Guidelines On The Effects of Cycling Operation On Maintenance Activities
Guidelines On The Effects of Cycling Operation On Maintenance Activities
Guidelines On The Effects of Cycling Operation On Maintenance Activities
Technical Report
Remember to
replace artwork
and delete this
box
10709367
10709367
Guideline on the Effects of Cycling
Operation on Maintenance Activities
1004017
EPRI • 3412 Hillview Avenue, Palo Alto, California 94304 • PO Box 10412, Palo Alto, California 94303 • USA
800.313.3774 • 650.855.2121 • [email protected] • www.epri.com
10709367
DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN
ACCOUNT OF WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH
INSTITUTE, INC. (EPRI). NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE
ORGANIZATION(S) BELOW, NOR ANY PERSON ACTING ON BEHALF OF ANY OF THEM:
ORDERING INFORMATION
Requests for copies of this report should be directed to EPRI Customer Fulfillment, 1355 Willow Way,
Suite 278, Concord, CA 94520, (800) 313-3774, press 2.
Electric Power Research Institute and EPRI are registered service marks of the Electric Power
Research Institute, Inc. EPRI. ELECTRIFY THE WORLD is a service mark of the Electric Power
Research Institute, Inc.
Copyright © 2001 Electric Power Research Institute, Inc. All rights reserved.
10709367
CITATIONS
Principal Investigator
A. Shibli
The report is a corporate document that should be cited in the literature in the following manner:
Guideline on the Effects of Cycling Operation on Maintenance Activities, EPRI, Palo Alto, CA:
2001. 1004017.
iii
10709367
10709367
REPORT SUMMARY
Cyclic operation can result in an increase in forced outage rates, higher operation and
maintenance (O&M) costs, and further wear and tear on components due to additional overhauls
and maintenance. Such operation may also increase unforeseen costs due to additional personnel
training requirements and the use of more sophisticated evaluation and inspection techniques.
These guidelines are intended to help plant operators and engineers assess the impact of cycling
on maintenance activities and take appropriate preventive measures when operating a plant in
cyclic mode. While the guidelines provide important direction on O&M practices when shifting
from baseload to intermittent operation, they do not set out to be a comprehensive listing of
individual plant maintenance activities.
Background
The severity of cyclic operation affects boiler, turbine, electrical, and auxiliary components. The
effect is largely design dependent, and older plants originally designed for baseload usage fall
into the less tolerant category. Such units were designed with heavy section headers and piping
with a poor response to thermal fatigue, which basically results from temperature changes during
startup and shutdown. Cycling tends to exacerbate such problems and will lead to an increased
incidence of stress corrosion and corrosion fatigue of feedwater heaters, economizers, and
turbine units. Clearly, cycling utilities need to understand that events which are tolerable on an
occasional basis during baseload operation—in terms of damage to the plant, risk to staff, or
impact on the local environment—would be quite intolerable if occurring daily. EPRI sponsored
development of these guidelines to provide direction for all cycling plants, but particularly those
shifting from baseload operation to intermittent operation.
Objective
To develop guidelines addressing the effects of cycling operation on maintenance activities.
Approach
Because of the range of technical issues involved, EPRI engaged a multi-disciplinary team to
develop these guidelines. Investigators based the guidelines primarily on United Kingdom
maintenance experience for fossil steam plants with drum boilers, as revealed in a survey of plant
operators and R&D organizations. Information from other countries—including Ireland, Italy,
France, Portugal, South Africa, Hong Kong, and the United States—was also acquired through
surveys. Included in the survey demographics was a large population of once-through boiler
owners. Finally, in developing the guidelines, investigators also relied on their own knowledge
and understanding of plant issues as well as published literature.
v
10709367
Results
These guidelines focus on a number of key issues connected with O&M activities which are
critical to successful cyclic operation. These issues include
• Minimization of plant damage and optimization of operation, with emphasis on turbines and
boilers
• Operability enhancements
• Problems experienced under two-shifting duty
• Engineering modifications to facilitate flexible operation
• The impact of cycling on maintenance practices
• Industry practices as they relate to regulatory codes
Also featured in the guidelines are responses to the survey questionnaire along with interviews
concerning routine maintenance practices, condition monitoring, and maintenance scheduling.
The appendices to the guidelines document several new and interesting developments in repair
techniques.
In all, the guidelines emphasize that cycling mode presents challenges in the way utilities view
O&M procedures. Under cyclic operation, nearly every O&M procedure must be analyzed in
detail and almost certainly modified to ensure safety, cost-effectiveness and economy,
reliability/repeatability, and minimization of plant damage.
EPRI Perspective
These guidelines are part of EPRI’s development efforts under Target 69, Plant Maintenance
Optimization (PMO). The PMO mission is to lead the industry by developing and demonstrating
products and services that improve the use of power plant maintenance resources and increase
profitability for generation businesses.
Flexible cyclic operation of large coal-fired units has been successfully carried out while
maintaining high plant availability, without excessive additional costs. These guidelines will help
utilities increase plant availability by taking a cost-effective, systematic approach to preventive
maintenance activities when operating units in cyclic mode.
Keywords
Cycling operation
Plant maintenance optimization
Preventive maintenance
vi
10709367
ABSTRACT
This overview is mainly based on the UK experience for fossil steam plants with drum boilers.
However, information from other countries including Ireland, Italy, France, Portugal, South
Africa, Hong Kong and the United States has also been utilized with this data including a large
population of once through boilers.
Some new and interesting developments in repair techniques have also been included in the form
of Appendices.
vii
10709367
10709367
DEFINITIONS
· Two-shifting means synchronising and desynchronising from the grid system once per day,
on a regular basis, although it can also imply a week end shutdown.
· Intermittent operation can mean anything from a few hours of shut-down from time to time,
to long periods when the plant is cold but available to be called into operation at agreed
notice.
· Load following is varying load to match grid requirements and requires ramping up and down
between unit peak capacity and minimum load capability as required. Many plants which are
two shifting will also be required to load follow, rather than running at a constant rate during
the two shift period.
ix
10709367
10709367
ACKNOWLEDGMENTS
This is to acknowledge the input from a number of European and U.S.A. colleagues in the
preparation of this Report and in the provision of information contained herein. Responses from,
and discussions with, a number of plant operators has provided further insight in to the
maintenance practices carried out in countries outside Europe and North America. This input is
gratefully acknowledged.
xi
10709367
10709367
CONTENTS
xiii
10709367
6 OPERABILITY..................................................................................................................... 6-1
xiv
10709367
11 INDUSTRY PRACTICES/CODES.....................................................................................11-1
13 CONCLUSIONS ...............................................................................................................13-1
xv
10709367
10709367
LIST OF TABLES
Table 10-1 Variation of Valve Maintenance Intervals Detailed by Some Stations ...................10-3
Table 10-2 Variation in National Boiler Inspection Intervals....................................................10-4
xvii
10709367
10709367
1
INTRODUCTION
Power plant maintenance was originally controlled in the UK by the 1926 Factories Act which
stated that boiler plant should be inspected within a time interval of 26 months. For power plant
this implied a major shut down every two years and the opportunity to link this in with general
maintenance. This changed with the Pressure Systems and Transportable Gas Regulations of
1989, which handed over responsibility for the time interval to the Inspecting Authority. The
impact of this change has been rather less than might be expected, since although in principle the
period between the boiler unit itself has been extended, the need to do maintenance on ancillary
equipment such as valves, turbines, pumps and alternators has tended to shorten this interval.
The other factor, which has tended to prevent inspection intervals being increased, is the switch
from base load to cycling or two shifting. To combat this, companies are turning to more refined
approaches to estimating the impact of cycling on the life of high temperature components.
This Report is intended to help plant operators and engineers assess the impact on maintenance
and take appropriate preventive measures when operating a plant in cyclic mode. This is done
with specific reference to European plant operating and maintenance practices. The Report
therefore provides guidance as to what needs to be done when moving from base load to
intermittent operation. However, it does not set out to be a comprehensive guide as to what
should be done on individual plants.
Any station moving from base load operation to intermittent operation will have done cold, hot
and warm starts, and will have procedures for so doing. In moving to more frequent starting and
shutting down, as in the case of cyclic operation, these procedures must be analyzed in detail and
almost certainly modified to ensure:
· Safety
· Economy
· Reliability/repeatability
· Minimization of plant damage
· Environmental acceptability
What is tolerable on an occasional basis, during base load operation, in terms of damage to the
plant, risk to staff or impact on the local environment would be quite intolerable if occurring
daily. Some of these problems can originate from an unprofessional approach to maintenance.
For example, the fact that a safety valve tends to lift during shut down, would be acceptable if it
occurred once or twice a year after a long period of base load operation. It would be another
1-1
10709367
Introduction
matter if the safety valve were to blow once a day. This type of issues are addressed in this
Report.
Coal fired units vary immensely in design. Different coals have significantly different
characteristics. Hence it is only possible to offer generalizations here. Whatever the design,
intermittent operation should be possible. In the UK, 660MW coal-fired units have been
regularly two-shifted, and some 500MW coal-fired units have been cycled more than once per
day on a regular basis. The same pertains to oil and gas fired steam units which of course are
spared the problems associated with handling a solid fuel and its waste by-products.
Much of the emphasis, hitherto, on two-shifting, has been with thermal stress and fatigue of large
plant items operating at high temperature. In short, two-shifting has been regarded as another
facet of normal plant operation, where a unit may be cycled just a few times a year. Nevertheless,
two-shifting can bring other problems too. For example, superheaters and reheaters, due to the
need to start up “against the clock”, can experience temperature excursions significantly above
design.
The UK represents a good example of how attitudes have changed towards the operation of
generating plant and the ability to two-shift it efficiently and quickly. The original specification
for 500 MWe plants, built in the 1960’s and 70’s, required that units be capable of both
continuous base load and two-shift operation. In terms of two-shifting, after a six-hour overnight
shutdown, when the temperature was expected to be of the order of 480°C, units were to be
capable of being brought to full load within 60 minutes. It was also essential that the temperature
of the steam to the turbine, on start up, should not be lower than the turbine metal temperature.
From ‘synchronized no load’, the boiler was to be capable of attaining its full rated capacity in
20 minutes, with a steady rise in superheater outlet conditions. In practice these objectives were
never fully achieved. A more typical figure for hot start was 100 to 150 minutes, with loading
rates of 10 MW/min, giving a time to full load of around 50 minutes. That is, 150 to 200 minutes
from request to full load. Improved operation was needed and two-shifting trials were carried out
in the 1970’s and 1980’s to assess two-shift operability. It was concluded that faster two-shifting
was possible, albeit with a need for some changes to plant equipment and instrumentation.
Since that time, privatization of the electricity supply industry in the UK in 1989, combined with
the competition from new low cost (and hence base load) CCGT plant, has acted as an incentive
for further improvements. Today, a typical 500 MW machine can be brought on line within
about 35 minutes of notice and run up to full load within about 40 minutes. That is about
75 minutes from request to full load. It is probable that some units achieve 60 minutes. This
reduction in time has been achieved by a combination of further modification to units, including
inter-stage drains, improved instrumentation of critical components, improved automated control
and anticipatory sequencing using current computer control software. (Note: Inter-stage drains
are usually located between the various sections of the superheater in the boiler to promote
progressive establishment of flow through the boiler as the boiler is fired. A higher firing rate is
possible allowing condensate to boil out in the inverted sections whilst the flow prevents
overheating of the hotter sections. A bypass drain can usefully be installed around the HP turbine
to the reheater to provide a ‘cooling’ steam flow to protect the reheater sections which could
otherwise overheat until the turbine stop valves are opened. Critical components will include
1-2
10709367
Introduction
most of the later stages of superheater headers, possibly the main steam pipework, boiler stop
valves and HP turbine casing - basically any thick sections susceptible to the generation of
thermal gradients).
The severity of cyclic operation affects boiler, turbine, electrical and auxiliary components.
The effect is largely design dependent, and older plant, which was originally designed for base
load usage, is in the less tolerant category. Such units were designed with heavy section headers
and pipe work with a poor response to thermal fatigue which basically results from temperature
changes during start up and shut down. Hence cycling tends to exacerbate such problems.
Cycling will also lead to an increased incidence of stress corrosion and corrosion fatigue of
feedheater, economiser and turbine units which are further explained in Appendix 1.
Advice, information and experience have been sought from a number of plant owners, operators,
researchers and associated organizations in Europe, Asia and South Africa. This Report is thus
based on the reviewers’ own knowledge and understanding of plant issues, published literature,
and information accessed from plant operators and technical experts.
There are five Appendices with this Report. Appendix 1 describes some practical aspects of
thermal stress, corrosion fatigue and stress corrosion as applied to critical plant items.
Appendix 2 describes a few innovative repair processes which the user of this Report may find
helpful. Appendix 3 describes a few ‘Alternator’ related issues and Appendix 4 reproduces the
Questionnaire sent out to some of the plant operators. Appendix 5 is perhaps the most significant
appendix in that it gives an overview of the maintenance related issues concerning various
components of cycling plant and the significance of these problems.
1-3
10709367
10709367
2
OBJECTIVES OF THE STUDY
This was to assist EPRI in the above study by supplying information on maintenance practices
and activities necessitated by the cyclic operation of fossil steam plants in Europe with the
emphasis on coal fired plant.
2-1
10709367
10709367
3
METHODOLOGY USED TO CONDUCT THE STUDY
Because of the range of technical issues involved, a multi-disciplinary team was engaged in
carrying out this study. The team has conducted surveys of plant operators and R&D
organizations. The surveys were conducted through the questionnaire shown in Appendix 3 and
through interviews and discussions/meetings with plant operators and experts. The countries
involved in the survey were the UK, Italy, Ireland, Hong Kong, Portugal and South Africa. Input
to the report also includes UK experience in maintenance through ETD staff/consultants’ own
experience.
A survey of available literature was carried out with reference to sources which included:
· Ex-CEGB UK
· Institution of Mechanical Engineers, London UK
· International Conferences, Authors & Private Communications Worldwide
· EU-funded projects EC
· European Pressure Equipment Research Council (EPERC) EC
3-1
10709367
10709367
4
RELIABILITY/REPEATABILITY OF CYCLIC
OPERATION
The first consideration for the operator and maintainer must be the commercial conditions within
which the plant operates. For many IPPs (Independent Power Producers) there are narrow
timeslots for synchronization and penalties for not meeting them, or even the risk of losing the
generation opportunity. These commercial penalties are a key input into the cost/benefit analysis
of any expenditure to improve reliability for intermittent operation. Commercial opportunities
may present themselves from the ability to deliver at short notice times and fast loading and
these must be weighed against the increased risk of failing to meet the targets and possible
increased plant damage.
The first condition of avoiding commercial penalties is reliability; can the start procedures be
gone through regularly with an acceptably low percentage of failures? The second condition is
repeatability: can the procedures deliver synchronization and load-up within the target times
despite changing conditions (e.g. with wet coal and dry coal, with the plant’s best coal and its
worst, with the best shift team and the worst) or can simple variations be devised which will
cater for known different conditions?
To achieve repeatability, the procedures must allow for varying conditions and therefore the
standard procedure to be adopted for each type of start will be neither the quickest nor most
economical achievable. It is usually easier to delay the late stages of a start-up that is going well
so as to avoid being too early, than to try to accelerate a too ambitious start procedure. However,
some designs of steam turbine run into differential expansion problems if run-up or load-up is
delayed.
The reliability of a unit depends on the product of the reliabilities of the individual critical-path
items. Given the large number of critical-path items (e.g. switches, actuators, motors, valves) on
a big unit, each critical path item needs a very high reliability for the overall reliability to be
satisfactory RCM (Reliability Centered Maintenance) practices can be applied to focus
predictive and preventive measures on critical components to better meet a unit’s mission at
minimum cost and highest availability (footnote reference to appropriate EPRI report #).
4-1
10709367
10709367
5
MINIMIZATION OF PLANT DAMAGE AND
OPTIMIZATION OF OPERATION
Reliable instrumentation in the right place with output made available in the control room is the
major requirement to minimize damage. Very often the original instrumentation package
provided by the plant manufacturer is not adequate for intermittent operation. Deducing plant
conditions (e.g. turbine differential expansion) from an envelope of input conditions and time
can be very risky when abnormal conditions arise. However, instrumentation is expensive and it
may be economically necessary to install extensive instrumentation on only one of a type of unit
so that the other units may be instrumented only in areas found to be critical.
Turbine
As a sweeping generalization, steam turbines can usually be cycled without significant damage
if attention is paid to temperature matching and drainage. For a running regime involving
shut-downs of up to 8-10 hours (often the definition of “hot start” for turbine starting), the best
procedure is usually to deload the turbine quickly, thus preventing significant cooling. During
start up steam temperatures should be matched on the positive side with the temperature of the
turbine. That is the steam before inlet valves should be towards the upper limit for matching, as
set by the manufacturer. This temperature should take into account the throttling loss through the
valve (i.e. typically about 50oC above the turbine metal temperature). This will permit the turbine
to be run up and loaded up.
Drainage is critical, as water in the wrong place can do severe damage. Reliance should not
be placed on steam traps or opening drains for a specific period of time. The ideal drainage
arrangement on important steam lines is for two drain valves (one for isolation, one for
controlling the drainage) in series together with a reliable thermocouple. Drains should not be
closed until a temperature margin above saturation has been achieved, and then may need to be
blown through again as a precaution at a later stage of the run-up/load-up. The length of drain
line between the item being drained and the first drain valve should be minimized to prevent
blow-back of collected water onto hot components in the event of a trip or sudden deload.
Drain locations will vary from plant to plant and will be provided by the Original Equipment
Manufacturer. All that is being suggested here is two drain valves in series on each drain line,
one to be used for isolation and the other for throttling the drainage (the latter one clearly subject
to throttling wear, hence the need for an isolating valve to minimize on-load losses).
The thermocouple should be placed where it will give a reading representative of the steam
temperature in the drain line. A bypass line controlled by an automatic drain valve may be
desirable in some locations. Some manufacturers have been known to rely on automatic drain
5-1
10709367
Minimization of Plant Damage and Optimization of Operation
valves only for e.g. turbine casing drains, but this is not recommended by the authors of this
report.
Thermocouples should be provided to allow differentials to be read between the inside and
outside of thick metal components such as the main steam valve chests and strict limits observed.
The limits were traditionally (pre ~1980) set based on manufacturers’ experience – 100oC
(180oF) was a typical figure. These were aimed at achieving a certain design life (e.g. a particular
number of hot, cold or warm starts). Where the manufacturer does not provide a limit or the
operator desires to change from the manufacturer’s design criteria (e.g. by changing the speed or
frequency of starts), an independent engineering study will be needed. These days of course, it is
possible to come up with a scientific value by the use of finite element analysis when it should be
possible to optimize the through wall limit. This technique can also be used to optimize the
geometry of the component to improve its tolerance to temperature differentials.
Top to bottom temperatures for turbine HP and IP cylinders may be desirable. Bearing
temperatures and vibrations are essential, and differential rotor/casing expansions highly
recommended. A reading of the critical metal temperature for steam matching must be available,
together with a fall-back temperature measurement that can reliably be used to deduce the critical
temperature should that thermocouple fail.
The failure mechanism is thermal fatigue and creep fatigue. A common failure mechanism would
be the overheating of the inner surface of a thick metal section relative to the outer section. The
outer section constrains the expansion of the inner section, causing the inner section to go into
plastic deformation (yield) in compression. When the section temperature is re-equalised (on
reaching steady state temperature or during or after cool-down), the inner surface will be in
tension. If the tension is sufficient, the inner surface may then crack under the tension. Once
cracks are formed, further temperature cycling (even within the normal limits) may propagate the
cracks leading to eventual failure by loss of cross-section. Many cracks will not propagate, but
this must be established by monitoring and/or engineering analysis.
Turning gear to prevent bowing of the shaft during cooling down and jacking oil pumps to lift
the rotor shafts off the bearings will spend much more time in service and have many more
starts. Attention should be given to provision of spares and to eliminating poor reliability or
excessive manual input requirement. An alternative means of turning the shaft in emergencies is
desirable. There should be a standard procedure dealing with failure to go on turning, including a
time limit beyond which attempts to re-establish barring will not be continued.
Boiler
Boilers vary considerably in design and, together with the fuel/combustion system, are most
critical and vulnerable for intermittent operation; they go through big heating/cooling and
expansion/contraction cycles. Initial water wall flow distribution can be poor. Superheaters
will accumulate water. Mills will go through the coal/air explosive range on each start-up and
shut-down.
5-2
10709367
Minimization of Plant Damage and Optimization of Operation
The following are checkpoints, but there is no substitute for a thorough engineering study,
including additional thermocouples and a test program by an experienced engineering team;
many such are available commercially.
Careful consideration needs to be given to each header to ensure the inner to outer temperature
limits are not exceeded. The economizer inlet header tends to be vulnerable as it will stay hot
during a short shut-down, but may be chilled during the start-up until the feed heaters come into
service. On drum boilers, the drum is generally not a limiting factor for hot starts unless
excessive feeding is required such that only cool water is available.
Drainage
Water will accumulate in the tubes of pendent and platen superheaters. It is possible to monitor
this by thermocoupling in the dead spaces which is much cheaper than trying to thermocouple
on the tubes in the furnace. (Note: Dead spaces are the regions of the boiler within the casing,
but outside the gas flow region where there is no heat transfer into the steam/water flow.
For example, the ‘Top Dead Space’ is the region above the furnace roof where headers and
interconnecting tubes are located; a small dead space usually exists at the throat of the furnace).
Direct drainage is not possible, therefore the rate of firing must be controlled until all the water
has boiled off and a cooling steam flow is available to all tubes. Lower headers on other
superheater banks will require a routine of blowing down.
Testing will be necessary to determine the optimum program of drum level control, starting
with the level to be achieved when the boiler is boxed up on the previous shut-down; this must
maximize the amount of hot water available for the restart but avoid the need for expensive
blowing down to control the level as the water swells under the start-up firing. The need to run
a feed pump to top up during the shut-down should be avoided if possible. All steam and water
valves on the boiler must be maintained such that leakage and loss of pressure and water is
minimized. Operators must make proper use of master/slave valve arrangements.
In a natural circulation boiler, there may be difficulties in getting even flow distribution
established and this will be exacerbated by uneven slag formations. Uneven flow distribution
can lead to differential expansion problems between tubes, overheating of tubes, stressing of
membrane walls (where applicable), overstressing of tube attachment clips and stress on tube to
header stubs; these must be contained by limiting the rate of initial firing. Steam division walls
and furnace wall superheaters and reheaters are particularly vulnerable to overheating and
failure.
5-3
10709367
Minimization of Plant Damage and Optimization of Operation
The main boiler outlet isolating dampers must be maintained so that they can be operated
reliably and kept gas-tight to minimize heat losses during shut-down. However, regular checks
during shut-down to ensure that start-up burners are not leaking oil or gas into the furnace are
strongly recommended, and there must be an effective furnace purging procedure before any
attempt is made to light the boiler.
Desuperheater Sprays
Superheater and reheater desuperheater sprays are a potential source of dangerous leakage into
the boiler and steam leads. Desuperheater control valves should not be relied on for isolation but
must be kept in good condition so that controlled low flows can be achieved during start-up.
Spray isolation valves should be shut until the desuperheater spray is required and the use of the
feed pump during shut-down should be minimized.
Sootblowing
The regular expansion and contraction of the boiler will often help to control slag build-up.
Selective blowing of either the water walls or the superheaters may be necessary before shut-
down to assist in achieving a good pressure/temperature raising balance on start-up. Airheaters
are likely to accumulate some unburned fuel during start-up and should be blown at the earliest
opportunity. Sootblowers supplied from reheat steam may need an alternative superheater source
to allow early use during start-up.
Some units are provided with turbine bypass systems which greatly facilitate drainage for
temperature raising compared with those that rely on superheater and leg drains. In either case,
superheater drains are likely to be used for initial temperature raising until the main steam lead
metal temperatures can be matched and flow established to turbine. For regular hot starting, drain
line sizes (and the capacity of the downstream systems) may need to be increased as, in general,
the more the flow the greater the rate of temperature raising. Boiler pressure needs to be kept
down to increase the temperature raising and to minimize the throttling loss that must be allowed
for establishing steam flow to the turbine. However, sufficient pressure must be maintained to
allow block load to be applied to the turbine on synchronization.
UK practice has been to discharge the main start-up drains from the legs and the boiler into
blowdown vessels. The steam is released to atmosphere. The water from the boiler drains is led
to waste and the water from the leg drains is recovered (the difference being in the chemical
purity). The mainland European practice has been to take the steam leg drains via a throttling
valve into the main turbine condenser thereby condensing the steam and recovering the water
when the desired purity was achieved. The bypass systems were generally designed to open
quickly in the event of the generator becoming detached from the grid system and allowing the
5-4
10709367
Minimization of Plant Damage and Optimization of Operation
boiler/turbine unit to be stabilised at a low load, supplying its own auxiliaries. Thus the bypasses
served two functions.
A major consideration in plant start-ups and shut-downs is to avoid overheating or chilling thick
metal components. Hence, for a hot start, the steam temperature at the boiler final superheater
outlet must be raised close to the temperature of the main steam legs to the turbine before steam
is admitted to the legs. Then steam will have to be passed down the steam legs into drains until
the steam temperature at the turbine end of the legs matches the metal temperature (depending on
the exact layout of the plant) between the throttle valves and the stop valves in the steam chest.
Again depending on the plant design, the next metal temperature to match will be a
representative temperature (probably the inlet belt metal) in the hottest section of the turbine,
allowance being made for the temperature loss caused by throttling through the turbine
admission valves.
The pressure/temperature raising characteristics will depend greatly on the balance between
radiant and convection superheaters in the boiler (for each fuel type). Experimentation will be
needed to determine the target boiler pressure to be achieved at shut-down to optimize the start-
up, though this will need to be modified to allow for passing valves and dampers. A water/steam-
tight boiler will retain its pressure better and will not need to be topped up with cold feed water .
Similarly if the dampers pass, there will be a cooling draft through the boiler to conduct the heat
away.
In general, less stress is caused to a hot steam turbine by running it up and loading it quickly
rather than slowly. The problem, particularly with coal-fired boilers, is to achieve a fast, but
steady increase in firing and to ensure that the firing is matched to the steam flow (to avoid steam
and boiler metal temperature excursions). The steam boundary layer effect in the superheaters
means that maximising the steam flow through the boiler is necessary to achieve the desired
steam temperatures.
To ensure that the desired run-up and load-up rate of the steam turbine can be achieved, it is
useful to have stored energy in the boiler (in the form of high pressure) so that delays in
increasing the firing rate (perhaps because of problems starting a mill) do not immediately
require reduction in the turbine loading rate.
There are three factors which limit the desirable boiler pressure during start-up:
· one is that if the firing rate increases faster than targeted, the boiler safety valves will be
lifted causing a number of undesirable effects.
· The second is that the higher the boiler pressure, the more throttling is necessary to control
the turbine run-up and load up rates. The increased throttling increases the temperature drop
in the steam. Since most boilers have difficulty in achieving high steam temperatures at low
throughputs, increased throttling will exacerbate the problems of steam to metal temperature
matching.
· The third is that the specific heat of the steam is increased at higher pressure, making it
harder to achieve the desired steam temperatures.
5-5
10709367
Minimization of Plant Damage and Optimization of Operation
There is therefore an ideal boiler pressure to be achieved at the start of the steam to legs, steam to
turbine process. To achieve this ideal pressure at the end of the initial boiler fire-up process most
economically, a target boiler pressure should be established to be achieved when the boiler is
boxed up at the end of the previous shut-down. The relationship between the ideal start-up
pressure and the shut-down pressure is determined by the shut-down time, the cooling and
leakage rates of the boiler during shut-down and the pressure/temperature raising characteristics
of the boiler during start-up.
It will generally be found that different mills have a markedly different effect on the rate at
which furnace and steam temperatures can be raised. Because of the different heat emission
characteristics of oil and coal flames, it will probably be necessary to fire a mill as soon as it is
safe to do so. However to prevent the risk explosions, pulverized coal should not be admitted to a
boiler until oil or gas start-up fuel has reached safe temperature limits.
Light-up Burners
These must be maintained for high reliability, particularly those that ignite the start-up mill
burners otherwise excessive delays and/or extensive use of back-up staff will be incurred. The
problems will depend on the manufacturer and type. Light up burners generally work better
when used frequently. Other than the fact that they are used more and therefore require more
maintenance, they may well be more reliable!
Some of the precautions to be taken and points to be considered are listed below:
· During shut-down, the lighting-up oil will typically be on recirculation at a low rate.
· During start-up, the pumping and heating will need to increase quickly and reliably to allow
the firing of a lot of burners in quick succession.
· For a hot start, a high percentage of the boiler’s light-up burners will be needed, so they need
to be reliable.
· Usually, not all the mills are useable for initial light-up and temperature raising. For safety
reasons, the early mills probably need all of the relevant oil burners to be in service; this
requires a very high reliability of these particular burners.
· Older designs of light-up oil burner involved the ignition and oil lances to be pushed forward
into position by rams as part of the starting process. Boiler casing distortion and ram failures
caused reliability problems with such burners.
· Recirculating tip oil burners avoid the need for oil lance purge sequences. Where other types
of burner are used, steam or air purging has to be effective or the burner tip may get
carbonised deposits during the shut-down and therefore fail to ignite when called for.
· Many big, coal-fired units with low NOx burners are successfully two-shifted. It should be
stated here that there is nothing unique with low Nox burners and the light up burners are
generally the same for all types of burner.
5-6
10709367
Minimization of Plant Damage and Optimization of Operation
Steam may be needed to heat and/or atomize the light-up oil, to seal turbine glands and to
operate condenser air extractors. A single auxiliary boiler may be adequate for occasional
start-ups, but may need augmentation for frequent starts. An alternative to additional auxiliary
boilers might be to install a steam supply from the unit itself or to allow adjacent units to supply
each other if not all units are shut down at the same time.
Regular shut-downs make it possible for reduction of the coal yard manning periods.
Close liaison between the coal handling staff and operations staff is necessary so that the bunker
filling can be done appropriately to the load regime. There must be enough coal in the bunkers to
allow the start-up and operation until the time when more coal can be put up, but not too much if
the bunkers are prone to losing flow. Poor quality or wet coal should not be put in the bunkers
such that it will be at the bunker mouth during the start and load-up. If such coal has to be put
up, the operations staff must be made aware so that they can modify their procedure or allow
more time.
Mills
Here the avoidance of fires and explosions and coal level control are the key issues. Good
operating practices must be maintained. Modern control systems and regular attention to the
level sensing devices may be necessary to avoid over and under filling. However, since through-
put is lower with intermittent operations than with base-load, lower maintenance or wear parts
can be expected. But potential problems with frequent starts and stops must closely observed.
Switchgear
Some high voltage switchgear, e.g. oil-filled, has a high maintenance requirement if used
frequently. Vacuum-filled circuit breakers need much lower maintenance and are generally
easier to isolate and earth (the opportunity of frequent shut-downs might be taken to do more
maintenance). A cost/benefit analysis for replacement of older switchgear should be considered.
Any unreliable low/medium voltage switchgear should also be considered for replacement.
Feed Heaters
These must be stable (not prone to flooding) and capable of being put into full, cascaded service
without any local manual intervention. The high pressure heaters will usually have very thick
section tubeplates which will stay hot on a short shut-down; the stress levels must be controlled
on start-up by careful admission of the feed water, which will be relatively cold.
5-7
10709367
Minimization of Plant Damage and Optimization of Operation
Dust Plant
If the fly ash dust is sold, it may be necessary to dispose of the dust created during start-up
separately because the initial boiler combustion will be poor, leading to carbon in dust levels
which may not be acceptable to the purchaser.
Sustained low feed flow rates will need to be achieved during start-up. This requires careful
maintenance of the main valves or the provision of a start-up valve bypassing the main valves.
The usual failure mechanism is cutting of the seat through sustained, heavy throttling with the
valve nearly closed. Vibration can be a problem. The important issue is to have a valve that is
specifically designed for this sort of application, but maintenance is important and problems can
be alleviated sometimes by e.g. hardening of the seats. A small, start-up feed regulator valve is
recommended.
This must be able to raise vacuum reliably and repeatably despite the strong possibility of greater
than normal air in-leakage arising from the more frequent expansion and contraction of the
relevant joints. Vacuum raising should be fast enough not to be on the critical path for the start-
up.
5-8
10709367
6
OPERABILITY
Unless labor is cheap and can be deployed plentifully, all valves and dampers must be power
operated and with control and indication on the operator’s panel, likewise electric motors.
All key instrumentation must also be readable at the operator’s panel.
The next major step is to make it possible for the operator to be able to take an overview and
monitor the key elements of the process and not be overwhelmed with the volume of information
presented and number of individual actions he has to perform. However many staff are deployed
to operate the unit, one person must have the overview and be able to spot the variations that
could lead to commercial loss or plant damage. Here, modern control systems and VDU
(Visual Display Unit) displays can make an immense difference compared with old control
systems and chart recorders. To give some examples:
· All the turbine bearing vibrations and shaft eccentricities can be displayed on a single,
histogram VDU display where the color of each block changes to yellow then red when its
value approaches the warning or danger level.
· A target critical path line can be displayed on a screen for start-up, run-up and load up, with
the actual achievement also displayed live during the process. In this way, the operator
quickly becomes aware of variations and can make timely corrections.
· Elements of the task can be combined into single-button-start subroutines, e.g. firing mills,
commissioning feed-water heaters, steam to turbine/generator set and run-up.
· Alarms can be graded according to their relevance at any point in the process. Early
computerized systems often churned out huge amounts of irrelevant alarm state information.
Complete replacement of existing control and information systems can be very expensive, but
there are now available commercial systems that will extract and process information from the
existing system and provide high level information and control facilities.
The demands for tight combustion control to meet environmental standards are likely to
necessitate a modern, automatic system.
6-1
10709367
10709367
7
OTHER CONSIDERATIONS
Noise
Starting up steam plant usually creates high noise levels over a sustained period. Depending on
the situation of the plant, noise measurements and modifications may be necessary. Neighbors
may need to be informed of the changed operating regime and its implications.
Efficiency
Once at steady load, the unit will usually need a period of optimization before the usual
efficiency is achieved; this needs attention to minimize cost increases. Pressure may need to be
applied to operators, whose major task has been to get the unit started and to the required load on
time, to continue the effort until the efficiency has been maximized.
Variable or sliding pressure control is a possible option to reduce heat rates during ramp up and
ramp down. Here the inlet pressure and flow rate to the turbine is controlled by altering the
pressure on the feed pump to the boiler. This enables governor valves to be set fully open
avoiding throttling loss. There will also be some reduction in feed pump power.
This option is more feasible on once through systems, which can respond quickly to changes in
output. With drum type systems, if the feed pump pressure is reduced too quickly, there is a risk
of the boiler priming. Furthermore the drop in pressure can lead to a temporary cessation of flow
in the boiler. In such situations hybrid operation may be feasible. Here the unit is operated with
variable pressure from about 60-85% of rated output. Above this rating the system is run under
constant pressure or throttle control.
Availability of the unit will probably suffer during the early period of flexible operation and
boiler tube leaks will need systematic analysis. However, many large flexing units achieve high
availabilities. This is interpreted as a ‘shake down’ effect - when you push a machine to its
limits in new ways there will be an increase in certain types of problem - in this case thermal
fatigue of attachments is typical. Once the problem is identified it is possible to do something
about it by modifying the operation, improving the design or simply setting higher standards of
maintenance. Most of the UK units encountered increased rates of attachment weld failures when
they started two shifting - this was overcome by modification of attachments and application of
rigorous quality control in welding detail. Attachment failures are now at acceptable levels.
7-1
10709367
Other Considerations
Maintenance
The shut-down periods create opportunities for doing optional or forced maintenance on more of
the plant than is available when on load. Plans for taking advantage of the opportunities need to
be made. Procedures for fast depressurization to allow pressure parts repairs should be made
available. Opportunities should be taken to inspect suspect areas of the plant. The statutory
pressure parts inspection regime may need to be changed, at least until more experience of
flexible operation is gained. This may be in conflict with the return to service lay-up procedures
aimed at minimizing start-up time and wear and tear on components so a balance has to be
reached. A Reliability Centred Maintenance (RCM) program enhanced by inspection data and
operating history that incorporates risk based inspection methods can help to prioritize
maintenance activities to match a selected operational regime. RCM is a technique for ensuring
that maintenance expenditure is focussed on achieving the desired plant performance.
Chemistry
Periods of shut-down allow aeration of some of the condensate and contamination from transient
or small condenser leaks. Improper shutdown and layup can lead to serious plant damage such as
pitting, oxidation and corrosion, which during operation can be the indicators of the major failure
mechanisms affecting plant availability. Damage due to cycling includes boiler tube failures,
condenser leaks, and low pressure turbine blade problems. The boiler and feed dosing regimes
will need reconsideration, and condensate dumping may need to be carried out before it can be
returned to the feed system. Silica and conductivity limits prescribed by the turbine
manufacturers for the steam may take too long to achieve on a regular basis; polishing plant
and/or temporary relaxation of the limits may be necessary. Poor shutdown and layup exacerbate
the problems of higher levels of corrosion products flowing from the feedwater system and
within the boiler.
8-10 hour shut-downs are likely to increase the demand for purified water for additional
blowing-down and drainage. Additional storage and/or production or purchasing capability may
be necessary. The choice of chemistry for a unit is important, not only when the unit is operating,
but also during the transient conditions of shutdown and startup according to the above
referenced EPRI Guidelines. For instance: the choice of oxygenated treatment (OT) over all
volatile treatment (OVT) for drum and once-through units reduced markedly the level of
corrosion products during startup; the choice of EPT (equilibrium phosphate treatment) over
CPT (congruent phosphate treatment) should remove the possibility of hideout and hideout
return during startup and shutdown respectively.
Regular starts and stops exercise protection equipment such as water level detectors and help
prevent inoperability from non-use. However, this regime also puts the plant into risky
conditions more often where the protection may be required. Therefore a regular schedule of
7-2
10709367
Other Considerations
protection testing is required. To some extent this can be combined with the operating regime,
by noting alarms as they routinely come up and by shutting down the plant on a routine basis
using the different trip functions.
As much auxiliary equipment as possible should be shut down when the unit is off load to save
works power. The procedures and maintenance should try to avoid the need for occasional
running of a feed pump during shut-down. Some of the larger systems, such as fuel oil or
Circulating Water (CW) will probably require too much effort and create too much additional
start-up risk to be worth shutting down for short unit shutdowns, but the number of pumps and
heaters in service should be minimized. It may be possible to maintain the CW system prime
with a special small pump rather than keeping a main CW pump running. Consideration should
be given to whether it is more economical to supply the auxiliaries of a shutdown unit from a
running unit, if available, rather than from an external grid source.
7-3
10709367
10709367
8
PROBLEMS EXPERIENCED UNDER TWO SHIFTING
DUTY
When flexible load operation is practised, the nature of plant failures can change. Long term
creep damage can become less of a concern, and fatigue type problems become more evident.
In high temperature components, the interaction between creep and fatigue can result and
exacerbate damage accumulation. Thus it becomes imperative to monitor major components
such as headers and other thick sectioned components.
Mechanical problems arise due to thermal gradients and to restricted expansion etc. Problems
ranging from boiler tube attachment failures to more complex problems affecting banks of tubes
can ensue. Similarly, turbine rotor expansion becomes more critical with gland sealing and rotor
blade clearances a concern.
It is impossible to identify the precise effects of cycling on plant due to the extensive variations
in the design of components and operating conditions. However, typical examples of actual
experience are summarised below. A few examples of unusual failures are also included to
highlight how design specific the impact of cycling can be. It should be noted that some of the
examples are from units with broad differences in cycling characteristics, from load cycling to
two shifting.
Attemperators
· Liner cracking
· Liner attachment weld cracking
8-1
10709367
Problems Experienced Under Two Shifting Duty
Tubing
· Typical cycling related failures e.g. at tube to tube attachments, at windbox welds, openings
etc., corners, (with and without corrosion influence).
· One unusual failure type involved localised hot spots at reheater bends (12CrMOV material)
due to oxide spalling of 1Cr0.5Mo lodging at bends – temperatures were particularly high at
start-up and cycling promoted oxide spalling which located at tube bend.
· Several instances of pendant superheater and reheater tubing replacement due to excessive
temperatures during start –up. Usually solved by upgrade replacement.
· Although not common one respondent cited corrosion fatigue failures (explosive) at neutral
axis of horizontal superheater drain bends (condensate gathered and caused pitting).
Steam Drum
· Cracking at penetrations e.g. risers, thermocouples etc., section changes at doors, attachment
welds etc. Chamfering of openings used to reduce stresses. Grinding and monitoring usually
adopted.
ID Fans
· Cracking at impeller welds.
Rotors
· Cracking reported at HP blade slots
· Cracking at gland steam heat relief grooves of some designs – removed via reprofiling
grooves.
· Corrosion fatigue cracking at section changes on disced LP rotors.
8-2
10709367
Problems Experienced Under Two Shifting Duty
· LP blade root hole cracking on some designs (finger and slot in Alstom designs – not certain
to be cycling related as similar cracking reported on baseload plant).
· Blade cracking at lacing wire holes and erosion shields.
Valves
· Internal Cracking at section changes, webs, thermocouple holes etc. Usually leave and
monitor or grind out and improve profile if possible. Craze cracking usually not considered a
problem.
· External cracking at section changes.
· Cracking of hard facing (stellite) on seats.
Inner Casings
· Cracking at section changes, bolt holes, openings etc. general policy is to grind out if minor
and improve profile and monitor. Some weld repairs reported.
Feed Heaters
· Cracking at hemiheads and at baffle plates and a few instances of small (fatigue?) cracks at
openings in shell resulting in brittle fracture of shell (UK) – generally fairly long time to
materialise. Problems with some designs resulting in erosion and fatigue related leaks.
8-3
10709367
10709367
9
ENGINEERING MODIFICATIONS TO FACILITATE
FLEXIBLE OPERATION
Older base load plant was not designed with the expectation of flexible operation. The majority
of UK plant is sub critical type with forced and natural circulation evaporative sections. Turbines
were designed with throttle control suitable for high load operation.
In order to meet the requirements of flexible operation, various modifications have been
implemented. The most notable modifications include:
1. Improved instrumentation and integrated system control to give better indication of plant
condition and performance.
2. Improved remote control of plant to meet the rapid changes required by load variation and
facilitate prompt and safe operation.
3. Improved boiler venting to assist progressive warming through of a boiler, steam legs and
turbine during two shift operation.
4. Improved thermal insulation to reduce thermal effects due to cooling in overnight shut down
periods.
5. Improved light up burner reliability and operability. This is especially the case for coal fired
units where smaller gas/oil burners are used for coal burner ignition and flame stabilisation.
6. Boiler Off Load and Economiser Recirculation to maintain an even temperature distribution
and reduce thermal stresses.
7. Boiler hot filling systems to facilitate hot water to be transferred from an operating boiler to
one which is off load and reduce the thermal shock to economiser inlet etc.
8. HP Turbine Bypass to provide a flow path for steam through the cold reheat steam legs and
into the reheater. and thus enable higher firing rates.
Whilst these modifications are aimed at improving operability, they have an impact on plant
maintenance either by their increased importance in the operation of the plant or by reducing
the duress experienced by the plant. Thus instrument maintenance and oil burner reliability are
now more critical in the reliability of the plant than when steady state base load operation was
practised.
9-1
10709367
Engineering Modifications to Facilitate Flexible Operation
A general indication of plant problems associated with flexible plant operation was given in the
ETD previous report on ‘Damage to Power Plant Due to Cycling’ in Appendix P, a copy of
which is attached here as Appendix 3. A general vulnerability index of the various components
to damage and need for maintenance has been added to this Appendix.
9-2
10709367
10
IMPACT OF CYCLING ON MAINTENANCE PRACTICES
– RESPONSES TO THE QUESTIONNAIRE AND
INTERVIEWS
A questionnaire (see Appendix 4) was sent to a number of utilities and the analysis of the
responses received is described below.
10-1
10709367
Impact of Cycling on Maintenance Practices – Responses to the Questionnaire and Interviews
Responses suggest that the type and frequency of condition monitoring of rotating components is
quite variable. Vibration and temperature monitoring, oil sampling, thermography are frequently
used on rotating parts such as the turbine and generator rotors and pumps, motors etc. The larger
components are typically monitored on a continuous basis but with smaller pumps etc.
monitoring is on a periodic basis e.g. once a month. Thermograghy is also in use (to a lesser
extent - around 15%) on valves and feedheaters and some electrical components e.g. switchgear.
Maintenance Scheduling
Major Outages
The majority of responses revealed that turbine cover lifts i.e. major outages, are generally based
on equivalent hours (i.e. run hours + (number of starts x factor). Three of the Stations included
the multiplying factors they use. In two cases the factor was 25 for all starts and for the other
case 20 and 70 for hot and cold starts respectively.
Actual overhaul intervals quoted ranged from 65,000 to 85,000 hours. Some Stations indicated
lower intervals i.e. 50,000 hours, for Low Pressure turbines and generators. Although not
reported it is suspected that this is possibly because of specific problem e.g. stator winding
looseness or excessive LP blade tip erosion which would tend to be machine specific. It was also
noted that condition monitoring was used to justify extending the interval between outages.
Minor Outages
The minor overhaul interval is typically 2-4 years but appears to be based on boiler related
statutory requirements and on turbine stop and control valve maintenance requirements.
The variation of valve maintenance intervals detailed by some Stations is shown below:
10-2
10709367
Impact of Cycling on Maintenance Practices – Responses to the Questionnaire and Interviews
Table 10-1
Variation of Valve Maintenance Intervals Detailed by Some Stations
Operation
Inspection
Station Valve Type Operating Mode Testing
Interval
Frequency
MSSV = Main Steam Stop Valve, MSSVBV = Main Steam Stop Valve Bypass Valve, MSCV = Main Steam
Control Valve, CRV = Combined Reheat Valves
Three of the above reduced the interval by one year when moving to cycling duty. One reduced it
by two years, the other did not change the interval.
Boiler maintenance is largely time-based. This arises primarily as a result of the prescriptive
nature of most of the National pressure vessel inspection regulations, refer Table 10-2. A unified
“European” approach to pressure vessel manufacture comes into effect in 2002 and there is
10-3
10709367
Impact of Cycling on Maintenance Practices – Responses to the Questionnaire and Interviews
Some reports indicated extended scope for inspections and more focused maintenance for
cycling units. Analysis of the questionnaires indicated that almost all respondents stated that 40 –
60 % of their maintenance was corrective with the remainder preventative. Few replies indicated
the use of condition based techniques on boiler components other than with associated rotating
equipment e.g. ID Fans etc.
Table 10-2
Variation in National Boiler Inspection Intervals
NDE = Non Destructive Examination, VT = Visual Testing, UT = Ultrasonic Testing MT = Magnetic Testing,
PT = Penetrant Testing, RT = Radiographic Testing (1) as required by inspector (MT, PT, UT, RT)
10-4
10709367
Impact of Cycling on Maintenance Practices – Responses to the Questionnaire and Interviews
It appears that in most instances boiler inspection findings will influence intervals but usually are
focused to specific problem areas. Responses to the most effective fixes were inconsistent and
ranged from detailed life assessment strategy to increased instrumentation and introduction of
more automated valves. This variability in response possibly stems from the background of the
individual completing the questionnaire i.e. whether on the operational side, maintenance or
engineering. Similarly varied responses were obtained regarding problem areas - turbine flexible
joints, LP disc serration related cracking, LP blade erosion were typical of the turbine problems
cited. A few indicated water chemistry related problems, in particular phosphate hideout. Air in-
leakage problem at low loads was also mentioned, but the most common was tube failures.
Further details of problems are presented in Section 8.
Most respondents indicated some increase in O&M costs but they were not more definitive in
terms of percentages or details of high cost areas other than valve replacements and automation
aspects.
10-5
10709367
10709367
11
INDUSTRY PRACTICES/CODES
At present there are no maintenance codes in Europe. As a result most of the European plant
operators use either manufacturers’ guidelines or in-house guidelines. In the UK where the
practice is more developed and advanced in-house guidelines are used, for example, by Innogy
and Powergen. However, other countries, such as EDP/PROET in Portugal for example, use
mainly codes such as ASME (adapted for repair work). For component replacement
PROET/EDP reported that they may use the applied Code of Manufacture, which is generally
ASME (for Forster-Wheeler steam boilers, which are the vast majority). For steam turbines,
where applicable, generally ABB internal standards are used by this company.
Generally it can be stated that most maintenance regimes are set up by the OEM (Original
Equipment Manufacturer). The only exception to this is where there is a statutory requirement
e.g. in the UK as set out under the Pressure Systems regulations. Similar Statutory requirements
exist in all countries and require certain components to be examined at periods as set out by law.
Most maintenance is based on custom and practice and can vary even within one company.
Organisations with a high level of commercial awareness will carry out ‘value for money’ type
assessments and optimize maintenance practices to suite. As an example, a few years ago the JIT
(Just In Time) approach was widely used whereby you aimed to replace or repair components
just before they failed - they often missed!
European standards which take in to account cycling are not known to exist at present but five of
the general maintenance standards are being developed which cover different aspects of
industrial maintenance. All are being developed under the auspices of CEN Standards
Committee TC319. The CEN TC 319 was divided into five working groups:
Further information on these can be accessed from the CEN site “http://www.cenorm.be” or
“http://www.ini.hr/efnms”.
11-1
10709367
10709367
12
NEW DEVELOPMENTS AND RELATED R&D EFFORTS
A number of new European projects, some funded by the European Commission, have been or
being undertaken in the field of maintenance. Although, as far is known none of these deal
directly with maintenance of cycling plant, many of these activities can help a plant operator to
benefit in terms of damage monitoring and engaging in predictive maintenance. Thus
developments have been made in terms of remote temperature and strain monitoring using fiber
optics, as in the case of the European Commission funded project ‘Forms’. Concepts of total
maintenance have also been developed which integrate corrective and preventive modes of
maintenance and introduce concepts of benchmarking. Similarly models which can offer
adaptive and dynamic frameworks have been developed. However, none of these is at the stage
of commercial application at present and therefore perhaps not suitable for further discussion in
this Report.
12-1
10709367
10709367
13
CONCLUSIONS
Cyclic operation is merely the more frequent application of existing procedures. This, however,
multiplies the plant and commercial risks and costs so all aspects – operations, maintenance,
chemistry and engineering - must be tackled systematically to produce a commercially optimal
result. Very flexible operation of large coal-fired units has been successfully carried out while
maintaining high plant availability and without excessive additional costs.
13-1
10709367
10709367
14
GENERAL REFERENCES
1. M P Shipley and R J Browne “Cost- Effective Maintenance for the New Millennium”
Power Station Maintenance 2000, I.Mech E 2000, London, UK.
2. W H Stroman “Cyclic Corrosion Concerns for HRSG Users” Int. Seminar on Cyclic
Operation of Power Plant - Technical, Operation and Cost Issues, London June 2001, ETD
Publication.
3. M G Dyson “Start Up Issues with Boiler Water Treatment and Their Solution” ibid.
4. EPRI TR-107754 (August 1998) Cycling, Startup, Shutdown, and Layup Fossil Plant Cycle
Chemistry Guidelines for Operators and Chemists.
5. ETD Report No. 1002-iip-1001, Damage to Power Plants Due to Cycling, Issued December
2000.
7. EPRI GS-7219, Project 1184, September 1993, Cycling of Fossil-Fueled Power Plants
Volume 6: Evaluation and Strategy.
8. EPRI GS-7219 Volume 2 Project 1184-20, May 1991, Cycling Operation of Fossil Plant,
Volume 2: Converting PG&E’s Moss Landing 6 & 7 to Cycling Duty.
14-1
10709367
10709367
A
EXACERBATION OF THERMAL FATIGUE, CORROSION
FATIGUE AND STRESS CORROSION IN WATER AND
STEAM SYSTEMS UNDER CYCLING CONDITIONS
For heavy section headers a significant temperature gradient can develop across the wall, during
start up, basically because the flow of steam heats up the inside surface of the headers. Once the
plant is running this temperature gradient will disappear, and with it the stresses What is
potentially much more serious is the fact that in superheater pendants, relatively cold condensate
can build up in some of the bottom loops. At the same time the rest of the loops will be free
steaming so that the header will have reached operating temperature and be quite hot. At some
point the condensate will begin to boil and it will then be flushed through the system, partly as a
result of the boiling process and partly as a result of the pressure differential across the
superheater. Whatever the mechanism of flushing through, this relatively cold fluid will quench
the inside surfaces of the header leading to a severe thermal cycle. Thermal fatigue cracking
tends to appear where there are stress concentrations. Accordingly, in headers cracking tend to
occur in the ligament regions between the nozzels.
The other main cause of thermal fatigue will occurs in the heat transfer section. Here in
equipment such as superheaters, a permanent through wall temperature gradient results, as a
result of the heat transfer, so that stresses develop. Under base load conditions, these stresses
gradually disappear as a result of a phenomenon called “stress relaxation”. Unfortunately on cool
down these stresses reappear but with a reversed sign, but again some stress relaxation will
occur. On heat up thermal stresses again appear. The effect on such equipment is to add long
cycle fatigue stresses to those due to the pressure alone so that there will be a reduction in the
creep life of the equipment.
A-1
10709367
Exacerbation of Thermal Fatigue, Corrosion Fatigue and Stress Corrosion in Water and Steam Systems Under
Cycling Conditions
There is little doubt that the risk of corrosion fatigue does increase under two shift conditions,
particularly when a plant is off line at the weekend. The commonly held view is that because of
the long shut down, deaerators are short of steam hence air is drawn into the system. This
increases the corrosivity of the water, providing the “corrosion” half of the corrosion fatigue
mechanism. It is also likely that water treatment units can be off spec during startup. The
“fatigue” is due to, once again, thermal fatigue from differential expansion between parts which
heat up faster or slower than the surrounding material.
Regions of this type, where differential temperatures can arise, are invariable associated with
stress concentrations. Hence the opposing view is that corrosion fatigue is not necessarily
dependent on high oxygen levels, and that it is due primarily to the cracking of the protective
magnetite film. There are now a number of techniques and processes which enable operators to
run with very low oxygen levels and it will be useful to keep track of whether corrosion fatigue
is less prevalent in such systems.
Stress corrosion of blading is always a possibility on the back end of steam turbines in the region
of the Wilson zone where the steam droplets first begin to appear. Here the effects of cycling can
be positive and negative. Because of the variable pressure under cycling conditions, it is likely
that the point at which condensation occurs will move around much more than under base load
conditions. Furthermore, during the start and shut down periods the steam will be quite wet at
times. Hence, at least some of the time the susceptible blade root fixing will be “washed” by
relatively pure droplets. Against this is the fact that the blading will be passing through critical
resonance periods to a much greater extent. Furthermore because of greater problems with water
treatment and the increased tendency for droplet carryover to occur steam purity is likely to be
lower under cycling conditions. For this latter reason a check should be made on the quality of
the steam before it enters the condenser.
In principle the risk of stress corrosion/corrosion fatigue should increase under cycling
conditions. This is partly due to steam quality problems, as indicated above. It will also be due to
low cycle fatigue induce through a combination of running the turbine up and down, and also
due to thermal stresses induced in the turbine rotor during start and shut down.
A-2
10709367
B
ADVANCES IN REPAIR WELDING TECHNIQUES
Repair welding of defects, without subsequent stress relief was originally developed by CEGB
Marchwood (UK) for Cr-Mo-V thick section pipe work and is now being extended to other
materials and structures by the TWI and associated organizations [1].
The technique should only be used where either the defect or crack is of sufficient magnitude to
preclude it being “lived” with or ground out. If a crack is present or needed to be cut away,
(i.e. ground or burnt out to prevent the crack growing or to enable the weld repair to proceed),
this procedure should be started at the ends of the crack. This will prevent the crack tending to
“run on”.
Despite avoidance of post weld heat treatment, which implies taking structural components up to
high temperature, a procedures need to be thought through. It is advisable to carry out trial weld
operations on test pieces that will oblige the welder to simulate the actual conditions. During the
work, the job should be supervised to ensure that procedures are adhered to.
Preheat will still be necessary, so that in confined conditions, it is necessary to take due
precautions to ensure that the welder can carry out his job properly. Heating elements, for
preheating, should be placed on the opposite face to that being welded, and temperatures
monitored with thermocouples.
The weld faces need to be buttered, using a two-layer technique, and close control of weld heat
in put levels is required. This avoids dilution and ensures that the requisite grain structure is
obtained. A low heat input stringer technique is recommended for the first layer, and this may
require some training.
Mitsui Babcock in the UK, in conjunction with other organizations, have developed weld repair
procedures, where there have been concerns about the risk of over-pressurizing a system or
causing brittle fracture to the component. The problems with this type of work increase with
older structures where there is a risk of hydrogen cracking, laminations in material and a risk of
poor toughness in HAZ regions, particularly if there is no PWHT [3].
B-1
10709367
Advances in Repair Welding Techniques
Toughness is in fact a major limitation and it may be necessary to carry out fracture toughness
tests on trial specimens containing HAZ. The properties of the sub-critical HAZ are, of course,
the ruling factor.
The work was initially done on a cracked boiler. Backing and wing plates were used to retain the
weldment. Finite element analysis was used to estimate the residual stress field, but a short
period was allowed between each weld sequence, to allow some stress relaxation to occur. In
contrast to the procedures described above and developed by ex-CEGB Marchwood (UK), it was
agreed to use standard welding practices, since this avoided welder induced defects, due to lack
of familiarity.
Hopkinsons Ltd have developed a machine for re-welding, grinding and lapping of valve seats in
main steam line isolation valves. Because these valves are of the gate type, it implies renewing
an annular valve seat that is concentric with the pipe diameter. All welding, cutting and grinding
actions have to be done in a plane that is at right angles to center line of the pipe [4].
The need for this comes from plants operating on two shift duty, which leads more wear of the
sealing surfaces, due to opening and closing, and more cracking of the seat due to thermal
fatigue. There is also the prospect in some designs of creep of the valve body which result in the
valve seat breaking away.
One issue, which is highly pertinent, is that although it is possible to relap valve seats several
times, eventually the lapping reaches a zone of the valve seat in which dilution of the seat
material occurs. This will then result in very high rates of wear and scoring during the next
period of service. During repair, Stelliting is practical.
B-4 References
2. S J. Brett, D J. Abson and RL. Jones “The Repair Welding of Power Plant Without Post
Weld Heat Treatment” International Conference on the Integrity of High Temperature Welds,
Nottingham University, UK, Nov 1998.
3. J B. Wintle, R I. Jones, P. Mills, and A M. Pennick “Repairs to Safety Critical Welds without
Pressure Testing”, Seminar 642 on Repair of Pressure Equipment, I.Mech.E. London,
December 1999
B-2
10709367
C
ALTERNATORS
There have been a number of changes to generator design and materials of construction, among
which has been the switch to end rings of the Fe-18Mn-18Cr type which has reduced stress
corrosion as an issue. ENEL, however, did detect stress corrosion of Fe-18Mn-18Cr (for
consistency) end rings, which was associated with copper deposits. It is therefore advisable to
keep the alternator enclosure dry at all times.
Despite this there has been a move towards utilizing air cooled designs, particularly for CCGT
plant where there is a strong drive to reduce manpower and ancillary equipment. However,
similar pressures have led to the design of air cooled alternators being pushed to the limit and
problems are beginning to emerge. Design stresses and temperatures are higher, which will
impact on maintenance and reliability. Some of the issues were covered in a recent paper and are
summarized below [1]. These problems can appear within a short time of the machine being run,
and it is recommended that the first inspection be carried out after 12 months operation.
Every effort should be made to identify whether such problems are occurring, by electrical
testing, otherwise repair may become impossible.
References
1. S. Tucker, J. Milsom, G. Griffith and G. Stone “Problems with Modern Air- Cooled
Generator Windings “ Power Station Maintenance 2000 I.Mech E 2000.
C-1
10709367
10709367
D
THE QUESTIONNAIRE
Survey Form
Survey Objective: This survey is aimed at describing the impact of changing operation modes
on equipment and describe the change in maintenance activities when a plant transitions from
base-loaded to load-cycling operation.
Usual Fuel.
D-1
10709367
The Questionnaire
a) Corrective
b) Preventative
Fuel handling
Sootblowing
Condensate
Turbine Rotor
Turbine Valves
Generator Rotor
Boiler steamside
Boiler waterside
Boiler air/gas
Feedwater
D-2
10709367
The Questionnaire
HT piping
Motors
Pumps
Contact Name
D-3
10709367
10709367
E
CYCLING EFFECTS ON MAINTENANCE ACTIVITIES IN
THE UK – SUMMARY OF TYPICAL CYCLING
PROBLEMS AND THEIR POTENTIAL IMPACT
The following categories have been given as an indication of each activity to suggest the
likelihood of the problem occurring, its possible impact on availability and the type of
expenditure involved. It should be noted that locations are likely to operate with differing
commercial constraints, which could change the way in which any of the problems are managed.
These categories are intended for preliminary guidance only. The list is not necessarily
complete. Not all problems will be experienced and some utilities will experience other
problems. The engineering strategy/solution is given as the basis for further development to suit
the specific objectives of a utility.
Column: Probability of problem occurring (this gives some idea of how likely it is that the
problem will occur).
E-1
10709367
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential
Impact
Column: Type of Action required (gives an indication of the nature of the solution to the
problem).
E-2
10709367
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
LP Feed System
Condenser No significant problems attributed to low
extraction load or thermal cycling.
No significant problems attributed to low
load or thermal cycling. Monitor tube erosion and fatigue. I
LP Feed Heaters Potential for increased corrosion and Selective visual/MPI inspection at outages (4) OH
10709367
fatigue of heater tubes due to water quality M R Address water quality issues. OP
problems.
No significant problems attributed to low
DC Heaters
load or thermal cycling.
Potential for cracking of shell welds and Selective visual/MPI inspection of main seam
attachment welds due to thermal and L S welds and attachment welds. (4) OH
pressure cycling. Monitor and modify water treatment to reduce
Deaerator Thermal fatigue cracking of internal M 0 threat of corrosion. OP
attachments and furniture Consider use of anti corrosion coatings (can
Corrosion fatigue due to poor water L 0 be expensive and make inspection more MOD
chemistry. difficult)
HP Feed System
Thermal cycling of pump bodies on start- M R Monitor pump alignments/vibration and seals
up. R
to determine extent of problem.
Alignment problems due to distortion of Review operating procedures to reduce
L R OP
pump casings and impellers during rapid thermal shock.
temperature changes on start-up.
Consider modifications to pump bearings and REN
Interference & clearance problems on seals for flexible operation.
Feed Pumps pump seals due to rapid heating of L R
Consider pump pre-warming and bearing MOD
impellers relative to casings.
housing heating and cooling.
Inefficient operation of pumps during low OP
Review suitability of steam/electric pumps for
load operation below about 80% MCR, M R flexible operation.
possible overheating and leak off problems R
at low load. Monitor pump performance and temperatures.
Reduced reliability of electric pumps for Consider variable speed operation and leak
L R off configuration. MOD
frequent start up and shut down.
E-3
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
10709367
Selective visual/MPI inspection of all small
openings 0 openings (4).
H OH
OP
Optimize feed water chemistry control
Corrosion fatigue due to variations in water OH
L 0 Carry out sample visual and MPI inspections
Feed Pipework chemistry. on bends at extrados/intrados and external
Fatigue from pressure cycling, especially NDE on neutral axis of bends at outages (4).
at bends M S
Consider modification of pipework to increase
flexibility of system.
MOD
Thermal fatigue cracking due to injection of N S OH
Selective internal visual & MPI inspection of
Feed Regulating cold feed water during boiler start-up or
valve internals at outages (4)
Valves boiler top up.
Replace valves with more tolerant design eg
Feed Check valves Increased wear and tear due to increased
H 0 use of high alloy steels. REN
usage.
Economiser
Ligament cracking due to injection of cold H E Internal camera inspection for ligament OH
feed water during boiler start-up cracking and internal corrosion/erosion (4).
Ligament cracking due to boiler top up with External NDT of stubs
cold feed water during shut-down. H E
Headers Selective visual, MPI & ultrasonic inspection OH
Stub to header cracking due to at outages (4)
temperature differentials between H B OH
economiser tubes during low flow and Replace headers or damaged sections with
boiler shut down. increased ligament efficiency
L 0 Modify stub design to reduce local stresses. REN
Internal erosion & corrosion of outlet stubs
M dif b il f d t t i kl f d t t
E-4
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
Fatigue cracking due to water hammer L S Modify boiler feed to trickle feed system to MOD
during start-up on steaming economisers. reduce thermal shock.
Quench cracking of inlet tees due to Install economiser recirculation to maintain
MOD
injection of cold feed water on start-up uniform temperatures during boiler shut down.
H 0
Thermally induced bending due to
stratification of water flow during low load MOD
operation or off load boiler top up (top to M 0
bottom temperature differential)
Opportunist inspection and repairs
OH
Tubing Thermo-mechanical cycling of tube Selective visual/MPI inspection at outages (4)
H 0
10709367
attachments Consider replacement of attachments with
MOD
more flexible designs.
Evaporative
Section
Selective visual/MPI inspection of stubs at OH
outages (4) Repair or replace as required.
Stub to header cracking due to
temperature differentials between water M S Consider design change to increase flexibility MOD
wall tubes during low flow and boiler shut between tubes.
Water Wall headers down. Visual inspection of header for evidence of
OH
Thermally induced bending due to bending.
stratification of water flow during low load Review observations and consider design
operation or off load boiler top up (top to M 0 changes to more flexible arrangement. I
bottom temperature differential)
Consider off load recirculation to even out or
reduce temperature distribution. MOD
Thermo-mechanical cycling of tube N B
attachments Selective visual/MPI inspection at outages (4) OH
Corrosion fatigue arising from water quality Modify tube to structure attachment to more
(especially O2 content) L 0 MOD
flexible design
Water Wall tubes External corrosion during extended Improve water quality control OP
shutdown periods M 0
Consider off load recirculation to even out or
reduce temperature distribution. MOD
Fireside oxide spalling due to temperature M 0
cycling Use of ‘rifled’ boiler tube to reduce DNB MOD
Localised overheating of tubes especially
E-5
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
10709367
M R Review design of attachments, expansion
boiler casing joints etc and modify to improve flexibility. MOD
Boiler structure
Failure of windbox and ductwork Regular inspection of supports at outages for
attachments L 0
evidence of broken or slack supports (4). OH
Failure of ductwork expansion joints Consider once off check weigh of boiler
Load migration in boiler sling rod supports L R supports to confirm support integrity. Check
MOD
and potential for failure L 0 weighing of supports hot and cold prior to
adjustments
Thermal cycling of pump bodies on start-
H R
up.
Alignment problems due to distortion of Selective visual/MPI inspection of pump OH
Circulating Pumps pump casings and impellers during rapid L R bodies at outages (4).
temperature changes on start-up. Consider off load recirculation to even out or
reduce temperature distribution. MOD
Interference & clearance problems on
pump seals due to rapid heating of
L R
impellers relative to casings
Drum
Selective visual/MPI inspection of stubs, OH
Thermal cycling of drum shell and localised attachments and body welds. (4)
Drum Shell
cracking at stress concentrations and L E Assess defects for profiling or local repair. OH
welds. Monitor large or repeat defects and carry out
engineering assessment of drum integrity. OH
Thermal fatigue cracking of attachments Selective visual/MPI inspection of drum
Internals L 0 OH
and seal welds furniture (4)
E-6
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
Primary
Superheater
Quench cracking of bottom headers OH
N E Internal camera inspection for internal defects
(especially horizontal sections) due to
condensation in higher elements during if quench cracking is a possibility.
Headers
start-up. Review drainage effectiveness and modify
(additional drains) if necessary. MOD
Water logging of sagging bottom headers
and possible circulation restriction during Consider replacing headers
H R
start up. REN
10709367
Monitor drainage of elements and inspect for OH
internal corrosion.
Install tube drying facilities
Potential for corrosion during off load MOD
Tubing H B Consider re-tubing with improved drainage if
conditions especially on horizontal sections REN
possible.
Improve method for off load storage e.g.
nitrogen capping MOD
Desuperheaters
Attemperator Sprays Erratic control of spray valves due to R Modify control system with anticipatory control
H MOD
temperature variations during start up logic
Secondary
Superheater
Identify headers at risk. I
Internal CCTV inspection and external NDE of OH
all headers followed by periodic inspections
Cracking of body welds and ligament N E (4)
cracking of thick walled headers due to
Headers thermal fatigue. Fit thermocouples and monitor temperatures I
and ramp rates.
Cracking of header stubs
H S Review operating procedures to control
Thermal fatigue of dissimilar welds temperature ramp rates. OP
(ferritic/austenitic) M B
Assess thermal stresses eg finite element
analysis of high risk headers. I
Repair headers as required. Options for local
i l t f ti fh d
E-7
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
10709367
headers.
Tubing Potential for steam side exfoliation of oxide Monitor steam side oxide layers for integrity of I
layers due to thermal shock. M 0 oxide layers
High stress levels in tube to tube Monitor and review tube attachments and
attachments and tube to structure due to I
N 0 alignment and consider design options.
high temperature differences
Consider redesign of tubes to introduce more
Thermal fatigue failures of fireside tube to flexibility eg expansion loops. MOD
tube attachments resulting in tube mis- OP
alignment N 0 Review operating procedures to reduce cyclic
effects
Thermo-mechanical fatigue of attachments N 0
and supports due to temperature
Review design to ensure adequate flexibility I
Structure differentials during start up and shut down.
M 0 to accommodate relative expansion.
Potential for load migration and
subsequent failure of boiler supports. Redesign seals with tube sleeves. MOD
Cracking at welded roof and wall seals N R
Reheater
Little evidence of thermal fatigue in thinner
walled headers. Occasional visual inspection with NDE as
Headers H B OH
appropriate to confirm integrity of headers
Cracking at drain openings
Local overheating due to high heat flux H B Selective visual/MPI inspection and thickness OH
during low flow at start up. checks of tubes. Check for evidence of
Tubing
Potential for steam side exfoliation of oxide overheating and monitor (4)
layers due to thermal shock. M 0
Monitor steam side oxide layers for integrity of
Thermal fatigue failures of fireside tube to oxide layers I
E-8
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
tube attachments resulting in tube mis- M 0 Monitor and review tube attachments and
alignment alignment and consider design options. OH
Thermo-mechanical fatigue of attachments M 0 Review design to ensure adequate flexibility
Structure and supports due to temperature
to accommodate relative expansion. MOD
differentials during start up and shut down.
Cracking at welded roof and wall seals H 0
Boiler Valves
Thermal fatigue cracking of thick walled N E
Selective visual/MPI inspection of valve OH
valve bodies especially at changes in
bodies and drainage ports at outages (4).
10709367
section.
Carry out local repairs or replacement with
Stop valves Cracking of valve seat and disk hard OH
N R rolling spares
facing
Consider replacement of valves with modern
Quench cracking at valve by-pass and MOD
design, possibly with P91 bodies
drain connections N B
Potential for excessive use due to poor R Selective visual/MPI inspection of valve OH
Safety valves boiler temperature and pressure control bodies and drainage ports at outages (4).
L R
during rapid start up, shut down and load Review boiler operability, control and
changes. instrumentation. OP
E-9
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
Main Steam
Pipework
Monitor steam pipework and drain I
Potential for local thermal quenching of temperatures and assess implications.
main steam pipework during boiler shut E
L Maintain pipework (and boiler) above
down. OP
saturation temperature as long as possible.
Thermal quenching at drains due to
Pipework
suction from drains system. S
H Reduce pressure in steam pipework to reduce
Thermal stressing due to temperature OP
mismatch during rapid admission of steam saturation temperature (sliding pressure
0 operation)
10709367
on start-up. M
Optimize start-up procedures to match steam
to metal temperatures OP
Monitor welds especially terminal welds, with
visual inspections and NDE as appropriate at OH
outages (4).
Welds Thermal fatigue and creep fatigue of welds L E
Consider need for metallographic replication
where creep and type IV cracking identified as OH
a threat.
Tendency for pipework to migrate to limit of
H 0 Examine pipe supports and pipe movements
travel on constant load supports OH
and compare with design intent to confirm
Increased cyclic stresses from restrained correct functioning of pipework supports.
Supports or defective pipe support system leading M 0
Consider need for detailed pipe stressing I
to over-stressing at welds.
assessment under cycling conditions.
Over stressing and cyclic stressing of
Consider need for additional pipe movement MOD
terminal welds at turbine, steam chest,
H 0 restraints
manifolds etc.
Turbine Steam
Chest
Increased wear and tear due to frequent Planned maintenance of valve operating gear
M R and valve seats at outages (4). Repair and OH
usage.
Cracking of hard surfacing on valve head replace as required.
M 0 Selective visual/MPI inspection of valve
Stop/Throttle valves and valve seats.
bodies at outages (4). OH
Wear of valve spindles and seals. Machine out shallow cracks and re-profile to
M 0
Thermal fatigue & quench type cracking at reduce stress concentrations. Monitor crack
valve body drains and by-pass portals. M 0 OH
growth.
E-10
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
10709367
HP & IP Turbine
Modify nozzle housing to accommodate MOD
differential expansion, eg include expansion
Low cycle fatigue due to rapid local slots.
Inlet nozzles L E
temperature rise on start up.
Install stop valve by-pass to facilitate pre-
warming MOD
Threat of low cycle fatigue not a major
problem
L R
Potential for alignment and axial expansion
problems during rapid temperature Monitor situation and modify operating OP
changes. M R procedures to minimize problems.
Rotor
Increased erosion of HP blades under wet Consider variable pressure (sliding pressure)
steam conditions on start up and sliding control to reduce temperature variations. MOD
pressure operation.
M R
Solid particle erosion from spalling oxide
layers
Install thermocouples and monitor I
temperature gradients and ramp rates. I
Assess design by finite element analysis and
Thermal fatigue and creep fatigue of thick
operational procedures to optimize stresses.
sections, section changes eg flanges and
Casing M E Inspect and optimize thermal insulation. R
at stress concentration features eg
Consider flange warming or heating elements MOD
grooves.
to balance steam to metal temperatures.
Consider options for replacement with modern
optimized design. MOD
E-11
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
Increased air in-leakage during turbine Consider use of off load gland steam system.
MOD
start up and shut down. H R Monitor gland performance
Seals R
Potential for seal damage due to Optimize procedures to minimize risk of gland
OP
differential expansion during start-up and H R rub occurring.
shut down. OH
Routine inspection and repairs at outages (4).
10709367
LP Turbine
Stress corrosion cracking of rotor discs M
E Review water/steam chemistry OP
and key ways
Replace discs with design less susceptible to MOD
Increased threat of transverse fatigue M SCC
cracking of shaft 0
Rotor Carry out opportunity inspection of shafts and
Fatigue cracking at keyways OH
M blading at outages.
Fatigue and corrosion failure of final row E
Machine out minor cracking.
blades OH
M Consider blade replacement (titanium blade
Blade erosion at low loads when exhaust R REN
option)
spraying used
Casing No significant problems attributed to low
load or thermal cycling.
Condenser
E-12
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
Water Treatment
Plant
10709367
Review water quality under variable load OP
conditions.
Variable conditions lead to instability of
Chemical dosing H R
chemical monitoring and dosing.
Consider chemical control equipment and
consider upgrading and/or replacement. MOD
Generators
E-13
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
Switchgear
10709367
increased wear and tear.
Increase inspection (condition monitoring) and
Switchgear R
maintenance of switchgear
Moisture and dust ingress during shut
down periods M 0
Electric Motors
H R
Increased wear and tear due to stop/start
R
operation Increase inspection (condition monitoring) and
maintenance of motors.
Increased duty on windings during motor
Motors M R
start operations Install variable speed controller gear
MOD
Moisture and dust ingress during shut Consider upgrading motors.
REN
down periods
M R
Control Systems
E-14
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
Unstable operation of plant H R Install new ‘algorithm’ based forward feed MOD
digital control systems
Erratic final steam temperature control H R
Install stress analyzer / life usage indication
Excessive thermal stressing of major H 0 for major components R
components
10709367
Coal Plant
E-15
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
Milling Plant
10709367
Mills
Review mill maintenance frequency to
Potential for pulverised fuel dust build up OP
M 0 minimise dust build up.
and fires/explosions.
MOD
Consider pulverized coal storage system to
minimize frequency of use of mills.
E-16
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
Combustion
10709367
burners to guarantee coal burner light-up Enhance routine maintenance.
Oil Burners H R R
and flame stabilisation, especially at low
load. Consider renewal of oil burners to more
REN
modern type.
Plant specific. Generally not a significant Monitor NOx emissions over operating range. I
problem. review observations on plant specific basis.
Low NOx
Some instances of severely reduced Seek specialist advice from plant and
performance equipment suppliers. OP
E-17
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
Fans
Low cycle fatigue of fan runners leading to Monitor fan runners and repair as required. R
cracking of blade to division/side plate L R
Consider design/material changes to reduce
ID & FD Fans welds. weld cracking. MOD
Distortion of shafts due to uneven cooling L R Review operating procedures to ensure
on shut down. uniform cooling e.g. baring gear. OP
Gas Pass
10709367
Carry out frequent cleaning and maintenance
of joints and seals. R
Additional wear and tear due to increased
Expansion joints and
frequency of cycling and ‘jacking’ effect of H R Consider redesign of worst affected joints to
seals
dust/ash entrapment within joints reduce dust entrapment, change materials or REN
seal type e.g. mechanical/flexible.
Monitor corrosion. OH
Increase in dew point corrosion of Patch repair or replace sections as required. R
Ductwork M 0
ductwork on cooling or at low loads.
Consider use of corrosion resistant materials
or coatings. MOD
E-18
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
Dust Extraction
10709367
Improve thermal insulation of structure to R
retain heat.
E-19
Cycling Effects on Maintenance Activities in the UK – Summary of Typical Cycling Problems and their Potential Impact
10709367
E-20
10709367
10709367
Target: About EPRI
Maintenance Task Selection Guidelines and EPRI creates science and technology solutions for
Technologies
the global energy and energy services industry. U.S.
electric utilities established the Electric Power
Research Institute in 1973 as a nonprofit research
consortium for the benefit of utility members, their
customers, and society. Now known simply as EPRI,
the company provides a wide range of innovative
products and services to more than 1000 energy-
related organizations in 40 countries. EPRI’s
multidisciplinary team of scientists and engineers
draws on a worldwide network of technical and
business expertise to help solve today’s toughest
energy and environmental problems.
EPRI. Electrify the World
1004017
EPRI • 3412 Hillview Avenue, Palo Alto, California 94304 • PO Box 10412, Palo Alto, California 94303 • USA
800.313.3774 • 650.855.2121 • [email protected] • www.epri.com
10709367