure

See item 3.10.14

051-080 HRSG pages .indd 68 13.07.15 11:54

 3. Operation of steam boiler 69

3.10.14. Example of a load change with duct burner

Fig. 59: Temperatures during cycling with duct burner

Fig. 60: steam mass flow during cycling with duct burner

051-080 HRSG pages .indd 69 13.07.15 11:54

 70 3. Operation of steam boiler

Fig. 61: steam pressure during cycling with duct burner

Fig. 62: temperature gradient during cycling with duct burner

3.10.15. Load change of the gas turbine

At the load change of the gas turbine the temperature changes are rather low, only the
flue gas mass flow is changing in a wide range and so if the boiler is running in a sliding
pressure mode with the flue gas mass flow the live steam mass flow is changing. With
live steam mass flow the pressure and evaporation temperature increases. But with
higher pressures the temperature changes is rather moderate.

051-080 HRSG pages .indd 70 13.07.15 11:54

 3. Operation of steam boiler 71

Fig. 63: Gas turbine temperature and mass flow – load diagram

[Gen1]

3.10.16. Duct burner

With a duct burner the load change can be made very rapidly. The input energy with the
fuel is converted very fast in steam because the flue gas temperature at the outlet of
the HRSG is normally the same or lower, so the complete energy is transferred to the
live steam mass flow. The live steam temperature is normally at HRSG with modern gas
turbines constant and controlled by spray coolers.

051-080 HRSG pages .indd 71 13.07.15 11:54

 72 3. Operation of steam boiler

3.10.17. Shut down

Like at a start up the temperature changes should not be too big. But at the shut down
it is not such a big problem because the temperature gradients for e.g. the drum can be
controlled very easy by the turbine valve. In this case it must be looked at the tempera-
ture at the outlet header of the last superheater. The superheater will be cooled by the
flue gas of the running out gas turbine and the steam. Therefore cooling may be too fast.

3.10.18. Run out of turbine

Fig. 64:
Protective load shedding
trip [ABB1]

Fig. 65:
Expected normal shut
down ABB[1]

051-080 HRSG pages .indd 72 13.07.15 11:54

 4. Appendix I Converting factors 73

4. Appendix I Converting factors

MULTIPLY BY TO OBTAIN
LENGTH mm 0.03937 in
m 3.2808 ft
in 0.083333 ft
in 25.4 mm
ft 12 in
ft 0.3048 m
AREA m² 10.764 ft²
cm² 0.155 in²
in² 0.0069444 ft²
ft² 0.092903 m²
SPECIFIC AREA m²/m 0.32808 ft²/ft
ft²/ft 0.3048 m²/m
VOLUME m³ 35.315 ft³
in³ 0.0005787 ft³
ft³ 0.028317 m³
gal (US) 0.13368 ft³
gal (IMP) 1.2009 gal (US)
litre (=dm³) 0.26417 gal (US)
MASS kg 2.2046 Ib
Ib 0.45359 kg
FORCE N 0.22481 Ibf
kp 2.2046 Ibf
Ibf 32.174 lb ft/s²
kp 9.80665 N
PRESSURE psi 0.068948 bar
bar 14.5036839 psi
Pa 0.0040218 in of water
kPa 4.0218 in of water
kPa 0.14504 Ibf in
kp/m² 0.039441 in of water
mm of water 0.03937 in of water
in of water 0.24864 kPa
in of water 0.036063 ibf in
in of wate, 0.0024539 atmospheres
VELOCITY m/s 3.2808 ft/s
m/min 3.2808 ft/min
m/hr 1.4667 ft/s

051-080 HRSG pages .indd 73 13.07.15 11:54

 74 4. Appendix I Converting factors

MULTIPLY BY TO OBTAIN
ft/s 0.3048 m/s
ft/min 0.016667 ft/s
MASS FLOW kg/s 7936.6 lb/hr
kg/hr 2.2046 lb/hr
lb/s 3600 lb/hr
lb/hr 0.45359 kg/hr
VOLUME m³/s 2118.9 ft³/min
FLOW ft/s 60 ft³/min
ft³/min 0.00047195 m³/s
MASS kg/(s m²) 737.34 lb/(hr ft²)
VELOCITY kg/(hr m²) 0.20482 lb/(hr ft²)
Ib/(s ft²) 3600 lb/(hr ft²)
Ib/(hr ft²) 4.8824 kg/(hr m²)
DENSITY kg/m³ 0.062428 lb/ft³
g/cm³ 62.428 lb/ft³
lb/in³ 1728 lb/ft³
lb/ft³ 16.018 kg/m³
SPECIFIC kJ/kg C 0.23885 Btu/Ib F
HEAT kcal/kg C 1 Btu/Ib F
Btu/lb F 4.1868 kJ/kg C
THERMAL W/m K 0.57779 Btu/hr ft F
CONDUCTIVITY cal/s cm K 241.91 Btu/hr ft F
Btu/hr ft²F/in 0.083333 Btu/hr ft F
Btu/hr ft F 1.7307 W/m K
kcal/hr m K 0.67197 Btu/hr ft F
W/cm K 57.779 Btu hr ft F
DYNAMIC Pa s 2419.1 Ib/hr ft
VISCOSITY cP 2.4191 Ib/hr ft
kg/hr m 0.67197 Ib/hr ft
lb/s ft 3600 Ib/hr ft
Ibf/s ft 115827 Ib/hr ft
lb/hr ft 0.00041338 Pa s
Ib/hr ft 0.41338 cP
KINEMATIC m²/s 38750 ft²/hr
VISCOSITY cSt 0.03875 ft²/hr
m²/s 10000 Stokes
ENERGY kJ 0.94782 Btu
kWhr 3412.1 Btu

051-080 HRSG pages .indd 74 13.07.15 11:54

 4. Appendix I Converting factors 75

MULTIPLY BY TO OBTAIN
kcal 3.9683 Btu
ft Ibf 0.0012851 Btu
hp hr 2544.4 Btu
Btu 1.0551 kJ
Btu 0.0002931 kWhr
Btu 0.252 k cal
J 0.00027778 W hr
POWER W 3.4121 Btu hr
kcal/hr 3.9683 Btu/hr
ft Ibf/hr 0.0012851 Btu/hr
hp 2544.4 Btu/hr
Btu/hr 0.29307 W
Btu/hr 0.252 kcal/hr
HEAT FLUX W/m² 0.317 Btu/(hr ft²)
kcal hr m- 0.36867 Btu/(hr ft²)
W cm 3170 Btu/(hr ft²)
cal s cm- 13272 Btu/(hr ft²)
Btu-hr ft 3.1546 W/m²
Btu. hr ft- 2.7125 kcal/(hr m²)
THERMAL m²K/W 5.6783 hr ft² F/Btu
RESISTANCE hr m²K/kJ 20.442 hr ft² F/Btu
hr m²K/kcal 4.8824 hr ft² F/Btu
s cm²K/cal 0.00013562 hr ft² F/Btu
cm²K/W 0.00056783 hr ft² F/Btu
hr ft² F/Btu 0.17611 m²K/W
hr ft² F/Btu 0.20482 hr m²K/kcal
HEAT W/(m²K) 0.17611 Btu/(hr ft²F)
TRANSFER kJ/(hr m²K) 0.048919 Btu/(hr ft²F)
COEFFICIENT kcal/(hr m²K) 0.20482 Btu/(hr ft²F)
cal/(s cm²K) 7373.4 Btu/(hr ft²F)
W/(cm²K) 1761.1 Btu/(hr ft²F)
Btu/(hr ft²F) 5.6783 W/(cm²K)
Btu/(hr ft²F) 4.8824 kcal/(hr m²K)
Temperature °C 1,8C+32 °F
°F (F-32)/1.8 °C
°C °C+273.15 K
°F °F+459.69 °R

Tab. 13

051-080 HRSG pages .indd 75 13.07.15 11:54

 76 5. Appendix II Disclaimer

5. Appendix II Disclaimer

KED and the author has attempted to provide accurate information on this script, infor-
mation on this script may contain technical inaccuracies or typographical errors. The
contents of any documents on this script are believed to be current and accurate as of
their publication dates. KED AND THE AUTHOR assumes no responsibility for the ac-
curacy of the information.

KED OR THE AUTHOR may change the programs or products mentioned at any time
without notice. For the most current and complete documents, please contact KED OR
THE AUTHOR directly. Mention of non-KED products or services is for information pur-
poses only and constitutes neither an endorsement nor a recommendation.

ALL INFORMATION, SOFTWARE, PRODUCTS AND SERVICES PROVIDED ON THIS

SCRIPT ARE PROVIDED “AS IS” WITHOUT WARRANTY OF ANY KIND, WHETHER
EXPRESS, IMPLIED, STATUTORY OR OTHERWISE. KED AND THE AUTHOR DIS-
CLAIMS ALL WARRANTIES, EXPRESSED OR IMPLIED INCLUDING, WITHOUT LIM-
ITATION, THOSE OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE
AND NONINFRINGEMENT OR ARISING FROM A COURSE OF DEALING, USAGE,
OR TRADE PRACTICE. Some jurisdictions do not allow the exclusion of implied war-
ranties, so the above exclusion may not apply to you.

KED AND THE AUTHOR SHALL NOT BE LIABLE FOR ANY INDIRECT, SPECIAL,
CONSEQUENTIAL, OR INCIDENTAL DAMAGES INCLUDING, WITHOUT LIMITA-
TION, LOST PROFITS OR REVENUES, COSTS OF REPLACEMENT GOODS, LOSS
OR DAMAGE TO DATA ARISING OUT OF THE USE OR INABILITY TO USE THIS
SCRIPT OR ANY KED PRODUCT, OR DAMAGES RESULTING FROM USE OR RE-
LIANCE ON THE INFORMATION PRESENT, EVEN IF KED OR THE AUTHOR HAS
BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.

051-080 HRSG pages .indd 76 13.07.15 11:54

 6. Literature 77

6. Literature

[Nes1] H. Nessler, R. Preiss; P. Eisenkolb; Developments in HRSG Technology; MAB

Anlagenbau Austria
[Pas1] A. Pasha, A. S. Ragland, S. Sun; Thermal and economic considerations for
optimizing HRSG Design; Proceedings of ASME TURBO EXPO 2002;
[Vis1] Viswanathan, R; Bakker, W.T.: Materials for Ultra Supercritical Coal Power
Plants, Part 1: Boiler Materials. Power Plant Chem 3(7), 2001.
[Bri1] Briggs, John R; Repowering a 2400 Psig Steam Cycle with Natural Circulation
HRSG’s; FWC;
[Brü1] Bründermann, G, Daublebsky von Eichhain, C; Cost reduction for finned tube
heat exchangers through optimisations of the fin characteristics; Rosink 2002;
[Got1] E. J. Gottung, T. P. Mastronarde, Alstom Power, Heat Recovery Steam
Generator Developments for the Global Market
[Swa1] Robert Swanekamp, PE, Editor-inChief SPECIAL REPORT: HRSGs blossom
in the competitive-power landscape,POWER Magazine, July/August 2000
[Gan1] Avoid heat transfer equipment vibration; V. Ganapathy; Hydrocarbon
Processing; June 1987
[Chen1] Y. N. Chen; Flow induced vibration and noise in tube bank heat exchangers
due to von Karman Streets; Trans. ASME Journal of Engineers for Industry;
Vol1. 1968 pp. 134- 146
[Berg1] W. Bergamann; Werkstofftechnik Bd. 2 Anwendungen; 1987 Hanser Verlag
München;
[Wei1] Weiermann, R.C.; Design of Heat Transfer Equipment for Gas-Side Fouling
Service. In Workshop on an Assessment of Gas- Side Fouling in Fossil Fuel
Exhaust Enviroments. Editors W.J. Marner and R.L. Webb. Publikation 82-
67, Jet Propulsion Laboratory, California Institute of Technology Pasadena,
California; 1982;
[Gan2] Ganapathy, V.; Cold end corrosion: causes and cure; Hydrocarbon Processing;
January 1989;
[Sim1] Simon, D.E.; Feedwater Quality in Modern Industrial Boilers – A Consensus of
Proper Current Operating Practices. 36th Annual Water Conference;
Pittsburgh, Pa.; 1975
[VGB1] Wasservorschriften für Wasserrohrkessel. Mitteilungen VGB 49, 3 S. 215- 217;
1969
[Mac1] Macbeth, R.V., Trenberth, R, Wood, R.W.; An Investigation into the Effect of
‘Crude’ Deposits on Surface Temperature, Dry- Out and Pressure Drop, With
Forced Convection Boiling of Water at 69 bar in an Annular Test Section. AEEW
– R 705; 1971
[Mac2] Macbeth, R. V.; The Effect of ‘Crud’ Deposits on Frictional Pressure Drop in a
Boiling Channel; AEEW- R 767; 1971;
[Mac3] Macbeth, R. V.; Boiling on Surfaces Overlayed With a Porous Deposit- Heat
Transfer Rates Obtainable by Capillary Actions; AEEW – R 711; 1971;
[Som1] Somerscales, E.F.C., Kassemi, M.; Fouling Due to In-Situ Corrosion Products.
ASME 22nd National Heat Transfer; Conf. HDT-Vol 35.1.;1984

051-080 HRSG pages .indd 77 13.07.15 11:54

 78 6. Literature

[Goo1] Goodstine, L., Kurpen, J.J.; Corrosion and Corrosion Product Control in the
Utility Boiler Turbine Cycle. Mater. Perf.; Vol 13. 1, 31; 1974;
[Ver1] F.H. Verhoff, J.T. Banchero; Predicting dew points of flue gases; Chem. Eng.
Progress 70; 1974; p 71/72
[Ras1] Rasch R.; Über die Reaktionen zwischen Eisenoxid und Schwefeldioxid; GIT
Fachz. Lab.; 1978; S. 667- 671;
[Wic1] Wickert, K.; Chemische Probleme bei der Verfeuerung schwerer Heizöle in
Hochleistungskesseln und Gasturbinen. Das Ölfeuerjahrbuch; 1961;
Verlag G. Kopf; Stuttgart
[Gen1] General Electric Data Sheet PG7241(FA) Gas turbine
[ABB1] ABB Gas Turbine GT 13 E2
[Als1] Alstom Power USA; Introducing OCC; Modern Power Systems; July 2001;
TRD Technische Regeln für Dampfkessel
National Fire Protection Association 8506
British Gas: Guidiance notes on the installation industrial turbines, associated
gas compressors and supplementary firing burners (06/89)
[Smi1] Smith O. J. M., “Closer control of loops with dead time”; Chemical Engineering
Progress, 53, 217–219 (1957); “A controller to overcome dead time”,
ISA Journal, 6, 2, 28–33 (1959).
[Kri1] Krischer, O., Kast, W.; Wärmeübergang und Wärmespannungen bei Rippen-
rohren; VDI Forschungshefte 474; 1959;
[Esc1] Escoa; Engineering Manual; Escoa Fintube Corporation; Oklahoma USA; 1979;
[VDI1] VDI Heat Atlas; VDI Verlag Düsseldorf; 1993;
[Eis1] Eisenkolb, P, Pogoreutz, M, Halozan, H; Modified Rankine HRSG beats triple
pressure system; ASME paper 96-TA-50;Jakara; ASME Turbo Asia 1996
[Eis2] Eisenkolb, P, Nessler, H, Daublebsky von Eichhain, Chr.;
PREVENTING CONDENSATE HAZARDS IN HRSG SUPER-HEATER
PANELS; PowerGen Orlando, FL; 2002;
[Pea1] Pearson, M., J Michael Pearson & Associates Co Ltd; Anderson, R. W. Florida
Power Corp; SPECIAL REPORT: HRSGs — Questions about condensate
quenching, prestart purging, POWER Magazine, July/August 2000
[Hel1] Hellwig, Udo; Instabilität; BWK Bd 40 Nr. 7/8; 1988;
[Fran1] Franke, J, Lenk, U., Taud, R, Klauke, F.; Successful Combiend Cycle Debut –
for Novel Horizontal Flow- Vertical Tube Benson HRSG;
Modern Power Systems; July 2000;

051-080 HRSG pages .indd 78 13.07.15 11:54

 7. Contact 79

7. Contact

Christian Daublebsky von Eichhain

Knollerstr. 5
80802 München
Germany

Tel :+49 89 123 6896

FAX :+49 1805 060 334 645 16
Mobil :+49 179 10 55 144
[email protected]

051-080 HRSG pages .indd 79 13.07.15 11:54

 80 3. Operation of steam boiler

051-080 HRSG pages .indd 80 13.07.15 11:54

Titel HRSG_2.indd 6 29.07.15 13:27
Titel HRSG_2.indd 7 29.07.15 13:27