5-1 5 Mechanical Design Considerations for High- and Low-Pressure Casings Introduction The turbine casing is essentially a cylindrical vessel and the main stationary portion of each expansion section. This casing encloses the rotating elements of the unit and at the same time locates the stationary blades, either directly, or through the location and support of an inner casing, which itself carries the stationary blades and/or diaphragms. The principle components of the casing are the shells, which provide the mechanical strength of the element and carry and locate other elements such as packing heads, diaphragms, and the inner casing or blade carriers. The casing is normally split along its horizontal joint at the centerline to facilitate assembly and provide access to the rotor and internal stationary portions of the unit. The shell halves are normally connected through a bolted flange at their horizontal joint and act to contain the working fluid while maintaining it in intimate contact with the steam path blade elements. Casings may also provide locations for internal packings or portions of the steam seal system and could, if moisture is present in the steam, be equipped with internal moisture collection and drainage systems. The high-pressure shells should also, in the case of minor failures, be capable of containing missiles that are generated from the rotor. Both the upper and lower portions of the casings can be arranged to provide connections for welded pipe stubs. To these stubs are connected external pipes that allow steam to be extracted for regenerative feed heating or other cycle or process uses. Such steam is extracted from the main steam flow. The casing may also be penetrated by other pipes that are used to introduce or extract steam for other parts of the cycle. It is normal for pipe connections to the upper half to be connected through flanges or other device that allows their quick disassemble at outages and then reconnection without the use of any form of heating or metal fusion techniques.
102
Embed
and Low-Pressure Casings Introduction The turbine casing is - Yimg
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
5-1
5
Mechanical Design Considerations for High- and Low-Pressure Casings
Introduction
The turbine casing is essentially a cylindrical vessel and the main stationary portion of
each expansion section. This casing encloses the rotating elements of the unit and at the same
time locates the stationary blades, either directly, or through the location and support of an inner
casing, which itself carries the stationary blades and/or diaphragms. The principle components of
the casing are the shells, which provide the mechanical strength of the element and carry and
locate other elements such as packing heads, diaphragms, and the inner casing or blade carriers.
The casing is normally split along its horizontal joint at the centerline to facilitate
assembly and provide access to the rotor and internal stationary portions of the unit. The shell
halves are normally connected through a bolted flange at their horizontal joint and act to contain
the working fluid while maintaining it in intimate contact with the steam path blade elements.
Casings may also provide locations for internal packings or portions of the steam seal
system and could, if moisture is present in the steam, be equipped with internal moisture
collection and drainage systems. The high-pressure shells should also, in the case of minor
failures, be capable of containing missiles that are generated from the rotor.
Both the upper and lower portions of the casings can be arranged to provide connections
for welded pipe stubs. To these stubs are connected external pipes that allow steam to be
extracted for regenerative feed heating or other cycle or process uses. Such steam is extracted
from the main steam flow. The casing may also be penetrated by other pipes that are used to
introduce or extract steam for other parts of the cycle. It is normal for pipe connections to the
upper half to be connected through flanges or other device that allows their quick disassemble at
outages and then reconnection without the use of any form of heating or metal fusion techniques.
5-2
Special provisions in the casing are necessary to admit the high-pressure, high-
temperature steam and make provision for the differential expansion that occurs between the
various portions of the shells. Such differential expansion occurs because of the different
temperatures or temperature gradient along the axial length of the casing and also because of the
different rates at which the various parts of the unit heat and cool with main steam temperature
changes. For double-shell construction, it is necessary for the main inlet pipes to pass through the
outer casing and introduce steam to the main steam inlet belt or nozzle box.
The high-pressure casings are normally supported at each end through arms that are
produced integral with and extend from the casing to pedestals that are located adjacent to, and
between, the casings or sections. Transverse and/or axial keys are used to maintain alignment of
the shells at these pedestals. Such keys have normally been hardened by nitriding and are located
on the bottom vertical centerline to ensure correct alignment is maintained at all loads and during
transient operating conditions.
Low-pressure casings are designed to contain the steam and to minimize the in leakage of
air when the exhaust pressure is sub-atmospheric. Because at exhaust from the turbine the
volumetric flow is large, it is normal for these low-pressures elements to be produced by
fabrication, and because such fabrications are not structurally strong, it becomes necessary to
support them for their entire perimeter at their horizontal joint or a similar location below this
joint.
Components Comprising the Turbine Casing
The turbine casings have a number of individual elements, which when assembled allow
the unit to operate safely and to achieve high levels of reliability and efficiency. A list of the
principle components follows.
5-3
The shells. The shells are the main structural components that are produced by casting,
fabrication, or in some designs by a combination of both depending upon the experience and
preference of the designer.
The shaft-end packing head. The packing head is attached to the shells, and carries the
gland rings that are located where the rotor passes through the shells. These heads are designed
to carry gland rings that minimize the outward leakage of the steam or the inward leakage of air.
The inlet section. The inlet to the steam path must be designed to allow free access of the
inlet pipes, transport the steam to the nozzle box, and minimize leakage of steam at those
locations. The inlet is designed to permit movement between the inlet pipes and the main body of
the shells.
The explosion diaphragm on low-pressure sections. In the low-pressure shells there is
a need to provide for the rapid removal of steam from the internals of the casing in the event
there is a sudden and high rate of pressure increase due to some transient condition.
A diffuser section at exhaust from the last stage. In an effort to maximize the energy
extracted from the working fluid, the final rotating blade is arranged to exhaust into a diffuser
section normally produced as part of the casing fabrication.
Functions of the Shells or Casings
The casings are the main containment vessel of the turbine, which defines their major
function of containing the working fluid. These elements, therefore, surround the steam path and
can be made to perform several other secondary functions. These functions are dependent upon
the steam conditions within the steam path, which conditions influence the arrangement,
materials, and support for these casings.
It is necessary to consider casings in two separate categories, arranged most suitably by
the temperature of the steam they contain. The high-pressure/temperature elements are those that
5-4
operate with steam temperatures above the range of about 700 degrees Fahrenheit (°F), and the
low-pressure/temperature elements are those that operate at temperatures below this value.
The high-pressure/high-temperature sections
The steam admitted to these high duty casings will have pressures up to 3500 pounds per
square inch absolute (psia) although units have been designed to 5000 psia. Temperatures are at
the 1000 to 1100°F value at the inlet and can be reheated to 1000 to 1050°F before readmission
to the intermediate pressure section. However, units have been designed to operate at higher
values of temperature.
Because of the initial steam conditions, the high-pressure/high-temperature casings are
subjected to high internal pressure, which produces significant tangential, longitudinal, and radial
stresses in the walls. These casings must therefore be designed so they are able to withstand
these conditions at normal operating conditions and during transients. It must also be recognized
that at the higher temperatures, the mechanical properties of the material from which the casing
is produced are lowered, which reduces the factors of safety of these major components.
The reheat casings are subject to lower steam pressures, but because of the increase in
specific volume of the steam at these lower pressures and the high reheat temperatures, these
casings can be subject to stresses of the same magnitude as exist in the high-pressure sections.
In addition to being a containment pressure vessel, the shells have certain secondary functions.
Fulfillment of these functions is important to the production of a successful design and is
necessary for the operation of the unit. The most important of these follow.
• The outer shells of high- and intermediate-pressure sections are part of the main
external structure. As such they must have sufficient strength they are able to transmit
the large differential expansion forces through the casing arms to slide the pedestals
on their sole plates. This they must do without any form of vertical or lateral
5-5
distortion that would affect the alignment of the steam path. The sole plates must be
able to be moved by the casing in such a manner they will not cause deflection,
excessive distortion, or misalignment. Alignment of the turbine generator must be
maintained at all times, under all loads and variations of steam conditions, and in all
directions.
• The casings must also be able to carry the loads developed on the stationary blade
rows (individual blades and diaphragms) and inner casing due the pressure deferential
within the steam path.
• The shells should have sufficient strength and weight that the casing is able to resist,
without change of alignment, external forces and moments imposed on it by station
piping. (Tavernelli and Coffin 1961) Figure 5–1 shows how various forces may be
imposed on the casing by expanding piping thrusts that tend to lift the casing from its
foundations and could be sufficient to cause misalignment.
Fig. 5–1 The Piping Thrusts Developed on a Casing
• The shells must maintain the stationary elements they carry in correct axial and radial
alignment relative to the rotor. That is, concentricity must be maintained together
with axial alignment.
• The outer shells must be sufficiently rigid that they are able to transmit and withstand
external forces due to excessive vibrations, including earthquakes and other high
intensity natural phenomena. These phenomena, although rare and highly unlikely in
most North American installations, could have catastrophic consequences if their
severity were sufficient to cause sudden and excessive misalignment within a casing
with the rotor at operating speed.
5-6
• The mass of the casings must be sufficient to make significant contributions toward
holding the unit firmly on its foundations and suppressing vibrations.
• In the event of a blade, wheel, or rotor failure, the inner and outer shells of the high-
and intermediate-pressure casings provide a strong containment vessel for the rotating
parts. These shells should be capable of absorbing high impact projectiles, thereby
minimizing the possibility of a projectile penetrating the casing and causing serious
injury to plant personnel.
• The outer shell provides a barrier by which heat is retained within the unit. This
barrier is reinforced by thermal lagging, which is attached to the outer surfaces of the
shell, and is the main barrier to radiant heat loss. The inner casing also provides a heat
barrier which reduces heat loss by minimizing temperatures on the inner surface of
the outer shell.
• Turbine shells are massive structures. They are thick sectioned, and due to their mass,
respond slowly to changes in steam temperature. This thermal inertia to the rate of
temperature change, gives rise to the need for special considerations of stationary to
rotating element clearances. The shell design must be adequate to accommodate this
thermal lethargy at all points of contact with potentially lower temperature elements,
such as valves, bearing housings, and the front standard.
• The thermal gradient developed in the casing walls will introduce thermal stresses
during operation. This is particularly so during temperature transients, when stresses
can be high.
The design of the shells should be such that the stresses induced provide a unit in which
the predicted life of the components is acceptable. To do this the material properties must be
carefully defined, and the design must eliminate, to the greatest extent possible, stress
5-7
concentration regions. This will help minimize the possibility of thermal or low-cycle fatigue
cracking.
The high-pressure, high-temperature casings are normally cast components. However,
fabricated elements have been used in some nuclear applications where the initial or nuclear
boiler delivery pressures are not more than about 1000 psia. Such nuclear casings do, however,
have free moisture in them that introduces another type of problem.
The low-pressure/low-temperature sections
The low-pressure or exhaust sections of a turbine unit are normally designed to accept
steam at an inlet pressure of about 200 psia and a maximum temperature of about 700°F. This
maximum temperature is set more by the material of the rotor than of the casings.
The normal design practice is to make the total expansion ahead of the low-pressure
portion of the unit occur in one or more sections. At the lower pressure end of the high-pressure
casing, pressures may be at the 400 to 600psia level and temperatures in the 700 to 600°F range.
At exhaust from the intermediate or reheat section, the pressure will normally be in the range 70
to 200 psia and the temperature at the 550 to 700°F level.
The normal arrangement of the low-pressure expansion sections of a high-output unit is
to have multiple double-flow sections, with an inner and an outer casing in which the axial thrust
is canceled. In these designs, the casings at their inlet are subject to a pressure differential across
their walls equal to the differential between the inlet pressure and atmosphere. There and also
many units in service with three low-pressure expansions—one accepting one-third of the steam
exhausting from the reheat or intermediate pressure section, and the other two-thirds going to a
double-flow low-pressure section.
The low-pressure casings have many of the same functions and characteristics as the
high-pressure, high-temperature components. However, due to their physical size and the fact
5-8
they are vessels required to maintain an internal pressure that is higher than atmospheric pressure
within the inner sections and a vacuum in their hood and between the inner an outer sections,
these requirements are modified. The basic functions of the low-pressure casing are:
• the outer shell must locate from the foundations and support the inner shell with
sufficient rigidity that it can maintain alignment of the steam path under all conditions
of transient load and steam conditions
• the inner casing must be able to carry and support the low-pressure diaphragms, to
maintain concentricity and axial alignment under all steam conditions and under both
steady state and transient loads.
• the casings must act as a transition and diverting structure to direct the steam
exhausting from the last stage blades to the condenser, minimizing the frictional loss
within the hood
• the low-pressure section casings must incorporate a seal system that limits the ingress
of air into the system and thereby help maintain vacuum integrity
• the casings must be sufficiently that robust they will not deflect by unacceptable
amounts due to vacuum pull during operation. Similarly the casing must be able to
resist vertical deflection due to heavy water loads in the condenser hot well.
• the casing, while mounted on the condenser with either rigid or flexible connections
and supported off the foundation, must have sufficient axial flexibility it is able to
accommodate temperature swings within the system and maintain alignment
• The casing must be designed with sufficient axial clearance to accommodate thermal
differential expansion at normal operating conditions and under short and long rotor
conditions
• The casings must be designed so steam extraction pockets can be used to remove
steam from the casings for regenerative feed heating
5-9
The high and reheat casings can normally be expected to contain any blades that detach
as missiles. In addition these cast steel casings should also contain the rotor, although the unit
would be wrecked. Such high condition casings therefore act as a containment vessel or safety
barrier in the case of a significant accident or material rupture. The low-pressure casing may
contain the blades, although last stage blade elements can cause significant damage if they
detach from the rotor. If a rotor or wheel bursts in the low-pressure casing, it is most unlikely the
casing will be able to contain the missiles that are generated.
High-Pressure/High-temperature Casings
There are a number of casing configurations that fall within the category of high-
pressure/high-temperature application. These include the following.
High-pressure sections for fossil application. These units are normally subject to a
maximum cycle condition of 3500 psia and 1000°F. Although pressures up to 5000 psia and
1200°F have been used on advanced cycles. These casings are always built to the configuration
of an inner and an outer shell so that a pressure, and more importantly, a temperature gradient
can be established across both the inner and outer components.
Intermediate-pressure sections for fossil application. There are still in operation
turbine units that do not utilize reheat at exhaust from the high-pressure section. Therefore, there
are turbine casings that are intended to operate on steam having conditions equal to those
exhausting from the high-pressure section. Such units were used when the cost of fossil fuel was
inexpensive at the time the plant was built and are often used where a plant is located near the
fuel source. The intermediate pressure sections are probably the least stressed type, which can be
called high-pressure/high-temperature, and which will be encountered in modern power plants.
5-10
Reheat pressure sections. It is normal in modern power plants to reheat the steam after it
has completed its initial expansion in the high-pressure section. Therefore, the steam entering the
intermediate pressure section casings has a pressure about 7 to 10% lower than the steam
exhausting from the high-pressure section, and the temperature of reheat is about 1000°F.
There are designs in which the high and reheat expansions are contained within a single
shell. In these designs the steam, after its initial (high-pressure) expansion, is returned to the
boiler reheat section, reheated, and returned to the same shell for a second expansion.
Second reheat sections. Some cycles are designed to utilize a second reheat section. In
this cycle the steam, after expanding in the first reheat section, is returned to the boiler where it is
given a second reheat and again returned to the turbine to continue its expansion in a second
lower pressure reheat section. Upon return from the boiler second reheat section, the steam has
had its temperature again raised to a value close to the initial temperature and is returned to this
second reheat section with a pressure reduced by 7 to 10% from that exhausting from the first
reheat section.
Those two turbine sections discussed previously are defined as the first and second reheat
sections and in certain applications are arranged for double flow.
High-pressure sections for nuclear application. With the advent of water-cooled
reactors producing low-quality steam, a high-pressure section was required that was capable of
handling large volumetric flows of steam that contained a small initial moisture content.
Therefore, as steam enters the turbine, it has an initial pressure of about 1000 psia and can have
an initial moisture content of 0.25%. In such casings, provision must be made to collect and
drain a considerable amount of water that will be deposited on the casings and other internal
parts of the unit as the pressure decays.
These turbine sections, in order to be able to accommodate the high volumetric flows
without exceeding axial velocity limitations for efficient expansion of the steam, have tended to
5-11
be used at 1800 revolutions per minute (rpm) for 60 Hertz (Hz) applications. While the majority
of 50 Hz application is as 1500 rpm, there are some 3000 rpm applications at lower ratings. The
need to go to half speed units has caused an increase in rotor and casing diameters to maintain an
acceptable velocity ratio _. This has tended to increase the stress levels in the casings because of
the larger diameter required of the casings.
The casings for fossil application discussed here normally contain a high speed
rotor—3000 or 3600 rpm—driving a two-pole generator. Because of its high speed of rotation
coupled with high operating temperatures, there are physical limitations to the diameter that can
be specified for the rotor. Currently, it is difficult to produce a rotor forging with suitable
material properties and capable of carrying the rotating blades much larger than 40 inches (in.).
Also the maximum length of blades must be limited because of the centrifugal loading.
Therefore, the maximum casing internal diameter would be limited to about 65 to 70 in.
In nuclear high-pressure sections and some fossil sections, particularly for cross
compound units, the sections can be arranged to drive half speed 1500 rpm or 1800 rpm four-
pole generators. Because of their lower speed, it is possible to increase the rotor diameter without
exceeding stress limitations in the rotor or blades. With this type of rotor, a limitation of
approximately 64 in., producing a total rotor diameter of about 95 in. exists.
These diametral limitations are for 60 Hz units. For 50 Hz units, the possible diameters
would be somewhat larger. However, the maximum diameter is often a function of rotor
manufacturing capability rather than stress levels. As manufacturing techniques improve, it is
possible larger diameter rotor forgings will be available and larger casings required.
Pressure Staging and Multiple Shells
The casings contain the high-pressure, high-temperature steam with a differential from
working condition to atmospheric. The duty on the individual casing shells is normally reduced
5-12
by the use of a double casing construction. This form of construction provides for a
temperature/pressure barrier to be established with conditions from high-pressure inlet to high-
pressure exhaust across an inner shell,and then from high-pressure exhaust to atmospheric over
an outer.
Shown as Figure 5–2 is a two-casing arrangement, in which the individual diaphragms, or
stationary blades, are located and carried in the inner shell. This inner shell is then supported
from flanges machined into the outer surface of the inner shell. These locate in special locating
grooves machined into the inner surface of the outer shell. It can also be seen from Figure 5–2
that there is a constant pressure and thermal gradient across the outer shell, and the inner shell is
subjected to a gradient dependent upon the differences between the stage conditions and the
high-pressure exhaust surrounding the inner shell.
Fig. 5–2 A Double Casing Unit With the Diaphragms Carried in Inner Casings or Blade
Carriers
Figure 5–3 is a casing design with the high- and reheat-pressure sections are contained
within a single casing. With this arrangement the high-pressure expansion has an inner casing to
carry the diaphragms, and the outer casing is subject to the same pressure gradients as seen in the
casing design of Figure 5–2. After reheating, the steam is returned to the reheat section, which is
a single casing design, with the diaphragms carried in grooves machined into the inner surface of
the shell. Therefore, this shell is subject to a decreasing gradient along its length from stage
conditions to atmospheric.
Fig. 5–3 A Combined High-Pressure and Reheat Section
5-13
For the lower-pressure/high-temperature shells, such as used for intermediate or reheat
sections, the major concern is with thermal gradients. In this type of section it is possible to make
the casing walls thinner and more flexible. A casing of this design is shown in Figure 5–4.
(Hummer and Drahy 1964) With this type of design, steam is introduced from the inlet pipe into
the inner casing. This unit has seals at a to prevent the excess leakage of steam while allowing
for the expansion and contraction of the inlet pipes. This allowance is provided to accommodate
pipe movement during start-up and shutdown, or whenever the inlet pipes will heat and cool
much faster than the surrounding casing.
Fig. 5–4 An Intermediate (Reheat) Section, With Inner Walls and Extraction Pockets for
Pressurizing the Outer Casing
The steam enters the nozzle box then expands through the steam path. At completion of
its expansion, steam at the high-pressure exhaust condition surrounds the accessible portion of
the inner casing, which is then subject to pressure and temperature gradients corresponding to the
difference between individual stage and section exhaust conditions. The outer shell is subject to
pressure and temperature differentials equal to the high-pressure exhaust conditions and local
ambient. Had this casing been of single-shell construction, the single outer casing would have
been subject to the total differential between stage and ambient conditions.
In this type of design, it can be seen that diaphragms are supported and carried in inner
casing rings with each supporting a number of stages. These are also termed blade carriers.
These diaphragm groups provide for access regions where steam can be removed from the unit
for regenerative feed heating. In these extraction belts or pockets, the steam exists at the stage
discharge conditions from the upstream carrier, making the outer casing inner surface conditions
5-14
equivalent to the extraction conditions. This reduces the temperature and pressure gradient across
the casing walls.
An alternate arrangement of the high-pressure casing is to eliminate the inner shell after
the high-pressure expansion is partially complete. When this is done the direction of the flow is
reversed, led to the other side of the nozzle box where the expansion is completed. Shown as
Figure 1–38 of chapter 1 is such a design with the flow reversed after eight stages to flow
through a final three before being returned to the boiler reheater section. With this design, the
steam path is split into two portions, an upper pressure portion and a lower pressure portion. The
arrangement of the shaft-end seals is the same except the pressure range across them will differ.
The only significant difference in such a design is that the outer casing will be subject to
a higher pressure and temperature differential over the first portion of the expansion. The second
or reversed portion has eliminated the inner portion of the casing, and the diaphragms are carried
by the single casing.
The reversal point—end of expansion portion—is selected based on three considerations.
These are:
1. the need to extract steam from the section for regenerative feed heating
2. the need to lower the temperature and pressure gradients across the individual casing
portions
3. the adjustment of the axial thrust developed in the two blade portions. These two
portions of thrusts are opposed, and will affect the thrust which needs to be carried by
the thrust block.
Reversal point selection affects shell pressure, temperature, and axial thrust. These effects
are best reviewed from the high-pressure expansion line of Figure 5–5 that shows that steam
enters the section at conditions Pin and Tin. The individual stage points of the high-pressure
5-15
section are established along this expansion line, and possible stage end points are shown as a0
… a3 providing for optimum velocity ratios _ of the stages.
Fig. 5–5 The Reversal Effect and the Selection of Pressure and Temperature at the
Reversal Point
The design evaluation will consider the impact of different reversal points and the effect
these will have on the turbine and cycle efficiency. Normally, the controlling consideration is
achieving an extraction point for regenerative feed heating as this extraction will normally be to
the top heater and will therefore set the final temperature of the feed water being returned to the
boiler. This temperature is fundamental in establishing the heat rate of the total installation.
There is some small degree of flexibility in selecting the pressure and temperature at the
reversal point. This flexibility is achieved by selection of stage diameters that will modify the
velocity ratio _ and the energy distribution across the individual stages above the reversal point.
From the expansion line of Figure 5–5, the possible reversal points are shown as a1, a2 or
a3, with a0 being the inlet to the first of the three alternates being considered. These alternate
stage points will influence both the turbine and cycle. From considerations of the expansion line
alternates, the effect on the unit can be seen. In Figure 5–5b, the three stages are in series, with
their individual thrusts Tn acting in the same direction. In Figure 5–5c, the last of these three
stages, the steam flow direction has been reversed. Therefore, the steam will reverse at exhaust
from the second stage, and enter the third in the opposite direction changing the thrust by an
amount 2xT3. Also the temperature at the reversal point will increase from To3 to To2.
This change will also increase the steam condition surrounding the inner casing and
modify the thermal gradients across both inner and outer casings.
5-16
The modern design many manufacturers utilize contains a nozzle box as shown in Figure
5–2 and in Figure 1–38 of chapter 1. These nozzle boxes are self-contained vessels, located
within but forming part of the inner casing. These boxes are normally produced by forging and
are designed to distribute the steam around a portion of the inlet annulus and discharge it through
the first stage nozzle plates as seen in chapter 6. In the case of the nozzle box, the highest steam
conditions sensed by the casing are those of the steam discharging from these first stage nozzles.
The total casing arrangement in a self-contained nozzle box is essentially that of a triple-
shell construction. Nozzle boxes are now used in practically all designs with an initial pressure
above 2000 psia and temperatures above 900°F. Depending upon the duty intended for the unit
and the system into which it will be electrically connected, the first stage nozzles may be
grouped in the following manner.
Segmental or nozzle control. If four or more physically separated inlet segments, as
shown in Figure 5–6a, are used the unit is termed nozzle controlled. With this design, admission
to each segment is controlled by a separate valve. The valves are each arranged to open or close
sequentially as unit output demand changes. The nozzle segments cover the complete 360° inlet
or whatever portion is required to access sufficient steam to the unit. With this design, there is a
small portion of inactive arc at the tangential transition from one nozzle segment to another.
Fig. 5–6 Alternate Methods for the Admission of Steam to the First Stage of a High-
Pressure Section
In this design, the valve opening sequence is V1, V2, V3, and finally V4. As each valve
opens sequentially, the active arc grows in tangential or chord length dependent upon the load
demand on the unit.
5-17
Two 180° arcs. A similar design employs two 180° segments, Figure 5–6b, with the joint
between the inlet arcs at the horizontal joint. In this arrangement, the inlet arcs may be fed by
one- or two-control valve arrangements. Again there are small inactive arcs, which in this design
are located at the horizontal joints.
As with the design shown in Figure 5–6a, steam is admitted to independent arcs, each
covering a nominal 180° of the tangential position. Up to 50% load steam is admitted to the top
half only with valves V1 and V2 open. Past 50% the other valves open to full load.
Full arc admission. One 360° segment or inlet arc is seen in Figure 5–6c. This
arrangement is similar to Figure 5–6b except there is a flow connection from the upper to the
lower chambers. This flow connection may be in the steam chest downstream of the control
valves but is more commonly made in a header adjacent to the control valves. Steam flow to this
common chamber is controlled by valves that admit steam to the entire inlet arc.
With this design the valves will open sequentially in response to load demands, but each
of the valves V1, V2, V3, and V4 feeds the complete 360° arc. This is termed throttle control.
There will normally be a small inactive arc at the horizontal joint. However, the effect of this on
the stimulus produced can be reduced by careful design of the joint partitions.
The Low-Pressure Casings
The term low-pressure/low temperature, in terms of turbine section arrangement is
applied to those expansion that accept incoming steam from a higher pressure section, and allow
it to expand to exhaust or condenser pressure. The casings that enclose this energy level steam
path tend, in modern units to be a separate, often double-flow section.
However, in many older and lower rating units without steam reheat, the lower steam
condition expansion occurred in a casing that was integral with the inlet or higher condition
5-18
casing at the inlet portion of the expansion. Often these casings are built using a cast high-
pressure section with a fabricated low-pressure section.
The older design single-casing units were produced by both casting and fabrication. In
many designs, the high-pressure section was produced from a steel casting and the lower
pressure portion was a fabricated structure bolted at a vertical joint to the high-pressure section.
This joint will often have a seal weld around its outer diameter to prevent the flow of air into or
steam out of the steam path.
The low-pressure casing is designed to accept steam from the exhaust of the expansion
immediately above it in terms of system pressure and temperature. The conditions of the steam
admitted to low-pressure casing are typically as follows.
In a fossil cycle. In these cycles the steam derives from the high, intermediate, or reheat
sections. Such steam is normally superheated. Its pressure is generally in the range of 70 to 200
psia. The initial temperature can be as high as 800°F. However, there are often limits placed on
this temperature not by considerations of the casing but rather by the operating temperature the
low-pressure rotor material can tolerate.
In a water-cooled nuclear cycle. In these cycles, the steam is admitted from the
intermediate system of the unit. Such an intermediate system will comprise a moisture separator
and possibly a reheater. Therefore, the steam conditions are typically in the range 70 to 250 psia,
and the temperature in the non-reheat cycle is at the saturation temperature corresponding to the
steam inlet pressure. In the case of the nuclear reheat cycle, steam is raised to a temperature less
than the initial cycle steam pressure saturation temperature by an amount equal to the terminal
temperature difference of the live steam reheater.
The steam at entry to the nuclear non-reheat, low-pressure section can contain moisture,
and the quantity is a function of the effectiveness of the moisture separator. For this reason, it is
probable the low-pressure, non-reheat cycle will have moisture present throughout the
5-19
expansion, and the casing must be designed to accommodate this moisture existing at high-
pressure levels and possibly having high velocities.
In an effort to maximize cycle efficiency and extract as much energy from the expanding
steam as effectively as possible, the exhaust pressure from the low-pressure section is passed to a
condenser that produces sub-atmospheric pressures in the low-pressure exhaust hood. The
condenser is normally optimized, designed, and selected to produce an exhaust pressure between
0.5 and 6.0 in. of mercury absolute (Hga) at all loads and with all cooling water temperatures.
As the exhaust pressure decreases there is an increase in the volumetric flow in the
discharge section of the L-0 blade system and casing. To minimize the frictional losses
associated with the resulting high-velocity flow of steam in the exhaust, the casing is normally
mounted directly above or adjacent and connected to the condensers. These exhaust casings can
also contain deflector plates designed to direct the steam into the condenser and distribute the
flow as evenly as possible over the entire flow down area.
The large volumetric flows associated with large modern units often requires multi-flow
exhausts be used so sufficient blade annulus area is available and the steam exhaust velocity is
limited to acceptable values.
Shown in Figure 5–7 is a double-flow low-pressure section with a monoblock rotor. From
this figure, it can be seen this section comprises a double-flow casing with five rows in each
flow. Both the inner and outer sections are fabricated. In this design the inner section is designed
to carry and support blade rings or diaphragms that are produced by the methods described in
chapter 6. The pockets used for the extraction of feed heating steam can also be seen.
Fig. 5–7 A Double-flow Low-Pressure Section
5-20
The inner shell is designed to locate the stationary blade rows and hold them in a correct
spatial position relative to the rotating blade rows. The stationary blade carrier elements or rings
are produced as outer diaphragm webs that carry one or more stationary blade rows. These outer
blade-ring carriers or webs locate directly in grooves machined into the inner casing fabrication
and permit adjustment within the inner casing to achieve optimum steam path alignment. These
fabrications also allow for space to remove steam for regenerative feed heating.
In the upper half of the low-pressure casings there are pressure relief or explosion
diaphragms. These diaphragms are designed to rupture and relieve any pressure that exceeds
atmospheric. Therefore, if for some operational or other reason vacuum is lost and the pressure
inside the low-pressure hood increases to a value above an acceptable limit, then the reversal of
pressure will deflect the diaphragms out and cause them to rupture. Rupture of the explosion
diaphragms will release the inner pressure of the casing, allowing the steam to escape from the
unit into the power-house or atmosphere in the case of an outside unit. The diaphragm rupture
pressure is normally between 15 and 30 psia. Rupture of these diaphragms will automatically
shut down the unit. If pressure were allowed to build up in the exhaust hood, levels of pressure
and temperature would increase to levels that would destroy the blade system.
Low-Pressure Casing Arrangement
The large number of multiple exhausts required for modern condensing units is
conveniently achieved by arranging for two or more double-flow sections in parallel. To achieve
a suitable temperature increase rate in the feed heating system and because of the large energy
range in low-pressure sections, three or four extraction of steam for regenerative feed heating are
normally required from the low-pressure expansions. Such an extraction requirement means that
the low-pressure hoods be produced so steam can be removed at a number of stage points in each
5-21
expansion. This requirement can complicate the general arrangement, and for some designs,
demands steam path and hood variation from one flow to the other.
If steam flow quantities are such that a single-flow section does not provide sufficient
discharge area, it is normal to arrange the low-pressure portion of the unit to employ a single or
multiple double-flow low-pressure sections. It was common at one time to employ designs with
three exhaust flows, with the one single expansion connected to the intermediate or reheat
pressure section discharge. This concept is not used extensively in the majority of modern units,
as it is more cost effective to develop modular designs of double-flow units, with specific
arrangements for steam extraction. These modular low-pressure designs also permit a better
mechanical arrangement of the low-pressure sections.
It is, therefore, becoming less common to employ an arrangement of three exhaust flows.
There are, however, still in successful operation a number units in which a single-flow low-
pressure section is connected directly to the intermediate section. This intermediate-
pressure/low-pressure (IP/LP) section can then be used with a single double-flow section to
provide a three-flow arrangement. In the three-flow arrangement, the first stationary blade row of
the low-pressure sections is set so that steam admitted to each of the three flows is controlled so
that with possible different steam extractions patterns in each. The exhaust flow from each
expansion last-stage blade row is the same.
The pressure range across the low-pressure sections is small when compared to the high
and reheat sections at one-fifteenth to one tenth their range. However, the energy extracted from
the low-pressure section can produce an output comparable to the sum of the output from the
other two expansions. Because of its large physical size and the fact that the space between the
outer hood and inner casing is maintained at vacuum pressure, there is a large downward force
resulting from the pressure differential between the inner hood and atmospheric. This total
pressure is sufficient to deflect the total casing vertically downward.
5-22
The low-pressure casing is, because of its size and the fact that it is not a massively rigid
structure like the cast high and intermediate or reheat sections, deflected downward. The extent
of this downward deflection is sensitive to the vacuum produced by the condenser. There can
also be a change in casing elevation as the level of water in the condenser hot well changes.
Older designs are still in use in which the total expansion from inlet condition to
condenser exhaust is achieved in a single casing. Shown as Figure 5–8 is the cross section of
such a unit, in which the casing is produced in sections that are bolted together to form a single
expansion. The low-pressure casing can be manufactured by either casting or fabrication. A seal
weld between the low-pressure and high-pressure casings may also be used. This casing is
designed to provide for the extraction of steam for regenerative feed heating and has a valve
chest produced integral with the high-pressure inlet.
Fig. 5–8 A Single Flow Unit With the Low-Pressure Casing Attached Directly to the
High-Pressure Section
Since the steam exhausting from the low-pressure section flows to the condenser, it is
convenient and economical to mount the low-pressure section above and connected directly to it.
While flexible connections exist, it is also convenient to weld the lower half casing to the
condenser shell to form a continuous structure.
The bearings supporting the low-pressure rotor can be constructed and supported in one
of two ways. These bearings are produced either with the bearing shells as an integral part of the
low-pressure fabrication or they are mounted external to the casing supported off the foundation.
There are two aspects of these two possible design alternates that should be considered and
evaluated.
5-23
1. When mounted from a pedestal on the foundation, the rotor elevation does not
change with vacuum or condenser hot well water quantity. It then becomes relatively
easy to predict the deflected shape and alignment requirements of the rotating
portion of the unit. However, because the casing will deflect downward under these
influences and must carry the stationary portion of the steam path, including the
sealing arrangement at the shaft end and diaphragms, there could be a need to
increase the radial clearance of the sealing systems in the low-pressure section to
allow for the difference in vertical deflection between the two sets of steam path
components.
2. When supported from the low-pressure fabrication, the radial seals at both the shaft-
end positions and the stationary blades can be maintained at or near optimum values
because the rotor will rise and fall with the bearings. This will minimizes leakage
losses. However, because the rotor will rise and fall as the vacuum changes, the
designer must have data on predicted deflection amounts to be able to establish the
normal running deflected form of the rotor.
With the bearings located within the exhaust hood, the rotor will have a shorter span,
limiting the bending stress induced in it.
Low-Pressure Casing Structures
The physical size of many low-pressure casings, particularly for 1500 and 1800 rpm
applications, are so large the casing must be constructed in several sections using vertical joints
in addition to the necessary horizontal joint split.
The two-casing (inner and outer) design of units consists of several portions, and these
should be considered separately because there are significant differences between them. These
casing segments are described next.
5-24
The outer upper shell
For the small exhaust-stage blade designs, it is possible to produce the entire upper
fabrication as a single structure. For large exhaust blade systems, the upper shell may consist of
two or more fabrications with these sections joined by bolted connection at the centerline. It is
necessary to break the structure into sections because of the shipping and handling restrictions.
There can also, in the largest fabrication, be limitations imposed by the size of machine
tools required to produce the components and the furnace size needed to complete any stress
relief requirements after welding.
The main structural components of an upper outer casing are shown in Figure 5–9. The
basic shell consists of a wrapper plate that provides the upper outer casing, and there may be
connections to this wrapper from the crossover pipe for steam admission. There will, in addition,
normally be provision for explosion diaphragms. The end walls are normally flat and must
provide sufficient distance from the exhaust or discharge line of the blades to the end walls. If
this space is not sufficient, there will be an unacceptable loss of the steam kinetic energy upon
impact with the walls causing an energy loss within the hood (see chapter 3). The wrapper will
require the use of reinforcing ribs and struts within the hood to provide strength against both
distortion and the downward atmospheric deflection.
Fig. 5–9 An Upper Outer Fabricated Casing
Design considerations for the upper hood require the wrapper and end walls be
sufficiently thick to resist deflection and distortion and be suitable for the vacuum pull. The outer
hood should also be designed so the bearings and steam seal components can be accessed
5-25
without the need to remove the outer casing. There will also be provision for access ports, so the
unit can be entered without the removal of the hood.
It is a normal design process to make a vibration analysis of the exhaust hood and then
use the reinforcing ribs and struts to de-tune the fabrication away from coincidence with any
natural frequencies developed within the structure.
The outer lower shell
The outer lower shell is the primary support structure carrying the low-pressure turbine
section. It must be capable of withstanding the vacuum load on both the side and end walls and
the vertical downward thrust transmitted to it by the upper half casing though the horizontal
joint. This structure must, if the low-pressure bearing is an integral part of the fabrication, carry
the bearings and support the weight of the rotor. The total downward thrust due to vacuum load
and weight must be carried through the casing while maintaining adequate bearing alignment.
The total low-pressure load is transmitted from the casing to the foundation by the support
brackets located at the sides and possibly the ends of the unit. The arrangement of a typical lower
half fabrication (half section) is shown in Figure 5–10. In Figure 5–11 is shown the lower half
casings of both the inner and outer portion with the double-flow rotor supported from the
bearings in the lower half.
Fig. 5–10 A Half Portion of a Lower Outer Casing
Fig. 5–11 An Open Low-Pressure Section Showing the Horizontal Joints of the Outer and
Inner Low-Pressure Casings
5-26
The outer lower shell is fabricated from carbon steel plates, and the shell is given rigidity
by the use of internal struts. These can be seen in Figure 5–12. A primary design consideration of
the lower half outer shell is its need to support and provide location to the inner casing and
possibly bearing cones. It is also important that it can maintain radial and axial alignment during
both normal and transient operation.
Fig. 5–12 A Lower Half Inner Casing Seen From Above the Horizontal Joint
The inner casing
The inner casing carries and supports the low-pressure section stationary blades and/or
diaphragms. These casings, in a double-flow configuration, contain at their center an inlet bowl
that accepts the incoming steam from the crossover/around pipes and directs it into the first stage
stationary blade row around the complete 360° flow annulus. The inner casing also contains
extraction steam belts that collect the feed heating steam from the main steam flow required for
regenerative feed heating. These belts extend around the complete blade outer circumference and
are connected to a pipe transporting the steam to the heaters.
In some designs, the lowest pressure heaters are located within the condenser body, but
the steam must still be transported from the extraction belt to the heater shell. Because the low-
pressure section will have moisture in several stages, there is also provision made in the inner
casing to collect centrifuged moisture or to locate the diaphragms that have provision for this
moisture collection. In this case, the low-pressure casing provides the drains that remove the
collected moisture. The lower half of an inner casing is shown in Figure 5–12, where the
fabricated arrangement can be seen.
This inner casing is normally a separate structure supported from the outer casing.
However, for older smaller rated units with a lower inlet temperature, it is possible to
5-27
manufacture the inner shell as an integral part of the outer. But for higher inlet temperatures, it is
necessary to manufacture the inner shell as a separate structure to accommodate the excessive
thermal gradients and differentials that can develop across the walls. In these older, lower rated
units with a lower inlet temperature, the single casing fulfills the requirements of the inner
casing. Also, it is connected directly to the condenser and therefore subject to the transient
thermal and load conditions normally experienced by the outer. These casings can be subjected
to high loading but the designer will allow sufficient margin that stress levels are well within
acceptable limits.
The inner shells are essentially open-ended, cylindrical pressure vessels with admitted
steam expanding axially in both directions in the double-flow configuration. In double-flow
designs, steam is admitted into an inner cylindrical annulus where it divides to flow axially out
through the steam path and exhausting to the condenser. Because the two flows are essentially
symmetrical, the axial thrust developed on the casing is also symmetrical and balanced. The
thrust developed in the tangential direction is in the same direction on both flows and therefore
additive. The casing must be keyed at its connection points to the other portions of the
foundation to ensure these thrusts are constrained. Relative to the weight of the structure these
thrusts are small, but there is normally some provision for containing them, particularly within
the individual blade rows.
The steam is admitted to the double-flow casing through one or two openings on top of,
at the bottom, or on the sides of the casing. The numbers and locations of these openings are
determined by the steam volumetric flow rate and conditions. The number of crossover/around
pipes, and their sizing is chosen so the mean steam velocity in the pipes and inlet annulus is not
greatly in excess of 150 feet per second (ft/sec).
Steam for feed heating is extracted from the inner casing at points immediately after the
rotating blade row. A circumferential opening into which the steam can flow is arranged around
5-28
the shell periphery. The opening of the circumferential annulus or belt and the extraction pipes
are sized so the steam velocity will not exceed about 150 ft/sec. A typical extraction arrangement
is shown in Figure 5–13. In this figure can be seen an arrangement incorporating a water catcher
belt that is arranged and positioned to collect and drain moisture carried in with the steam,
centrifuged into the belt from the rotating blades, or carried into it from the outer flow walls of
the casing.
Fig. 5–13 Details of the Fabricated Structure of Wrappers and Carrier Rings Required to
Achieve a Satisfactory Structure
The inner shell is normally surrounded on its outer surface by wet steam with a
temperature corresponding to the saturation temperature of the condenser pressure or exhaust
steam. The inlet temperature to the inlet bowl can be as high as 800°F although a more normal
value is 700°F. Therefore, it is clear there can be relatively large thermal differentials developed
across the inner casing at some locations. Many manufacturers elect to design their inner casing
with a heat shield surrounding the inner section to minimize this thermal gradient effect.
The circumferential bowl at inlet to the double-flow low-pressure section is located in the
center section of the inner shell, and the steam extraction pockets are spaced axially along the
length of the fabrication with each of these succeeding pockets at a lower temperature reducing
toward the exhaust. Depending upon the extraction points within the expansion, there can be
temperature differentials across the separating walls as high as 350°F. However, the actual
differentials across the walls may not be as high as the indicated steam temperature differentials
because of the moisture film coefficient on either side of the plate. The outer wrapper plate can
have the inner surface exposed to 800°F steam adjacent to the crossover bowl, and two inches
away in the axial direction the inner surface might be exposed to steam at 450°F.
5-29
The outer surface at this same location is exposed to the wet, cool condensing steam if no
thermal barrier is used. These multi-directional thermal gradients can induce extremely high
stresses and possibly cause casing distortions. If the stresses induced by these thermal gradients
are in excess of the yield strength of the material, then the distortions of the inner casing could be
permanent. Such distortions of the inner casing could result in clearance rubs and possibly
broken welds on reinforcing ribs and struts and therefore leakage at the various steam tight joint
faces. Should these stresses induce ruptures in the joining welds within the casing these can be
extremely difficult to access for weld repair.
Cast Low-Pressure Sections
While the majority of low-pressure casings are produced as fabrications, there are a
number of manufacturers that find that casting is still suitable because it is economical, reliable,
and capable of producing an effective product. The material used can be either cast iron or steel.
The principal material is cast iron. There are two forms of iron in use—graphite iron and the
spheroid graphite type. Cast iron is a material that is very suitable for casting, and it produces a
good quality form. Unfortunately the simple graphite cast iron cannot be easily upgraded if
defects are found. However, the spheroid graphite is readily welded and is therefore a suitable
material. Cast steel can be easily upgraded. Many modern two-casing low-pressure designs will
employ fabrication for the outer casing and casting for the inner casings.
Shown as Figure 5–14 is a large casting for an inner casing being turned after completion
of machining. The outer section into which this casing would be mounted would be produced by
fabrication. Shown as Figure 5–15 is the cast outer portion of a smaller unit.
Fig. 5–14 A Cast Inner Low-Pressure Casing Being Moved After Machining
5-30
Fig. 5–15 The Cast Iron Exhaust Hood and Low-Pressure Casing for a Small Output Unit
The material specifications
When steel is used for the production of an inner casing, the material requirements for
these elements are generally not as stringent as for the higher steam conditions. A typical
chemical composition is shown in Table 5–1.
Table 5–1 Typical Composition of Low-pressure Turbine
In addition to these elements, a minimal amount of aluminum will be permitted for
deoxidation. Some manufacturers will also specify a small level of copper (0.30 to 0.60%) to
help combat and minimize the effects of washing erosion. The physical properties of this low
carbon steel are shown in Table 5–2. These properties are established from test coupons cast
integrally with the main casting.
Table 5–2 The Mechanical Properties of Low-pressure Turbine
The procedures for producing patterns molds and cores for these castings are identical to
those used for the high-temperature, high-pressure elements discussed in previously in this
chapter. With this type of casting, internal chills are not used and external chills are used only to
help achieve a logical solidification pattern and material structure. At completion of cooling, the
casting is shaken out from the mold, and the feeder heads are removed before the casting has
cooled below 400°F. The feeder heads must be removed in such a manner the steel is not burned.
Before machining, the casting is given a visual inspection for major defects.
5-31
After rough machining the casting is given a nondestructive examination by magnetic
particle methods in accessible areas and radiographic examination in any weld preparation
regions
Acceptance level of casting defects
It is necessary to have established acceptance standards available for any casting faults
that might be found. The following criteria are intended to provide guidance only. There may be
other standards established by individual manufacturers for their units based on their
requirements and experience.
Visual acceptance standards. Folds, cavities, and clustered porosity with a depth greater
than 5% of the wall thickness should be removed by grinding. If the depth of the resulting cavity
is less than 10% of the wall thickness and the locations of the cavity are not in a region subject to
high stress levels, then these can often be accepted. It is best if the cavity is acceptable to blend it
out at its edges. If the cavity is greater than 10% of the wall thickness, then it should be weld
rebuilt and the requirements of stress relief applied.
Magnetic particle acceptance standards. Acceptable linear indications for critical and
non-critical regions are shown in Table 5–3. Surface defects are not permitted at planned weld
positions or at defect excavations. Discontinuous linear indications are considered acceptable
where the separation between adjacent indications is at least four times the length of the larger of
the two indications.
Table 5–3 Acceptable Magnetic Particle Inspection
5-32
If as a consequence of this magnetic particle examination unacceptable defects are found,
then they should be excavated and weld repaired. After weld repair the casting should be
subjected to a stress relief cycle.
Defects in machined areas. The casting must be free from sub-surface defects which
would be exposed on machining. Defects classified as being greater than ASME schedule 1
found by radiography in scheduled weld regions are not acceptable. Any such defects in this area
should be repaired by welding.
Welding repairs. If it becomes necessary to weld repair defects in the cast shells, the
faults must first be excavated by some suitable means such as grinding and/or chipping,
machining, or arc-flame gouging. In some areas it is necessary to grind smooth the excavations
before the repairs proceed. The normal method of weld repair is manual metal arc. It is also
necessary to preheat the casting before repairs begin and to maintain the preheat temperature
throughout the repair procedure. Preheat temperature is from 150 to 300°F. Depending on the
material and whether localized preheat is used, this must extent for a least 10 in. in all directions
surrounding the repair. As the filler material is laid in, it must be continually inspected to ensure
no cavities remain in regions where they could lead to cracking as the unit ages.
Heat treatment. When the casting requires heat treatment, it should be loaded into the
oven and heated at a temperature ramp rate that should not exceed 200 to 225°F/hour (hr). For
annealing, the temperature should be raised to about 1700°F and for stress relieving to 1100°F.
These actual temperatures depend on the material. Once the treatment temperature has been
achieved evenly throughout the oven, these temperatures should be maintained for a period of
one hour for every inch of thickness of the thickest wall in the casing, but not less than a
minimum period of 12 hours.
Machining. The large physical size and weight of these castings, particularly for the half
speed (1500 and 1800 rpm) units, makes their handling and turning a complex operation because
5-33
they are normally much larger than the high and reheat section elements. The machining process
is essentially the same as for the high-pressure elements, and final machining requires the halves
be firmly bolted together and preferably supported in the manner as they are to be installed in the
field. If this is not done, these casings will tend to deflect when installed in the outer casing, and
the steam path grooving or support surfaces will no longer be concentric because the casing will
have a different sag form.
The machining of cast casings is essentially the same as for the fabricated. Shown in
Figure 5–16 is a large low-pressure section set up for internal boring where the two halves are
firmly bolted and machined as a pair.
Fig. 5–16 The Final Machining of a Low-Pressure Inner Casing
Thermal Gradient and High-Pressure Shell Design
There is, because of the energy expenditure within the steam as it flows through the
steam path, a considerable thermal gradient along the axis of any casing. There is also a thermal
gradient through the thickness of the walls of the shell due to the differential temperature that
exists across them. Under normal operating conditions, the casing can adjust to and
accommodate these gradients, and the shells can continue to operate satisfactorily for many
years. However, during operation there are changes in the temperatures to which the various
components are exposed, dependent upon the condition causing the change and the rate at which
these changes occur.
The change of steam conditions with the greatest influence on the casing are those
changes that occur rapidly and cause an increases in the levels of stress developed in the casing
walls. These stresses can be sufficient to induce failure due to the phenomena of low cycle or
thermal fatigue. Such failures can occur after a few thousand or even a few hundred such cycles,
5-34
dependent upon the severity of the temperature change and the stress levels and the normally
severe concentration of stress that exists at these points. Figure 5–17 shows typical cracks in a
high-pressure unit due to the phenomena of low cycle fatigue. (D.P. Timo 1970) This crack
initiates at a sharp female corner where stress concentration is high.
Fig. 5–17 Portion of a High-Pressure Shell Showing the Circumferential Cracks Formed
in the Filet Radii Positions as a Consequence of Thermal Cycling
It is of interest to consider the magnitude of stress occurring across any component due to
temperature mismatch between an inner hot surface and an outer cooler surface. Consider an
element of shell wall shown in Figure 5–18 where the temperature on the inner hot surface is
shown as T1 and on the outer cooler surface as T2. In this wall, the temperature gradient or
mismatch is _T. Consider the mean gradient or change of temperature in three cases.
Fig. 5–18 Temperature Profiles Through the Walls of a Casing Under Various Heating
Cycles
Linear temperature degradation. Figure 5–18a shows a casing under normal operating
conditions with a gradient that is practically linear from T1 to T2. In this case a compressive
stress will exist between the hotter wall and the neutral axis and a tensile stress from the neutral
axis to the colder surface. From the zero stress of the neutral axis to the maximum stress in the
outer fibers of the wall material, there will be a local temperature gradient _T equal to 1/2(T1-
T2), and the stress will have a maximum value of fs.
fs = ∆T . µ . E
1 - S (5.1)
5-35
where
fs = stress
_ = linear coefficient of thermal expansion
E = young modulus
S = poisson ratio
Parabolic degradation. Normally the temperature gradient will not be linear, but will
follow some other distribution such as the parabolic seen in Figure 5–18b. In this case, the
maximum local temperature gradient _T is 2/3(T1-T2).
A condition that can subject the casing and other portions of the unit to high thermal
stress and where manufacturer’s recommendations should be followed in detail, is control of the
temperature ramp rates at start-up and shutdown when correct procedures can be controlled and
followed.
Hyperbolic degradation. The most severe conditions however exist at start-up or
shutdown when hot steam is initially admitted to the unit, washing the cold metal inner surface
with hot steam. Under these conditions hot steam flows suddenly through the unit, washing the
cold surfaces. The temperature gradient _T then approximates T1-T2 as seen in Figure 5–18c and
results in a significantly higher stress level in the outer fibers of the wall material. These stresses
become particularly significant in any region where there is high stress concentration such as at
section changes or where there are small fillet radii.
Sudden temperature changes caused by load shedding or boiler excursions also introduce
this situation. Sudden temperature changes when the unit is hot are possibly more sever than at
start-up because the material is hotter and therefore has poorer mechanical properties.
To the greatest extent possible manufacturers will avoid fillet radii that are too small.
Unfortunately some design requirements demand these be present. Also some manufacturing
5-36
techniques require and produce such radii as a function of the technique itself. Cast surfaces,
particularly those internal to the steam inlet annulus that cannot be inspected visually and are
difficult to access, are prime candidates for producing regions where stress concentration can be
high.
The temperature gradient, and therefore the thermal stresses developed in portions of the
shell, can be limited by adjusting the rate at which the boiler conditions change or by adjusting
the turbine start-up rate. Unfortunately it is inevitable that during the life of the unit, there will be
some start-ups in which the thermal stress exceeds the yield strength of the material in some
portions of the shell. Similarly, there will be uncontrollable excursions where these stresses are
exceeded. Such situations are regrettable but can and must be accepted. The immediate and
cumulative effects of the excessive plastic strains induced must be understood and be readily
measurable.
In an effort to understand and limit this effect, many turbine builders have introduced
systems of measuring, recording, and aggregating the contribution of each start-up temperature
change or excursion, whether the induced thermal stress is of a high or low magnitude, toward
initiating a surface crack. The level of temperature mismatch between main steam and initial
metal temperature at start-up—temperature change—is converted to a low-cycle fatigue index
(LCFI). A typical curve of such an index is shown in Figure 5–19.
Fig. 5–19 LCFI as a Function of Temperature Changes
When the summation of all individual indices over the years of the unit’s operating life
reaches 100%, there is a possibility a surface crack will have initiated. This may not be harmful,
and it may require a further 100% aggregation for the crack to propagate to a significant depth
and even more operation before rupture would occur. If cracks are discovered early in their life,
5-37
then they can normally be removed by grinding. However, the removal of material will not solve
the problem created by temperature transients. If the crack is in a position where grinding can be
undertaken, then grinding can slow the rate at which the crack will propagate, but will not
prevent its reoccurrence.
High thermal stresses and the resulting accumulation of high individual indices can be
prevented by ensuring the temperature of the steam washing the inner surface is only slightly in
excess of the temperature of the core of the metal. If possible, the mismatch temperature should
be limited to values between -50 and +100°F, although acceptable outer limits are -175 and
+270°F. Here, a minus sign (-) indicates the main steam is cooler than the internal metal
temperature and a plus sign (+) indicates the main steam is hotter than the metal. The actual
values of acceptable temperature differentials for any unit will depend upon various factors
including the thickness of the metal section and the thermal conductivity of the shell material.
These recommended temperature differentials are normally the limiting factor to unit
start-ups, and the average thermal gradient during operation should not exceed the recommended
if an acceptable life is to be expected from the equipment. There may be some parts of the unit
where steady-state temperature differences larger than this will occur during normal operation. In
such cases, parts will have been designed to accommodate this and be of suitable materials and
form to provide the degree of flexibility required to prevent excessive stress.
Estimating Low-Cycle Fatigue Life
The model just discussed for considering the production of thermal stresses and their
effect on casing life together with the introduction of low cycle fatigue cracks, although very
simple, provides a relatively simple tool for estimating casing life consumption due to start-up,
shutdown, and during operation transients when large temperature changes occur.
5-38
The life expenditure that occurs during any unit transient is a function of three
factors—the magnitude of the thermal stress induced in the shell, the material properties of the
shell, and the environmental temperature at which the change has occurred.
Any component repeatedly subjected to stresses beyond the yield strain of the component
material at its operating temperature will develop cracks in a finite number of cycles. The
number of cycles required to initiate these cracks is a function of the stress level.
Turbine shells are a relatively complex form, and they contain regions where during
operation there is a considerable degree of stress concentration in the parts that are subject to
biaxial loading. Therefore, the calculation of actual stress levels by traditional methods is
difficult, although finite element methods have allowed a much better understanding of the loads
and stresses involved in casings. Because of these difficulties, designers find it is of considerable
advantage to calibrate experimental values of stress against calculated values, which are
normally determined by finite element methods.
Figure 5–20 shows a portion of an outer casing scale model equipped with strain gauges
used to predict actual values. These stresses are then compared with calculated values, which
permits experimental factors to be established which can be applied to other casings with similar
geometries to obtain an adequate degree of accuracy.
Fig. 5–20 Scale Model of an Outer Shell Instrumented With Strain Gauges to Help
Establish Stress Levels
In any repeated stress/strain situation in which the stresses developed within the
component are in excess of the yield stress in tension and compression, it is usual practice to plot
the total (elastic plus plastic) strain elastic plus static (EET) against cycles in determining the life
factor for the component. Two curves for a typical shell material are shown as Figure 5–21. The
5-39
materials selected for any component have a considerable effect on the low-cycle fatigue. In
Figure 5–21curve A being a low strength relatively ductile alloy and curve B a stronger, less
ductile one.
Fig 5–21 Elastic + Plastic Strain as a Function of Stress Cycles for Two Different Shell
Materials
It has been shown that at room temperature, low-cycle fatigue life may be predicted by
the following expression. (Tavernelli and Coffin 1961)
∆εp = 0.5 Ln [100/(100 - %Ra)]
N (5.2)
where
__p = plastic strain range
%Ra = percent reduction in area measured in a tensile test specimen
N = the number of cycles to cracking
This expression is valid in the high strain range where the ratio of plastic to elastic strain
is high. In any casing form, it is normal for the designer to determine the number of cycles N to
initiate a crack. This number is factored into the total design considerations including the
selection of the material to be used.
Thermal Gradient in the Low-Pressure Inner Casing
In many low-pressure sections, problems are encountered due to thermal gradients
causing permanent distortion of the inner casing. It is normal for these gradients to occur in both
the axial and radial direction. Therefore, the distortion which results can occur in a complex form
5-40
in the casing structure causing permanent set in both directions. In addition to the predictable
temperature differentials from the expansion and extraction of steam, it is known that the skin
surface temperature of the outer surface of the inner casing varies in an unpredictable manner
and is influenced by changes in load and varies from the upper to the lower halves at any
transverse section.
At exhaust from the last stage blade annulus, the steam is deflected to flow into the
condenser. However, there are spaces between the inner casing outer surface and the inner
surface of the top half outer casing. These spaces fill with flowing steam, which passes through
them to the condenser and the surfaces therefore attain steam temperature. This represents a
thermal gradient on the inner casing walls. However, many designs of inner casings are arranged
to include a thermal barrier attached to the outer surface of the inner casing. This barrier helps
ensure the temperature gradient across the wall is not as severe as that caused by the outer
surface of the inner casing attaining steam temperature.
There are within the exhaust hood factors that cause temperature variation and non-
symmetric flow. These include:
• unit load. As the unit load varies, so will the quantity of steam flowing through the
steam path, which will in turn modify the flow velocities and patterns through the
spaces between the hoods.
• exhaust pressure. As the exhaust pressure produced by the condenser changes, there
will be a change in the steam specific volume and the volumetric flow will change.
Also, as the condenser pressure changes so will the saturation temperature of the
steam that covers the metal surfaces. There will be changes in steam velocity and
temperature associated with condenser pressure changes.
• Rotational effect. While the flow pattern of the steam at exhaust from the last stage
blade annulus will be substantially axial, it will possibly have some tangential and
5-41
radial component to its flow. These effects will produce a total flow distribution that
is different from side to side within the casing. This side-to-side flow difference will
influence the total flow patterns within the hood, which will be sensitive to any small
change in either steam pressure or quantity.
In designing a low-pressure section, sufficient flow area must be made available to the
exhausting steam to minimize the pressure drop from the blade annulus to the condenser. There
are two important considerations to this requirement.
1. The exhaust blades will discharge their flowing steam into a diffuser that is produced
as part of the low-pressure section fabrication. The diffuser form is selected to
minimize losses associated with removing the steam away from the exhaust plane and
not impede further flow from the blades.
The diffuser is normally constructed from rolled plate that is either welded or bolted
to the inner casing. The axial distance from the exhaust blade annulus to the casing
end wall is limited, and it is difficult to achieve a perfect arrangement within the axial
length available. However, designs can be provided that allow a diffuser section to be
used and can be accommodated within the available axial space to help minimize the
losses which occur.
2. The hood structure must turn and divert the steam, normally downward, to the
condenser, causing a minimal frictional loss within the hood. Hoods are designed so
strategically placed diverter plates can turn the steam in an effort to keep the flow
density at any point relatively constant, avoiding excessive velocities and minimizing
frictional losses.
5-42
During operation, the inner casing must remain sufficiently rigid it is able to maintain
concentricity in the radial direction and retain axial alignment. It must do this and yet remain
sufficiently flexible it is able to respond to large temperature swings that occur within short
periods of time. These requirements of flexibility and rigidity are obviously contradictory.
However, it is important that alignment and rigidity requirements are addressed in the design
phase. If any forced compromise is required by one requirement, then it must be recognized by
and accounted for in defining the requirements of the other. These requirements can be
aggravated by any large temperature gradients, and the designer must anticipate the most severe
condition when defining the low-pressure sections.
There are various approaches that have been considered to solving the thermal gradient
problems encountered in designing and manufacturing the low-pressure casings. An attempt
could be made to reduce or eliminate the radial and axial gradients by insulating the various
members of the fabrication. Also, the radial gradient could be reduced by insulating the inner
surface of the wrapper plate. However, any insulation used would be exposed to wet steam, and
unless this insulation was impervious to water soak, it would immediately loose its insulation
properties on becoming wet. A ceramic insulation would overcome water soak problems but
would be unable to expand and contract adequately to accommodate casing movement.
To reduce or eliminate axial temperature gradients, the shell would have to be made of
several cylindrical sections to minimize conductive heat transfer. This would require making the
inlet bowl and each extraction belt a separate fabrication. This is obviously an expensive solution
since it would require a multiplicity of transverse flange faces and could present considerable
alignment problems. In addition, thermal cycling of the casing affects the bolting on flange faces,
and each separate fabrication would need to be supported individually from the lower half outer
casing.
5-43
Since there is considerable difficulty in reducing or eliminating gradients, the normal
engineering approach has been to design the structures so they are able to accept the anticipated
gradients and not have stresses induced in them which exceed the yield strength of the material.
This is the design approach currently pursued by manufacturers. It has so far proven to be an
acceptable solution, but the costs of producing the casings are increased by the use of more
expensive material and thicker sections in some locations than are required from the simple
consideration of normal (non transient) gradients.
The schematic of an inner shell, shown as Figure 5–22, indicates the predicted steam
temperatures and pressures at various locations within an inner shell arranged for steam
extraction pockets. These conditions are consistent with normal operation and will change during
transient operation. In this type of design, the only connections from the inner support sections to
the cooler wrapper plate are relatively thin supporting ribs. These ribs are free to deflect and
move axially under the influence of both thermal growth and diaphragm thrust. This type of
design eliminates the compressive stresses that would be present in the ribs if these had been
massive structures and the extraction pockets or belts had not been circumferential, thus
permitting limited axial movement.
Fig. 5–22 Temperatures at Various Locations in a Fabricated Low-Pressure Casing
High-Pressure Turbine Shell Materials
Advancing steam conditions and increases in diameter, particularly for half-speed
machines, have required a continual improvement in both the composition and mechanical
properties of the material and the manufacturing techniques used to produce steam turbine casing
shells. This is particularly important when applied to high-temperature, high-pressure units.
Casings are produced from alloy steels, and the castings are carefully controlled both to ensure
5-44
mechanical strength and freedom from casting defects which have the capability to compromise
the integrity of the shell.
For temperatures up to about 750°F, a material that is produced to American Society for
Testing and Materials (ASTM) A27 Grade 65-35 will normally be acceptable. The mechanical
and chemical specifications for this material may be modified by closer control of the chemical
constituents and the heat treatment undertaken. However, the material will generally meet the
overall requirements of this specification. For increased temperatures a more suitable
specification is one that accords closely with the requirements of ASTM A356 Group 8, and for
the highest steam temperatures up to about 1100°F the ASTM A356 Group 9 specification is
most suitable.
Typical chemical constituents of these materials are shown in Table 5–4 and the
minimum acceptable mechanical properties in Table 5–5. Turbine builders will modify these
basic requirements to suit their particular philosophies, applications, and design requirements.
This is acceptable and reflects the experience from many years of operation.
Table 5–4 Turbine Shell Castings Nominal Chemical Composition