17AE11-Propulsion - I Unit 3 - COMPRESSORS

17AE11-Propulsion – I
UNIT – III COMPRESSORS

• Principle of Operation of Centrifugal Compressor
• Work Done and Pressure Rise
• Velocity Diagrams/Velocity Triangles
• Diffuser Vane Design Considerations
• Concept of Prewhirl, Stall and Surge
• Elementary Theory of Axial Flow Compressor
• Degree of Reaction,
• Compressor Blade Design
• Centrifugal and Axial Compressor Performance
Characteristics.
17AE11-Propulsion – I
UNIT – III COMPRESSORS
UNIT - III
• COMPRESSORS:
Principle of Operation of Centrifugal Compressor – Work Done
and Pressure Rise – Velocity Diagrams – Diffuser Vane Design
Considerations – Concept of Prewhirl, Stall and Surge, Elementary
Theory of Axial Flow Compressor – Velocity Triangles – Degree of

Reaction, Compressor Blade Design, Centrifugal and Axial
Compressor Performance Characteristics.
Multi Stage Centrifugal Compressor
Centrifugal Compressor Impeller
Gas turbine
Axial flow Compressor rotor
Axial flow Compressor rotor blades
• This chapter is to treat the rotating modules of both aero engines and
gas turbines. These four chapters discuss the compressors, fans, and
turbines. Since all of these are identified as turbomachines, it is
necessary to introduce a definition for turbomachines here.
• The word (turbo) or (turbines) is of Latin origin, meaning “that which
spins or whirls around.” Turbomachines are often referred to as rotor
dynamic devices that transfer energy to or from a working fluid through
the forces generated by a rotor.
• For compressors or fans, energy is transferred from the rotor to the
fluid. The reverse occurs in turbines that deliver shaft power in exchange
of thermal energy taken from the working fluid.
• In this chapter, details about dynamic compressors will be discussed.
Brown classified compressors as either intermittent flow or continuous
flow ones (Figure).
• Intermittent flow compressors are positive displacement ones. They are
classified as either reciprocating or rotary types. Such positive
displacement compressors achieve increase in pressure by trapping fluid
in a confined space and transporting it to the region of higher pressure.
• Continuous flow compressors are classified as either ejectors or
dynamic compressors. Dynamic compressors are further classified as
radial flow, mixed flow, or axial flow compressors.
• Dynamic compressors have close similarity to propellers and prop fans.
In both, energy is transferred from the blades to the working fluid.
• However, they differ in influenced fluid quantity. Dynamic compressors
accommodate a finite quantity of working fluid that flows steadily in an
annular duct that comprises a hub and casing.
• Consequently, it is defined as enclosed machines. On the contrary,
propellers and prop fans influence unbounded quantity of air; in other
words, they are turbomachines without shrouds or annulus walls (or
casing) near the tip. Thus, these machines are termed as extended
turbomachines.
General Operating Specifications
• Dynamic compressors work by converting velocity to pressure in a

continuous flow. They are efficient, compact, and handle vast quantities
of working fluid.
• Considering air as the working fluid at 1.013 bar and 288 K, the inlet
flow is typically 150 kg/s (122 m3/s or 259,000 ft3/min) per square meter
of inlet area at Mach 0.4.
• The intake diameters of various types range from 0.1 m (4 in.) to 2.44 m
(8 ft). The range of mass flow varies from about 1.2 to 700 kg/s.
• The isentropic efficiency ranges typically from 0.7 to 0.9.
Centrifugal Compressor
• Generally, centrifugal compressor is called so because the flow through the
compressor is turned perpendicular to the axis of rotation.
• This type of compressor is composed of three main elements, namely, the
rotating part or impeller, the stationary part or stator, and a manifold or a
collector (denoted by scroll or volute).
• The impeller has several blades referred to as vanes. The diffuser may have
vanes or may be vane less.
• An important part upstream of the impeller is the inducer duct. The nature
of flow through the impeller is strongly dependent on the duct
arrangement upstream.
• This could be a simple straight duct, a curved inlet duct, a curved return
passage of a multistage compressor, or the final stage of an axial
compressor.
• These alternative configurations can lead to distorted flow conditions at the
impeller inlet, which in turn will lead to an overall performance
deterioration of the compressor.
GOVERNING EQUATIONS
• The thermodynamic state of an ideal fluid is described by two
properties, which may be the pressure and temperature.
• For fluid in motion, its velocity (magnitude and direction) must also
be specified.
• The centrifugal compressor illustrated in Figure defines three states,
namely, the impeller inlet (state 1), impeller outlet (state 2), and
diffuser outlet (state 3). These states are also plotted on the T–S
diagram in Figure .
• Each state is defined by two points, namely, the static and total
(stagnation) conditions. Thus, two pressure lines (static and total) are
seen for each state. The compression processes are assumed
adiabatic irreversible; thus entropy increase is found in each process.
• The velocity triangles at the inlet and outlet of the impeller are shown in
Figure. The velocity triangles in any turbomachine are governed by the
relation
• As shown in figure, the flow at approach to the eye may be axial (C1) with
a typical velocity of 150 m/s for air or may have a swirl angle (in case of
prewhirl as will be discussed later).
• The flow velocity when added to the impeller tangential (rotational)
velocity (U1) will give the relative velocity (W1), which varies in magnitude
and direction from eye hub to tip since (U1) varies with radii.
• This requires the inducer vane to be twisted to align with the flow at all radii. It is
normal for the relative Mach number at the eye tip to be below 0.9, which falls to
below 0.5 for most of the subsequent passage length to minimize frictional losses.
• This inlet velocity is drawn in the axial–tangential plane. At the impeller outlet, the air
leaves the passage with a relative velocity (W2), which when added vectorially to the
tangential speed (U2), will give the absolute velocity (C2), which is greater than (C1) and
represents the effect of work input to the rotor shaft.
• The tip tangential velocity (U2) at outlet is slightly greater than the whirl
velocity (Cu2); the difference is called the slip. It is interesting to note here
that the outlet velocity triangle is at the radial–tangential plane. Thus, the
inlet and outlet velocity triangles at the inlet and outlet of impeller are in two
different planes as shown in Figure.
Velocity triangle
• In turbomachinery, a velocity triangle or a velocity diagram is a triangle
representing the various components of velocities of the working fluid in a
turbomachine.
• Velocity triangles may be drawn for both the inlet and outlet sections of
any turbomachine.
• The vector nature of velocity is utilized in the triangles, and the most basic
form of a velocity triangle consists of the tangential velocity, the absolute
velocity and the relative velocity of the fluid making up three sides of the
triangle.
Velocities involved
• A general velocity triangle consists of the following
vectors:
• V : Absolute velocity of the fluid.
• U : Blade Linear velocity.
• Vr: Relative velocity of the fluid after contact with rotor.
• Vw: Tangential component of V (absolute velocity), called
Whirl velocity.
• Vf: Flow velocity (axial component in case of axial
machines, radial component in case of radial machines).
The following angles are encountered during the analysis:
• α: Angle made by V with the plane of the machine
(usually the nozzle angle or the guide blade angle).
• β: Angle of the rotor blade. Absolute angle
• The impeller vanes may be radial at outlet or they may be inclined
backward or forward, thus identified as backward-leaning (commonly
now described as backswept) or forward leaning compressor (Figure)
3 Basic Fan Laws
• Fan Law 1 tells us that the change in air flow rate of a fan is proportional to the change in
speed of the propeller. If the propeller speed is increased by 10%, the air flow rate will also
increase by 10%.
• Fan Law 2 tells us that the change in total static pressure (SP) of the ventilation system will
increase by the square of the change in propeller speed of the fan. If the propeller speed is
increased by 10%, the total static pressure will increase 21%.
• Fan Law 3 tells us that the change in horsepower (HP) required by the fan to turn the propeller
will increase by the cube of the change in propeller speed of the fan. If the propeller speed is
increased by 10%, the horsepower required to turn the propeller will increase 33.1%.
CFM stands for Cubic Feet per Minute

Concept of Prewhirl
• Intake or inlet passage to the compressor may or may not be fitted with
inlet guide vanes (IGV).
• IGVs are designed to give either positive or negative prewhirl (Figure).
• IGVs giving a positive prewhirl are frequently found in high-speed
compressors, say in aero engines, to minimize the possibility of air
reaching or exceeding the sonic speed.
• In this case, possible shock waves formed will lead to great losses in the
intake. Negative prewhirl is employed to increase the inlet relative speed
at some applications. These IGVs may also be set at fixed angle or they
can be rotated to vary the inlet flow angle.
Axial-Flow Compressors and Fans
• In axial compressors, the air flows mainly parallel to the rotational axis.
• Axial-flow compressors have large mass flow capacity, high reliability, and high
efficiency, but have a smaller pressure rise per stage (1.1:1 to 1.4:1) than
centrifugal compressors (4:1 to 5:1). However, it is easy to link together several
stages and produce a multistage axial compressor having pressure ratios up to
40:1 in recent compressors.
• Integrally bladed rotors permit blade speeds significantly above conventional
rotors and hence stage pressure ratios greater than 1.8 [1].
• Axial compressors are widely used in gas turbines, notably jet engines, wind
tunnels, air blowers, and blast furnaces. Engines using an axial compressor are
known as axial flow engines; for example, axial flow turbofan.
• Almost all present day jet engines use axial-flow compressors, the notable
exception being those used in helicopters, where the smaller size of the
centrifugal compressor is useful.
• The fan in turbofan engines is also an axial-compression module, which is treated
as an axial compressor having a fewer number of blades of very large height, wide
chord, and large twist.
• These fans may be single or up to three stages in low-bypass turbofan engines.
Nowadays, jet engines use two or three axial compressors for higher pressure
ratios.
• Nearly all the turbojet engines are of the two-spool type, which have two
compressors identified as low- and high-pressure compressors. Two-spool
turbofan engines have the fan as the low-pressure spool. Sometimes, this fan
represents the first stage of the low-pressure compressor.
• The fan in this case is followed by few stages of an axial compressor that
represent the first module of the engine core. In other turbofans, the fan by its
own represents the low-pressure compressor, which is followed by the high-
pressure compressor.
• Three-spool turbofan engines have the fan, intermediate pressure, and high-
pressure compressors. Low-pressure compressor turns at the lowest rpm while
high-pressure compressor turns at the highest speed.
• Atypical axial compressor depicted in Figure has a series of rotating “rotor” blades
followed by a stationary “stator” set of blades that are concentric with the axis of
rotation.
• The compressor blades/vanes are relatively flat in section. Each pair of rotors and
stators is referred to as a stage, and most axial compressors have a number of
such stages placed in a row along a common power shaft in the center.
• The stator blades are required to ensure reasonable efficiency; without them, the
gas would rotate with the rotor blades, resulting in a large drop in efficiency.
• The axial compressor compresses the working fluid (here, only air will be treated)
by first accelerating the air and then diffusing it to obtain an increase in pressure.
The air is accelerated in the rotor and then diffused in the stator.
• This is illustrated in Figure 13.1, where the absolute velocity (C) increases in the
rotor but decreases in the diffuser. For successive stages, a saw-teeth pattern for
the velocity is obtained, while the static pressure continuously increases in both
the rotor and stator rows of all stages.
• The axial compressor is built up of a series of stagers, each consisting of a disc of
rotor blades followed by a ring of stator vanes. The axial compressor is generally
composed of four main elements: front frame, casing with inlet (stator) vanes,
rotor with rotor blades, and rear frame.
• The front frame in turbojet engines is a ring-shaped single piece lightweight
structure made up of aluminum alloy or steel, usually cast and then machined.
• It is composed of an outer ring, an inner hub, and 6–8 streamlined supporting
struts. If the compressor is a part of the turbofan engine, then this front frame is
replaced by a row of inlet guide vanes (IGVs).
• The compressor casing is a tube-like construction split lengthwise to facilitate
engine assembly and maintenance. To retain the stator blades and variable IGVs in
modern engines, the inner surfaces of the casing are machined with
circumferential T-section grooves.
• The final ring of the stator blades (vanes) may be called outlet guide vanes (OGVs),
as they guide the flow to the axial direction to suit the compressor outlet.
• After installing the rotor, both casing halves are bolted together through
longitudinal flanges. The compressor casing is made up of lightweight titanium.
• The third constituent of the compressor is the rotor assembly.
• The rotor blades (similar to the stator ones) have aerofoil section similar
to the aircraft wing, but they are highly twisted from root to tip to obtain
the optimum angle-of-attack to the flow everywhere along the blade
length.
• The reason for that twist is that the root section travels much slower
than the tip section.
• Thus for a constant axial velocity, the relative velocity and angle change
from root to tip.
• The length of the blades decreases progressively downstream in the
same proportion as the
• pressure increases. The rotor blades are attached by a dovetail root, pin
fixing, fir tree, and straddle T root.
• These sets of rotor are mounted on the rotor shaft through drums,
individual discs, bolted discs, or clamps.
• The discs are assembled with a number of tie-bolts, with the bolt-circle
diameter selected to produce a dynamically stiff rotor and good torque
transmission.
• The last constituent is the compressor rear frame that guides and delivers
the pressurized air stream to the combustion chamber. The center of the
rear frame accommodates the rearward bearing of the rotor that absorbs
the longitudinal thrust of the rotor.
• Moreover, the struts of the rear frame provide structural strength,
lubrication, and venting of the bearing as well as supply of bleed air.
• As the flow usually has a constant axial velocity through the compressor,
the cross-sectional area of the annular passage reduces toward the outlet
(or from the front to the rear).
• This is due to the decrease in the volume of air (thus increasing the density)
as compression progresses from stage to stage.
• Three different layouts are possible (Figure). The first layout maintains a
constant casing diameter of successive stages and increases the hub
diameter of successive stages. This design provides the maximum Mach
number and pressure ratio.
• In the second pattern, the hub radius is kept constant and the casing
diameter is increased, which provides the best geometry of a compressor.
• The final pattern has a constant mean diameter, which is associated with a
decrease in casing diameter and an increase in the hub diameter. This typical
layout is generally employed in deriving the basic theory of axial
compressors.
COMPARISON OF AXIAL AND CENTRIFUGAL COMPRESSORS
• The centrifugal compressor achieves part of the compression process by causing
the fluid to move outward in the centrifugal force field produced by the rotation of
the impeller.
• Thus, the pressure rises due to the change in the potential energy of the fluid in
the centrifugal force field of the rotor.
• On the contrary, the pressure rise in axial-flow compressor rotors and stators is
achieved by the exchange of kinetic energy with thermal energy in a diffusion
process.
• The pressure rise in a centrifugal compressor is therefore less limited by the
problems of boundary layer growth and separation in adverse pressure gradients.
• The essential feature of airflow in the impeller of a centrifugal compressor is that
all the fluid leaves the rotor at the tip radius rather than over a range of radii as in
the axial compressor.
• Probably for this reason, the centrifugal compressor, which first attained
a range of pressure ratio and efficiency useful for turbojet engines, was
used in the von Ohain engine (1939) and Whittle engine (1941).
Ohain engine (1939) Whittle engine (1941).

ADVANTAGES OF THE AXIAL-FLOW COMPRESSOR OVER
THE CENTRIFUGAL COMPRESSOR
1. Smaller frontal area for a given mass rate of flow (may be 1/2 or
1/3); thus, the aerodynamic drag or nacelle housing the engine is
smaller as shown in Figure.
2. Much greater mass flow rates (e.g., present day axial
compressors); have mass flow rates up to 200 kg/s (up to 900 kg/s
for high-bypass ratio turbofan engines), while centrifugal
compressors have mass flow rates less than 100 kg/s.
3. Flow direction of discharge is more suitable for mitigating; thus,
suitable for large engines.
the aerodynamic drag or nacelle housing the engine is smaller
4. May use cascade experiment research in developing
compressor.
5. Somewhat higher efficiency at high pressure ratio (perhaps
4%–5% higher than centrifugal compressor).
6. Higher maximum pressure ratio, which was about 17 in the
1960s and achieved up to 45 for the present transonic
compressors.
ADVANTAGES OF CENTRIFUGAL-FLOW COMPRESSOR OVER THE
AXIAL-FLOW COMPRESSOR
1. Higher stage pressure ratio (5:1 or even 10:1).
2. Simplicity and ruggedness of construction.
3. Shorter length for the same overall pressure ratio.
4. Generally less severe stall characteristics.
5. Less drop in performance with the adherence of dust to blades.
6. Cheaper to manufacture for equal pressure ratio.
7. Flow direction of discharge air is convenient for the installation of an
intercooler and/or heat exchanger in gas turbines.
8. Wider range (margin) of stable operation between surging and choking
limits at a given rotational speed as shown in Figure.
MAIN POINTS FOR
COMPARISON OF
CENTRIFUGAL
AND AXIAL
COMPRESSORS
MEAN FLOW (TWO-DIMENSIONAL APPROACH)
• The flow is considered to occur in the tangential plane at the mean
blade height where the peripheral rotation velocity is U (Figure).
• This two-dimensional approach means that the flow velocity will have
two velocity components:
(1) axial denoted by subscript (a) and
(2) peripheral denoted by subscript (u).
• Here, the radial velocity component is neglected.
• The mean section for both the axial compressor and axial turbine is
shown side by side in Figure , which illustrates how the hub-to-tip ratio
of axial compressor is much smaller than its value in axial turbine.
• For an axial compressor having a constant casing diameter, the radius
of the mean section is increasing in the rearward direction as shown in
Figure.
• The velocity triangles at the inlet and outlet of a single stage are shown
in Figure .
Lect-31
In this lecture...
• Centrifugal compressors
• Thermodynamics of centrifugal
compressors
• Components of a centrifugal
compressor
2
Prof. Bhaskar Roy, Prof. A M Pradeep, Department of Aerospace, IIT Bombay
Lect-31
Centrifugal compressors
• Centrifugal compressors were used in the
first jet engines developed independently
by Frank Whittle and Hans Ohain.
• Centrifugal compressors still find use in
smaller gas turbine engines.
• For larger engines, axial compressors
need lesser frontal area and are more
efficient.
• Centrifugal compressors can develop
higher per stage pressure ratios.
3
Lect-31
Centrifugal compressors
• Besides small aero engines, centrifugal
compressors are used in the auxiliary
power units (APUs) in many aircraft.
• Some of the aircraft air conditioning
systems employ centrifugal compressors.
• In a few engines, centrifugal compressors
are used as the final stage of the
compression process downstream of a
multi-stage axial compressor. Eg. GE T
700, P&W PT6, Honeywell T53.
4
Lect-31
Centrifugal compressors stage
Typical centrifugal compressor rotors
5
Lect-31

3
2
1
Schematic of a typical centrifugal compressor

6
Lect-31

Total losses
P02
Impeller losses P03
T
T02=T03 02 03 P3
C 32
02s 2c p
C 22
3 P2
2c p 03s
P01
T2
2 P1
T01 C12 01
2c p
T1
1
s
T-s diagram for a centrifugal compressor
7
Lect-31

The torque applied on the fluid by the rotor
τ=m  [(rC w )2 − (rC w )1 ], where 1 and 2 denotes the
compressor inlet and outlet, respectively.
The total work per unit mass is therefore,
 = Ω[(rC w )2 − (rC w )1 ]
w = Ωτ / m
or, w = (UC w )2 − (UC w )1 in which, U = Ωr
From the steady flow energy equation,
C 22 C12
w = h02 − h01 = h2 − h1 + −
2 2
C 22 C12
or, h 2 − h1 = (UC w )2 − (UC w )1 − +
2 2
8
Lect-31
b U2
U1
r2
r1
Ω
9
Lect-31

The above equation gets transformed to,
U22 U12  V22 V12 
h 2 − h1 = − −  − 

2 2  2 2 
 Ω 2r 2  dV 2
i.e., dh = d  −
 2  2
Since, Tds = dh − dP / ρ
dP  Ω 2r 2  dV 2
= d  − − Tds
ρ  2  2
dP  Ω 2r 2   V2 
For an isentropic flow, = d  − d 
ρ  2   2 
10
Lect-31

• For axial compressors, dr≈0 and the above
equation reduces to dP / ρ = −d(V 2 / 2)
• Thus in an axial compressor rotor, pressure
rise can be obtained only be decelerating
the flow.
• In a centrifugal compressor, the term
d(Ω 2r 2 / 2) > 0, means that pressure rise can
be obtained even without any change in
the relative velocity.
• With no change in relative velocity, these
rotors are not liable to flow separation.
11
Lect-31
• However most centrifugal compressors do

have deceleration and hence are liable to
boundary layer separation.
• Centrifugal compressor rotor is not
essentially limited by separation the way
axial compressor is.
• It is therefore possible to obtain higher per
stage pressure rise from a centrifugal
compressor as compared to axial flow
compressors.
12
Lect-31
Conservation of Rothalpy
• If we were to assume steady, viscous flow
without heat transfer
C12 C 22
h1 + − U1C w1 = h 2 + − U2 C w 2 = I
2 2
• Here, I, is the rotational enthalpy or rothalpy.
• It is now known that rothalpy is conserved for
the flow through the impeller.
• Any change in rothalpy is due to the fluid
friction acting on the stationary shroud (if
considered in the analysis).
13
Lect-31
Impeller
• Impeller draws in the working fluid. It is
the rotating component of the centrifugal
compressor.
• The diverging passages of the impeller
diffuses the flow to a lower relative
velocity and higher static pressure.
• Impellers may be single-sided or double-
sided, shrouded or un-shrouded.
• In the impeller, the working fluid also
experiences centripetal forces due to the
rotation.
14
Lect-31
Impeller
• In principle, there are three possibilities for

a centrifugal compressor rotor.
– Straight radial
– Forward leaning
– Backward leaning
• Forward leaning blades are not used due
inherent dynamic instability.
• Straight and backward leaning blades are
commonly used in modern centrifugal
compressor rotors.
15
Lect-31
Impeller
β2 β2
Ω Ω Ω
U2 U2 U2
V2 V2 V2
C2 β2 C2 C2
β2
Forward leaning blades Straight radial Backward leaning blades
(β2 is negative) (β2 is positive)
16
Lect-31
Inducer
• Inducer is the impeller entrance section

where the tangential motion of the fluid is
changed in the radial direction.
• This may occur with a little or no
acceleration.
• Inducer ensures that the flow enters the
impeller smoothly.
• Without inducers, the rotor operation
would suffer from flow separation and high
noise.
17
Lect-31
Inducer
m m
rt
rm
rh
Inducer Ut Um Uh
βt Ct
V’t
βh
βm
Leading edge velocity triangles
U1
β1 C1
V1
Section m-m
18
Lect-31
Inducer
• It can be seen from the above that

Vt' = V1t cos β1t
Where, V ' denotes the relative velocity at the
inducer outlet.
• It can be seen that V ' < V1 , which indicates
diffusion in the inducer.
• Similarly, we can see that the relative
Mach number from the velocity triangle is,
M1rel = M1 / cos β1t
19
Lect-31
The diffuser
• High impeller speed results in a high
absolute Mach number leaving the impeller.
• This high velocity is reduced (with an
increase in pressure) in a diffuser.
• Diffuser represents the fixed or stationary
part of the compressor.
• The diffuser decelerates the flow exiting the
impeller and thus reduces the absolute
velocity of the working fluid.
• The amount of deceleration depends upon
the efficiency of the diffusion process.
20
Lect-31
The diffuser
• The fluid flows radially outwards from the

impeller, through a vaneless region and then
through a vaned diffuser.
• Both vaned and the vaneless diffusers are
controlled by boundary layer behaviour.
• Pipe and channel type diffusers are used in
aero engines due to their compatibility with
the combustors.
21
Lect-31
The diffuser
Diffuser vanes
Vaneless space
r3
r3>r2>r1
r2
Ω
r1
Impeller
22
Lect-31
The diffuser
Logarithmic spiral
Radial direction
Cr
C
α
CW
Streamlines in a radial diffuser

23
Lect-31
The diffuser
Let us consider an incompressible flow in a vaneless region of

constant axial width.
From continuity , m
 = ρ(2 πrh)Cr = constant.
From conservation of angular momentum,
rC w = constant
∴ C w /Cr = constant = tanα, where α is the angle between the
velocity and the radial direction.
Thus, the velocity is inversely proportion al to radius. This means
that there is diffusion taking place in the vaneless space.
24
Lect-32
In this lecture...
• Centrifugal compressors
• Coriolis acceleration
• Slip factor
• Performance characteristics
• Stall and surge
2
Lect-32
Coriolis acceleration
• We have discussed earlier that pressure

change due to the centrifugal force field is
not a cause of boundary layer separation.
• This can also be explained by the Coriolis
forces that are present in centrifugal
compressor rotors.
• Let us consider a fluid element travelling
radially outward in the passage of a rotor.
• We shall examine the velocity triangles of
this fluid during a time period dt.
3
Lect-32
Ωr
Ω
dCw
Ωdr Vdθ
dC V
C’ C
dθ
Ωr
Ω(r+dr)
4
Lect-32
• The magnitude of the relative velocity is

unchanged, but the particle has suffered
an absolute change of velocity.
dC w = Ωdr + Vdθ
or, dC w = ΩVdt + VΩdt,
Thus, the Coriolis acceleration , aθ = 2ΩV
and it requires a pressure gradient in the tangential
1 ∂P
direction of magnitude, = −2ρΩV
r ∂θ
5
Lect-32
• The existence of the tangential pressure
gradient means that there will be a positive
gradient of V in the tangential direction.
1 dP
=−
d V2 / 2
=−
(
V dV )
ρ rdθ rdθ r dθ
1 dV
Therefore, = 2Ω
r dθ
• This means that there will be a tangential
variation in relative velocity.
6
Lect-32
Cw2
C2
V2
U2
Straight radial blades
7
Lect-32
Slip factor
• Towards the outlet of the impeller, as the
Coriolis pressure gradient disappears, there
will be a difference between Cw2 and U2.
• This difference in the velocities is expressed
as slip factor, σ s = C w 2 / U2
• The slip factor is approximately related to
the number of blades of the impeller.
• For a straight radial blade, the slip factor is
empirically expressed as σ s ≈ 1 − 2 / N, where
N is the number of blades.
8
Lect-32
Slip factor
• As the number of blades increases, the slip
factor also increases and thus the slip lag at
the tip of the impeller reduces.
• The effect of slip is to reduce the magnitude
of swirl velocity and therefore the pressure
ratio.
• The presence of slip means that to deliver
the same pressure ratio, either the impeller
diameter or the rotational must be
increased.
• This in turn may lead to either increase in
frictional losses or stresses on the impeller.
9
Lect-32
Performance characteristics
• The centrifugal compressor performance
characteristics can be derived in the
same way as an axial compressor.
• Performance is evaluated based on the
dependence of pressure ratio and
efficiency on the mass flow at different
operating speeds.
• Centrifugal compressors also suffer from
instability problems like surge and
rotating stall.
10
Lect-32
• The compressor outlet pressure, P02, and
the isentropic efficiency, ηC, depend upon
several physical variables
P02 , η C = f(m
 , P01 , T01 , Ω, γ, R, ν, design, D)
In terms of non - dimensionl ess parameters,
P02 m γRT01 Ω D Ω D 2 
, ηC = f  , , , γ, design 
P01  P01D 2 γRT ν 
 01 
P02 m T01 N 
The above reduces to , ηC = f  , 
P01  P01 T 
 01 
11
Lect-32
Usually, this is further processed in terms of the

standard day pressure and temperature.
P02 m θ N 
, ηC = f  , 
P01  δ θ 
 
T01 P01
Where, θ = and δ =
(T01 )Std. day (P01 )Std. day
(T01 )Std. day = 288.15 K and (P01 )Std. day = 101.325 kPa
12
Lect-32
Surging limit
B
D
Constant speed line
Pressure ratio
A E
Choking limit
Mass flow C
13
Lect-32
Surge line
P02 Locus of points of
P01 maximum efficiency
N
θ
m θ
δ
14
Lect-32
η
N
θ
m θ
δ
15
Lect-32
• There are two limits to the operation of
the compressor.
• Operation between A and B are limited
due to occurrence of surge.
• Surging: sudden drop in delivery pressure
and violent aerodynamic pulsations.
• Operation on the positive slope of the
performance characteristics: unstable
• Surging usually starts to occur in the
diffuser passages.
16
Lect-32
• The pressure ratio or the temperature rise in
a centrifugal compressor also depends upon
the blade shaping.
• There are three possible types of blade
shapes: forward leaning, straight radial and
backward leaning.
• Theoretically, the forward leaning blading
produces higher pressure ratio for a given
flow coefficient.
• However such a blading has inherent dynamic
instability.
• Therefore, straight radial or backward leaning
blades are popularly used.
17
Lect-32
P03 / P01 or,

ΔT0 Forward leaning
2
U 
(γ − 1) 2  T01
 a01  Straight radial
Backward leaning
 or  Cr 2 
 
m U 
 2 
Performance characteristics for different
blade geometries
18
Lect-32
Rotating stall
• Rotating stall might also affect the
compressor performance.
• In this case a stall cell (that might cover
one or more adjacent blades) rotates
within the annulus.
• Full annulus rotating stall may eventually
lead to surge.
• Rotating stall may also lead to
aerodynamically induced vibrations and
fatigue failure of the compressor
components.
19
Lect-32
Rotating stall
Propagation of rotating stall

20
Lect-32
Choking in a compressor stage

• The other limiting aspect of centrifugal
compressors is choking.
• As the mass flow increases, the pressure
decreases, density reduces.
• After a certain point, no further increase in
mass flow will be possible.
• The compressor is then said to have
choked.
• The right hand side of the constant speed
lines together form the choking line.
21
Lect-32

• Choking behaviour for rotating passages is
different from that of stationary passages.
• Inlet:
• Choking takes place when M=1
T 2
=
T0 γ +1
Assu min g an isentropic flow, the choking mass flow rate is
( γ + 1) / 2( γ − 1)
m
  2 
= ρ 0 a0  
A  γ + 1
• Since ρ0, a0 refer to the inlet stagnation
conditions and are constant, the mass flow
rate is also a constant: choking mass flow.
22
Lect-32

• Impeller:
• In rotating passages, the flow conditions
are referred through rothalpy, I.
• During choking, it is the relative velocity, V,
that becomes equal to the speed of sound.
I = h + 12 (V 2 − U2 ) → T01 = T + (γRT / 2c p ) − (U2 / 2c p )
( γ + 1) / 2( γ − 1)
T  2  U2  m
  T 
∴ =   1 + and = ρ 01a01  
T01  γ + 1  
2c p T01  A  T01 
( γ + 1) / 2( γ − 1)
m
  2 + (γ − 1)U2 / a01
2

or, = ρ 01a01  
A  γ + 1 
23
Lect-32
• In an impeller, the choking mass flow is a

function of the rotational speed.
• Therefore, the compressor can, in
principle, handle a higher mass flow with
an increase in speed.
• This also requires that no other
component like the inlet or the diffuser
undergoes choking at this new rotational
speed.
24
Lect-32

• Diffuser:
• The choking mass flow in a diffuser has an
equation similar to that of an inlet:
( γ + 1) / 2( γ − 1)
m
  2 
= ρ 0 a0  
A  γ + 1
• The stagnation conditions at the inlet of
diffuser depend upon the impeller exit
conditions.
• It can be shown that the choking mass
flow is a function of the rotational speed
and therefore can be varied by changing
the rotational speed.
25
Lect 34
Centrifugal Compressors
Design of Centrifugal Compressor

elements – Impellers, Vanes etc.
2
Lect 34
a) Impeller
b) Diffuser
Vanes
c) Vaneless
diffuser
e) Inlet Guide
vanes
f) Volute
3
Lect 34
Other Design Possibilities:

a) Double-sided impeller :
b) Multi-staged compressor
4
Lect 34
Other
important
issues to be
designed
5
Lect 34
The velocity
Diagrams
6
Lect 34
Slip factor
In a real compressor relative velocity vector V2 is at
angle β2 because of non-radial exit from the impeller tip
as the real viscous flow detaches near the tip from the
impeller vane (trailing) surface
Stanitz formula , 0.63.π / N No dependence

σs = 1 -
1 - φ2 . tan
β 2* on backsweep
Cr2
Where φ2 = & N= no. of blades
U2 0.63.π β2 <-450 ; N>8
which, for a radial vane, σs = 1-
N
π
( /N) cosβ2
Stodola Definition σ = 1 − 00 < β2 <-600
1 − (Vr2/U2 ) tanβ2
π /N) cosβ
Wisner’s definition σ = 1 − ( β2 >-450 ; N>20
2
7
Lect 34
Forward Small High High speed

Curved Volume Pressure High noise ,
Vanes ratio Low
Efficiency
Backward Large Low to High High
curved Volume and Pr Ratio Efficiency,
Vanes size Low Noise
Radial Vanes Medium Medium to Good
Volume and High Pr ratio Efficiency
Size
Radial Vaned CCs have been used in A/C engines for 50
years. Now, well designed backward curved vaned CCs
are increasingly being used for higher efficiency.
8
Lect 34
In highly
forward and
highly backward
curved
(β2 >-600 )
impellers
slip factor
looses its
meaning
9
Lect 34
At the compr. entry face

Ca1 U1 = ω.reye where reye varies from
tanβ1 =
U1 the root to the tip of the eye
Thus for a high speed compressor (or large sized)

β1 shall vary hugely from root to tip of the eye.
To be
Under off-design operations,
β ( - β1
ir = *
) decided by
at any radius, incidence, ir 1 r
designer
High positive incidence i (≥ +50) may precipitate

early flow separation inside the impeller vane
passage, even near the eye, specially if high
diffusion (i.e. high adverse pressure gradient) is
being attempted inside the impeller vanes.
10
Lect 34
At the exit plane of the impeller, the exiting

flow deviates from the trailing edge and lag
behind in rotational mode. This is often
referred to as the lag or deviation angle.
which is an average at the
δ-=
β * passage exit, and β2* is the
av 2 2 impeller vane exit angle set
by design
Diffusion Limit :
An upper limit of realistic diffusion limit V2/V1≈ 0.6

In rotating diffuser V2/V1 < 0.6
In Impeller design, ρ1 A1 / ρ2 A2 > 2.0
11
Lect 34
Design of the vaneless space

• Vaneless space
is often used to
decelerate
impeller exit
flow from
supersonic to
subsonic speed
• A completely
vaneless diffuser
is lighter , has
broader mass
flow operating
range but has a
lower efficiency
12
Lect 34
Reduction in deviation angle at the impeller exit

under off-design operating conditions is to be
designed in to the impeller and the vane designs.
Vaneless diffuser
Backward curved vanes +

splitter
13
Lect 34
The general relationship for Compressor Pressure ratio is

given by
γ

03 (σ . Uη0c-.(
Uγ -.C 2
)  γ -1
π0C =
pΨ.
= 1 +
1). s
2
2 1 w1

p01  a 01 
• Theoretical energy density (Hth) transfer is highest with
forward curved vanes, in which most of the energy would be
available in kinetic form, Hdyn at the impeller exit.
• While a radial impeller gives almost 50-50 split of static
(Hstatic ) and dynamic heads (Hdyn) at the impeller exit, the
backward curved vanes give high static pressure development
in the impeller.
• Pre-swirl ( α1 > 0 ) reduces the work done by compressor
14
Lect 34
The theoretically
obtained points to
the right of b are
considered
choked, i.e. the
compressor
cannot process
greater mass
flows. The
compressor is
said to go in to
stall at m a , this
happens when
high pressure rise
is attempted at
low mass flow With Mass Flow Control only
15
Lect 34
• In aircraft engines,
rotating speed is
variable during
actual running.
•Thus the zone of
operation is bounded
between the points
a,b,c and d .
•The ηmin lines and
the speed lines, nmax
and nmin, define the
boundaries (shaded
area) of operation.
With Speed Control and Flow Control
16
Lect 34
• If more control variables are available it may

be possible to extend the zone of operation of
the compressor. All possible means of extending
these boundaries further are being explored.
• Variable geometry (stagger) Inlet and exit
(diffuser) guide vanes to be explored
17
Lect 34
Centrifugal Compressor characteristics with multiple controls

18
Lect 34
Stall and Surge control

•Surging tends to originate
in diffuser passages where
frictional effects of the vane
retard the flow.
• Flow reversal may vary
from one blade passage to
the next.
• The surging is reduced by
making the number of
diffuser vanes an odd
number mis-match of the
impeller vanes. In this way
pressure fluctuations are
more likely to be evened
out over the annular
vaneless circumference.
19
Lect 34
Losses : Ideal and

Real Characteristics
• Most of the losses

are still found by
rigorous rig test.
• CFD gives good 1st
cut estimation of
loss analysis
20
Lect 34
• Efficiency, η is borne out of loss analysis,

whereas work done factor Ψ, is borne out
of flow analysis as shown in the last slide.
A value of σs is also arrived at by either
CFD analysis or a first cut value by simple
flow analysis.
• The flow parameters need averaging both
at the compr. inlet (eye) along the vane
height as well as the impeller exit along
the depth of the vane.
21
Lect-2
In this lecture...
• Axial flow compressors and fans

– Thermodynamics of compression
– P-v and T-s diagrams of compressors
– Thermodynamics of compression
process
– Multi-stage compression
– Basic operation of axial compressors/fans
– Velocity triangles
– Work and compression
2
Lect-2
Introduction
• Simplified aero-thermodynamic analysis

• Optimised cycle design to precede the
detailed component design
• Prediction of work requirements
• Efficiency of the compressor
• Enables faster design modifications
3
Lect-2
Thermodynamics of compression
(i) Adiabatic (process 1-2/) , Pvγ=c

(ii) Isothermal process (1-2//), Pv=c
∞
(iii) Isochoric (Process 1-2///), Pv =c
4
Lect-2
Thermodynamics of compressors
i) Isentropic process (1-2/)

ii) Polytropic process (1-2)
iii) Isothermal process (1-2//)
iv) Isochoric Process (1-2///)
5
Lect-2
• The compression
process is usually
expressed in H-s or T-
s diagrams.
• The ideal compression
process is assumed to
be isentropic.
• Deviation from this is
expressed as
isentropic efficiency.
6
Lect-2
X1 , X2 are the
losses in the rotor
and the stator
respectively
Compression in terms
of static parameters
7
Lect-2
Compression in terms
of total parameters
8
Lect-2
Thermodynamics of multi-stage compressors
• The flow at the rotor

exit with high kinetic
energy is still to be
converted to static
pressure through
diffusion.
Averaged T-s • The exit kinetic energy
characteristics
of a compressor is of
the same order as the
Rotor isentropic, stator isothermal entry kinetic energy
Rotor polytropic, stator isothermal
and the entire work
input is expected to be
converted to pressure.
9
Lect-2
Basic operation of axial compressors

• Axial flow compressors usually consists of a
series of stages.
• Each stage comprises of a row of rotor
blades followed by a row of stator blades.
• The working fluid is initially accelerated by
the rotor blades and then decelerated in the
stator passages.
• In the stator, the kinetic energy transferred
in the rotor is converted to static pressure.
• This process is repeated in several stages to
yield the necessary overall pressure ratio.
10
Lect-2

• The compression process consists of a series of
diffusions.
• This occurs both in the rotor as well as the
stator.
• Due to motion of the rotor blades two distinct
velocity components: absolute and relative
velocities in the rotor.
• The absolute velocity of the fluid is increased in
the rotor, whereas the relative velocity is
decreased, leading to diffusion.
• Per stage pressure ratio is limited because a
compressor operates in an adverse pressure
gradient environment.
11
Lect-2
• Turbines on the other hand operate under

favourable pressure gradients.
• Several stages of an axial compressor can be
driven by a single turbine stage.
• Careful design of the compressor blading is
essential to minimize losses as well as to
ensure stable operation.
• Some compressors also have inlet Guide Vanes
(IGV) that permit the flow entering the first
stage to vary under off-design conditions.
12
Lect-2
Velocity triangles
• Elementary analysis of axial compressors begins
with velocity triangles.
• The analysis will be carried out at the mean height
of the blade, where the peripheral velocity or the
blade speed is, U.
• The absolute component of velocity will be
denoted by, C and the relative component by, V.
• The axial velocity (absolute) will be denoted by Ca
and the tangential components will be denoted by
subscript w (for eg, Cw or Vw)
• α denotes the angle between the absolute velocity
with the axial direction and β the corresponding
angle for the relative velocity.
13
Lect-2
Velocity triangles
1 2 3
V2
β2
V1
β1 α2
C2
V1 α3
V2 C3
U
U
C1 C2
Rotor Stator
  
C = U +V
14
Lect-2
Velocity triangles
V2
β2 Vw2
α2
V1 Vw1
ΔCw
U
β1 C2 Cw2
α1
C1 Cw1
Ca
15
Lect-2
Property changes across a stage
Total enthalpy
h01 h02 h03
Absolute velocity
C1 C2 C3
Static pressure
P1 P2 P3
16
Lect-2
Work and compression

• Assuming Ca=Ca1=Ca2, from the velocity
triangles, we can see that
U U
= tan α1 + tan β1 and = tan α 2 + tan β 2
Ca Ca
• By considering the change in angular
momentum of the air passing through the
rotor, work done per unit mass flow is
w = U(Cw 2 − Cw1 ), where Cw1 and Cw 2are the tangential
components of the fluid velocity before and after the
rotor, respectively.
17
Lect-2

The above equation can also be written as,
w = UC a(tan α 2 − tan α1 )
Since, (tan α 2 − tan α1 ) = (tan β1 − tan β 2 )
∴ w = UC a(tan β1 − tan β 2 )
In other words, w = UΔCw
• The input energy will reveal itself in the
form of rise in stagnation temperature of
the air.
• The work done as given above will also be
equal to the change in stagnation enthalpy
across the stage.
18
Lect-2

h02 − h01 = UΔCw
UΔCw ΔT0 UΔCw
T02 − T01 = ⇒ =
cp T01 cp T01
Since the flow is adiabatic and no work is done as
the fluid passes through the stator, T03 = T02
Let us define stage efficiency , ηst , as
h03s − h01
ηst =
h03 − h01
This can be expressed as
T03s ΔT0
= 1 + ηst
T01 T01
19
Lect-2
In the above equation, ΔT0 = T03 − T01

In terms of pressure ratio,
γ / ( γ −1 )
P03  ΔT0 
= 1 + ηst 
P01  T01 
This can be combined with the earlier equation
to give,
γ / ( γ −1 )
P03  UΔCw 
= 1 + ηst 
P01  cp T01 
20
Lect-2

• From the above equation that relates the
per stage temperature rise to the pressure
ratio, it can be seen that to obtain a high
temperature ratio for a given overall
pressure ratio (for minimizing number of
stages),
– High blade speed: limited by blades stresses
– High axial velocity, high fluid deflection
(β1-β2): Aerodynamic considerations and
adverse pressure gradients limit the above.
21
Lect-2

• From the above equation that relates the
per stage temperature rise to the pressure
ratio, it can be seen that to obtain a high
temperature ratio for a given overall
pressure ratio (for minimizing number of
stages),
– High blade speed: limited by blades stresses
– High axial velocity, high fluid deflection
(β1-β2): Aerodynamic considerations and
adverse pressure gradients limit the above.
22
Lect-19
In this lecture...
• Performance characteristics of
centrifugal compressors
• Surging and choking
• Tutorial on centrifugal compressors
2
Lect-19
• The centrifugal compressor performance
characteristics can be derived in the
same way as an axial compressor.
• Performance is evaluated based on the
dependence of pressure ratio and
efficiency on the mass flow at different
operating speeds.
• Centrifugal compressors also suffer from
instability problems like surge and
rotating stall.
3
Lect-19
P02 ,ηC = f (m , P01 , T01 , Ω, γ , R,ν , design, D)
In terms of non - dimensionless parameters,
P02  m γRT01 ΩD ΩD 2 
,ηC = f  , , , γ , design 
P01  P01 D 2 γ RT ν 
 01 
P02  m T01 N 
The above reduces to ,η C = f  , 
P01  P01 T 
 01 
4
Lect-19
P02  m θ N 
,ηC = f  , 

P01  δ θ 
T01 P01
Where,θ = and δ =
5
Lect-19
Surging limit
B
D
Constant speed line
Pressure ratio
A E
Choking limit
Mass flow C
6
Lect-19
• There are two limits to the operation of
the compressor.
• Operation between A and B are limited
due to occurrence of surge.
• Surging: sudden drop in delivery pressure
and violent aerodynamic pulsations.
• Operation on the positive slope of the
performance characteristics: unstable
• Surging usually starts to occur in the
diffuser passages.
7
Lect-19
• Rotating stall might also affect the
compressor performance.
• In this case a stall cell (that might cover
one or more adjacent blades) rotates
within the annulus.
• Full annulus rotating stall may eventually
lead to surge.
• Rotating stall may also lead to
aerodynamically induced vibrations and
fatigue failure of the compressor
components.
8
Lect-19
Propagation of rotating stall

9
Lect-19
• The other limiting aspect of centrifugal
compressors is choking.
• As the mass flow increases, the pressure
decreases, density reduces.
• After a certain point, no further increase in
mass flow will be possible.
• The compressor is then said to have
choked.
• The right hand side of the constant speed
lines together form the choking line.
10
Lect-19
Surge line
P02 Locus of points of
P01 maximum efficiency
N
θ
m θ
δ
11
Lect-19
η
N
θ
m θ
δ
12
Lect-19
Problem # 1
• A centrifugal compressor has a pressure
ratio of 4:1 with an isentropic efficiency of
80% when running at 15000 rpm and
inducing air at 293 K. Curved vanes at the
inlet give the air a prewhirl of 25o to the
axial direction at all radii. The tip diameter
of the eye of the impeller is 250 mm. The
absolute velocity at inlet is 150 m/s and
the impeller diameter is 600 mm.
Calculate the slip factor.
13
Lect-19
Problem # 1
C1 25 V1
Cw1 U1
14
Lect-19
Solution: Problem # 1
Exit stagnation temperature is
T02 = T01 (π c ) = 293(4)
( γ −1) / γ (1.4 −1) / 1.4
= 435.56 K
Therefore the isentropic temperature rise,
∆T0 s = 435.56 − 293 = 142.56 K
The actual temperature rise, ∆T0 = ∆T0 s / η c
∆T0 = 178.2 K
Work done per unit mass is, w = cP ΔT0
w = 1.005 ×178.2 = 179 kJ/kg
15
Lect-19
Peripheral velocity at the tip of the eye,

U1 = πdN / 60 = π × 0.25 ×15000 / 60 = 196.25m / s
Cw1 = C1 sin 25 = 63.4m / s
Peripheral velocity at the tip of the impeller,
U 2 = πDN / 60 = π × 0.60 ×15000 / 60 = 471.2m / s
16
Lect-19
We know that power input is, w = U 2Cw 2 − U1Cw1

179 × 103 = 471.24 × Cw 2 − 196.35 × 63.4
or , Cw 2 = 406.27 m / s
Therefore, the slip factor is,
σ s = Cw 2 / U 2 = 0.862
17
Lect-19
Problem # 2
• At the inlet of a centrifugal compressor
eye, the relative Mach number is to be
limited to 0.97. The hub-tip radius ratio of
the inducer is 0.4. The eye tip diameter is
20 cm. If the inlet velocity is axial,
determine, (a) the maximum mass flow
rate for a rotational speed of 29160 rpm,
(b) the blade angle at the inducer tip for
this mass flow. The inlet conditions can be
taken as 101.3 kPa and 288 K.
18
Lect-19
Problem # 2
U1
β1 C1
rt V1
rm
rh
19
Lect-19
The rotational speed at the inducer tip is
U1 = πdN / 60 = π × 0.2 × 29160 / 60 = 305.36m / s
From the velocity traingle, we can see that
V1 C12 + U12
M 1rel = =
γRT1 γRT1
T1 = T01 − C12 / 2cP = 288 − C12 / 2010
C12 + U12
M 1rel =
γR(288 − C12 / 2010)
C 2
+ 305 . 63 2
0.97 2 = 1
115718.4 − 0.2C12
Simplifying, C1 = 114.62m / s
20
Lect-19
T1 = T01 − C12 / 2cP = 288 − C12 / 2010 = 281.464 K

γ /( γ −1)
P01  T01 
=  
P1  T1 
Substituti ng, P1 = 93.48kPa
∴ ρ1 = P1 / RT1 = 1.157 kg / m 3
π
Annulus area at the inlet, A1 = d 2 (1 − rh / rt )
4
A1 = 0.0264m 2
21
Lect-19
Since the flow is axial,

Ca1 = C1
∴ m = ρ1 A1C1 = 1.157 × 0.0264 × 114.62 = 3.5kg / s
The blade inlet angle at the tip is
tan β1 = C1 / U1
∴ β1 = 20.57 o
22
Lect-19
Exercise Problem # 1
• The design mass flow rate of a centrifugal
compressor is 7.5 kg/s with inlet stagnation
temperature of 300 K and pressure of 100
kPa. The compressor has straight radial
blades at the outlet. The blade angle at the
inducer inlet tip is 50o and the inlet hub-tip
ratio is 0.5. The impeller is designed to have
a relative Mach number of 0.9 at the inducer
inlet tip. If the tip speed is 450 m/s,
determine (a) the air density at inducer inlet,
(b) the inducer inlet diameter, (c) the rotor
rpm (d) the impeller outlet diameter.
• Ans: 0.988 kg/m3, 0.258 m, 17100 rpm,
0.502 m
23
Lect-19
• A centrifugal compressor runs at 10000
rpm and delivers 600 m3/min of air at a
pressure of 4:1. The isentropic efficiency
of the compressor is 0.82. The outer
radius of the impeller is twice the inner
radius. The axial velocity is 60m/s. If the
ambient conditions are 1 bar and 293 K,
determine (a) the impeller diameter at
inlet and outlet (b) the power input (c) the
impeller and diffuser angles at inlet.
• Ans: 0.92, 0.461, 2044 kW, 13.9o, 7.1o
24
Lect-19
• 30 kg of air per second is compressed in a
centrifugal compressor at a rotational speed
of 15000 rpm. The air enters the compressor
axially. The compressor has a tip radius of 30
cm. The air leaves the tip with a relative
velocity of 100 m/s at an angle of 80o.
Assuming an inlet stagnation pressure and
temperature of 1 bar and 300 K, respectively,
find (a) the torque required to drive the
compressor, (b) the power required (c) the
compressor delivery pressure
• Ans: 4085 Nm, 6.417 MW, 6.531 bar
25
Lect-11
In this lecture...
• Performance characteristics of axial

flow compressors
• Single stage characteristics
• Multi-stage characteristics
2
Lect-11
Single stage performance characteristics
• Let us consider a typical axial compressor

stage comprising of a set of rotor blades
followed by a set of stator blades.
V1
V2
U
U
C1 C2
α3
C3
Rotor Stator
3
Lect-11
V2
β2 Vw2
α2
V1 Vw1
ΔCw
U
β1 C2 Cw2
α1
C1 Cw1
Ca
4
Lect-11
• From the above velocity triangles,
Cw 2 = U − Ca tan β 2 and Cw1 = Ca tan α

Since, Δh0 = U∆Cw
Δh0 = U [U − Ca (tan α1 + tan β 2 )]
∆Cw Δh0
or , = 2 = 1−
Ca
(tan α1 + tan β 2 )
U U U
5
Lect-11

• Change in the design mass flow rate
affects Ca, change in rotor speed affects U.
• Change of either Ca or U changes the inlet
angle β1 at which the flow approaches the
rotor.
• The above equation shows that the blade
performance depends upon the ratio Ca/U.
The stage performance is a function of the loading coefficien t,
flow coefficien t and the efficiency .
Thus,
Stage performance = f(ψ, φ, η)
6
Lect-11
∆Cw Δh0
= 2 = 1 − a (tan α1 + tan β 2 )
∆h0 ∆Cw C
=
U2 U U U U
1
(tan α1 + tan β 2 )design
Measured
Stage loading
 Ca   Ca 
   
 U  design U 
1.0
η st
Measured
Stage efficiency
 Ca   Ca 
   
 U  design U 
7
Lect-11
V1 V1 V1
β1 β1 β1
V1 V1
V1
U β1 U β1 U
β1 α1
α1 α1 C1
C1 C1
Ca Ca
Ca
Design condition : Off - design condition : Off - design condition :
Normal operation Positive incidence flow separation Negative incidence flow separation
 Ca   Ca   Ca   Ca   Ca   Ca 
 =   <   > 
 U   U  design  U   U  design  U   U  design
8
Lect-11
Multi-stage performance characteristics
• Let us know consider a multi-stage

compressor. Inlet station is denoted by 1
and exit of the compressor by 2.
• Therefore the overall pressure ratio of
the compressor is P02/P01.
9
Lect-11

P02 ,ηC = f (m , P01 , T01 , Ω, γ , R,ν , design, D)
In terms of non - dimensionless parameters,
P02  m γRT01 ΩD ΩD 2 
,η C = f  , , , γ , design 
P01  P01 D 2 γRT ν 
 01 
For a given design, we can assume that γ and ν do not
affect the performance significantly. Also, D and R
are fixed. Therefore the above reduces to
P02  m T01 N 
,η C = f  , 
P01  P01 T 
 01 
10
Lect-11

P02  m θ N 
,ηC = f  , 

P01  δ θ 
T01 P01
Where,θ = and δ =
11
Lect-11
P02
P01
m θ
δ
12
Lect-11
P02
P01
m θ
δ
13
Lect-11

• In a multi-stage compressor, a small departure
from the design point at the first stage causes
progressively increasing departure from design
conditions from the first stage onwards.
• Thus, a small reduction in (ca/U)design at the first
stage could lead to positive incidence separation
at the last stage.
• Similarly, a small increase in (ca/U)design could
lead to negative incidence separation in the last
stage.
• The most extreme mismatching of the front and
rear stages occur during starting.
14
Lect-11

Design velocity triangles
V2 V2
β2 β2
α2
α2 V1
U U
V1 C2 β1 C2
β1 α1
α1
C1 C1
Ca Ca
First stages Last stages
15
Lect-11
• Decreased Ca with α1 and β2 constant,

results in increased α2 and β1 or
increased loading on both rotor and
stator blades.
• In the case of increased Ca, it results in
the opposite effect.
• Designers use several solutions to allow
compressors to self-start: use of bleed
valves allowing some of the incoming air
to escape, variable IGVs, multi-spooling.
16
Lect-11
• Axial compressors suffer from two

possible modes of unstable operation
– Rotating stall: non-axisymmetric, aperiodic
– Surge: axisymmetric, periodic
• Rotating stall: progression around the blade
annulus of a stall pattern, in which one or more
adjacent blade passages are instantaneously
stalled, then are cleared for unstalled flow as
the stall cell progresses.
• Rotating stall causes alternate loading and
unloading of the blades: fatigue failure.
17
Lect-11
P02
P01
m θ
δ
18
Lect-11
∆P0
ρU 2
Stage characteristics
Throttle characteristics
Ca
U
19

17AE11-Propulsion - I Unit 3 - COMPRESSORS

Uploaded by

Copyright:

Available Formats

17AE11-Propulsion - I Unit 3 - COMPRESSORS

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

17AE11-Propulsion - I Unit 3 - COMPRESSORS

Uploaded by

Copyright:

Available Formats

17AE11-Propulsion – I

UNIT – III COMPRESSORS

Theory of Axial Flow Compressor – Velocity Triangles – Degree of

• Dynamic compressors work by converting velocity to pressure in a

CFM stands for Cubic Feet per Minute

Ohain engine (1939) Whittle engine (1941).

Centrifugal compressors stage

Typical centrifugal compressor rotors

Centrifugal compressors stage

Schematic of a typical centrifugal compressor

Centrifugal compressors stage

Centrifugal compressors stage

Centrifugal compressors stage

Centrifugal compressors stage

Centrifugal compressors stage

Centrifugal compressors stage

• However most centrifugal compressors do

• In principle, there are three possibilities for

• Inducer is the impeller entrance section

• It can be seen from the above that

• The fluid flows radially outwards from the

Streamlines in a radial diffuser

Let us consider an incompressible flow in a vaneless region of

• We have discussed earlier that pressure

• The magnitude of the relative velocity is

Straight radial blades

Usually, this is further processed in terms of the

P03 / P01 or,

Propagation of rotating stall

Choking in a compressor stage

Choking in a compressor stage

Choking in a compressor stage

Choking in a compressor stage

• In an impeller, the choking mass flow is a

Choking in a compressor stage

Design of Centrifugal Compressor

Other Design Possibilities:

Stanitz formula , 0.63.π / N No dependence

Forward Small High High speed

At the compr. entry face

Thus for a high speed compressor (or large sized)

High positive incidence i (≥ +50) may precipitate

At the exit plane of the impeller, the exiting

An upper limit of realistic diffusion limit V2/V1≈ 0.6

Design of the vaneless space

Reduction in deviation angle at the impeller exit

Backward curved vanes +

The general relationship for Compressor Pressure ratio is

• If more control variables are available it may

Centrifugal Compressor characteristics with multiple controls

Stall and Surge control

Losses : Ideal and

• Most of the losses

• Efficiency, η is borne out of loss analysis,

• Axial flow compressors and fans

• Simplified aero-thermodynamic analysis

(i) Adiabatic (process 1-2/) , Pvγ=c

i) Isentropic process (1-2/)

Thermodynamics of multi-stage compressors