University of Tehran, Faculty of Engineering

Radial Compressors
Advantages
• Suitable for handling a small
volume flows
• Shorter length than an
equivalent axial compressor
• Better resistance to FOD
• Less susceptibility to loss of
performance by built-up of
deposits on the blade surfaces
(more robust)
• The ability to operate over a
range of mass flow at a
particular speed
• Easier to manufacture
University of Tehran, Faculty of Engineering
 University of Tehran, Faculty of Engineering
Radial Compressors
 University of Tehran, Faculty of Engineering
Radial Compressors
• Impeller: work is
transferred to
accelerate flow
and increase
pressure
• Diffuser: recover
high speed
generated in
impeller as
pressure
University of Tehran, Faculty of Engineering
University of Tehran, Faculty of Engineering
 University of Tehran, Faculty of Engineering

Shrouds
• Removes losses in clearance.
• Not used in gas turbines
– Add additional mass
– Unacceptable for high
rotational speed where high
stresses are produced
 University of Tehran, Faculty of Engineering

Radial compressor
operation
• Typical design takes 50 % of
increase in static pressure in
diffuser
• Euler equation governs
performance:

Rate of change of angular momentum 

C w 2 r2  C w1r1 

prewhirl term

Theoretica l work   C w 2 r2  C w1r1  

for cases without prewhirl 
  C w 2 r2  C w 2U 2
 University of Tehran, Faculty of Engineering

Slip factor

Due to inertia of flow Cw2 < U:

Theoretical work   U 2
2

Stanitz formula for estimating σ

Cw2 0.63
 1
U n
n  number of impeller vanes
 University of Tehran, Faculty of Engineering

Power input factor - 

• Converts energy to heat => additional loss =>

Work done  U 2

and
U 2

T03  T01  ,
cp
1.035    1.04
 University of Tehran, Faculty of Engineering

Overall pressure rise:


 /( 1)  /( 1)
p03  T03   c (T03  T01)     U 2
 1
    1    1 c


p01  T01   T01   cpT01 

• p03 is here used to denote the pressure at

compressor exit. p02 is reserved for the
stagnation pressure between the impeller
and the diffuser vanes.
 University of Tehran, Faculty of Engineering

Example 4.1
• ψ =1.04, σ = 0.90
• N = 290.0 rev/s,
• D = 0.5 m
• Deye,tip = 0.3 m, Deye,root = 0.15 m
• m = 9.0 kg/s
• T01 = 295 K
• P01 = 1.1 bar
• ηc = 0.78
• Compute pressure ratio
and power required
 University of Tehran, Faculty of Engineering

Solution
Impeller tip speed is: U    0.5  290  455.5 m / s
Temperature equivalent of the work done is:

U 2 1.04  0.9  455.52

T03  T01    193 K
cp 1005
 /( 1)  /( 1)
p03  T03   c (T03  T01) 
3.5
 0.78193
    1    1    4.23
p01  T01   T01   295 
Power required  m cp (T03  T01)  9 1.0051931746 kW
• Calculate vanes inlet angle
To find the inlet angle of the vanes it is necessary to determine
the inlet velocity C C 1 a1
 University of Tehran, Faculty of Engineering

Solution
Ca1 must satisfy the continuity equation i.e. m  1 A1Ca1
A trial and error process is required.
 (0.32  0.152 )
Annulus area of impeller eye, A1  2
 0.053 m
4
Based on stagnation conditions:
p01 110 m 9
1    1.30 kg / m , Ca1 
3
  131m / s
RT01 0.287 295 1 A1 1.30 0.053
Since Ca1  C1 , the equivalent dynamic temperature is:
2 2
C1 1312 C1
  8.5 K, T1  T01   295 8.5  286.5 K
2cp 2 1005 2cp
p01 1.1 p1 99.2
p1    0.992bar,     1.21 kg / m3

T01 / T1  /( 1) 295/ 286.53.5 1

RT1 0.287 286.5
 University of Tehran, Faculty of Engineering

Solution
m 9
Check on Ca1 Ca1    140 m / s
1 A1 1.21 0.053

Final Trial: Try Ca1  C1  145 m / s

2 2
C1 1452 C1
  10.5 K, T1  T01   29510.5  286.6 K
2cp 2 1005 2cp
p01 1.1 p1 96.8
p1    0.968 bar,     1.185 kg / m3

T01 / T1  /( 1) 295/ 284.53.5 1

RT1 0.287 284.5

m 9
Check on Ca1 Ca1    143 m / s
1 A1 1.185 0.053
 University of Tehran, Faculty of Engineering
Solution
Peripheral speed at the impeller eye tip radius
   0.3  290  273m/ s
Peripheral speed at the impeller eye root radius
   0.15  290  136.5 m / s

143
 at root  tan1
 46.33
136.5
1 143
 at tip  tan  27.65
273

The shape of the impeller between eye and tip is a matter of trial
and error. The aim is to avoid flow separation.
• Calculate depth of the impeller
University of Tehran, Faculty of Engineering
 University of Tehran, Faculty of Engineering
Solution
To calculate the required depth of the impeller we assume that the
1) Radial component of velocity is equal to the axial velocity at
inlet to the eye and 2) Half the total loss occurs in the impeller.
Cr 2  Ca1  143 m / s Cw 2   U  0.9  455.5  410 m / s
Cr 2  Cw 2 1432  4102
2 2 2
C2 0.5(1  c )  0.11
   93.8 K
2c p 2c p 2010
 /(  1)
p02  c (T02  T01 ) 
3.5
 0.89  193 
 1    1    1.582
3.5

p01 2  T01   295 

C
T2  T02  2  488  93.8  394.2 K , T02  T03  193  295  488 K
2c p
 /(  1)
p2  T2 
3 .5
p2  394.2 
  ,   , ( p2 / p01 )  ( p2 / p02 )( p02 / p01 )
p02  T02  p02  488 
394.2 3.5
( p2 / p01 )  (1.58  )  2.35, p2  2.35  1.1  2.58 bar
488
 University of Tehran, Faculty of Engineering
Solution
p2 2.58100
2    2.28 kg / m3
RT2 0.287 394.2
The required area of cross-section of flow in radial direction at the
impeller tip is: m 9
A   0.0276 m 2
 2 Cr 2 2.43  143
0.0276
Hence the depth of impeller channel is   0.0176 m  1.76 cm
  0.5
Backswept Curved Vanes are used in
for high-performance compressors

• Higher pressure ratio

• Higher efficiency
• Wider operating range of air-flow
University of Tehran, Faculty of Engineering
 University of Tehran, Faculty of Engineering
The Diffuser

• An efficient combustion system is eased if the velocity of the air entering the
combustion chamber is as low as possible.
• It is necessary, therefore, to design the diffuser so that only a small part of
the stagnation temperature at the compressor outlet corresponds to kinetic
energy. Usually the velocity of the air at the compressor outlet is in the
region of 90m/s.
• It is much more difficult to arrange an efficient deceleration of flow than it is
to obtain an efficient acceleration. If the divergence is to rapid, this may
result in the formation of eddies formation with consequent transfer of some
kinetic energy into internal energy and a reduction in useful pressure rise.
• A small angle of divergence, however, implies a long diffuser and a high
value of skin friction loss. Experiments have shown the optimum included
angle of about 7 degrees, although angles of up twice this value can be
used with diffusers of low length/width (or radius) ratio without incurring a
serious increase in stagnation pressure loss.
 University of Tehran, Faculty of Engineering
The Diffuser
In order to control the flow of air effectively and carry out the
diffusion process in as short a length as possible, the air leaving
the impeller is divided into a number of separate streams by fixed
diffuser vanes. The angle of the diffuser vanes at the leading
edge must be designed to suit the direction of the absolute
velocity.
 University of Tehran, Faculty of Engineering

Vaneless Space

The picture can't be display ed.

To find the correct inlet angle for the diffuser vanes, the flow in the vaneless
space must be considered. No further energy is supplied to the air after it
leaves the impeller tip, so that, neglecting the effect of friction, the angular
momentum r C w must be constant. Hence C w decreases from impeller tip to
diffuser vane ideally in inverse proportional to the radius. The radial velocity C r
will also decrease from the impeller tip to diffuser vane, in accordance with the
equation of continuity. If both C w and C r decrease, then the resultant velocity C
will decrease from the impeller tip, and some diffusion evidently takes place in
the vaneless space. The consequence increase in density means that C r will
not decrease in inverse proportion to the radius as does C w and the way in
which C r varies must be found from the continuity equation.
 University of Tehran, Faculty of Engineering

Example 4.2
Consider the design of a diffuser for the compressor dealt with in
the previous example. The following additional data will be
assumed:
Radial width of vaneless space 5 cm
Approximate mean radius of diffuser throat 0.33 m
Depth of diffuser passage 1.76 cm
Number of diffuser vanes 12
Requirements:
●The inlet angle of the diffuser
vanes
● The throat width of the
diffuser passages
 University of Tehran, Faculty of Engineering
Solution
For simplicity it will be assumed that the additional friction loss in the short
distance between impeller tip and diffuser throat is small.
a) Conditions at the radius of the diffuser vane leading edges

r  0.25  0.05 r C w  const  C w 2  r1C w1 / r2  0.25  410 / 0.3  342 m / s

The radial component of velocity can be found by trial and error. The final trial
is given here. Try C r 2  97 m / s
2 2 2
C2 Cr 2  Cw 2 3.42 2  0.97 2
   62.9 K
2c p 2c p 0.201
Ignoring any additional loss between the impeller tip and the diffuser vane
leading at 0.3 m radius, the stagnation pressure will be that calculated for the
impeller tip, namely it will be that given by p 02 / p 01  1.582 3.5
425.1 3.5
2
T2  To 2  C 2 / 2 c p  488  62 .9  425 .1 K p 2 / p 02  ( )
Proceeding as before we have:
488
(p 2 / p o1 )  (p 2 / p o 2 )(p o 2 / p o1 )
p2 425.1 3.5
 (1.582  )  3.07, p 2  3.07 1.1  3.38 bar
p o1 488
 University of Tehran, Faculty of Engineering
Solution
p2 3.38 100
2    2.77 kg / m 3
RT2 0.287  425.1
Area of cross-section of flow in radial direction Area r  2  0.3  0.0176  0.0332 m
2

m 9
Check on C r 2 Cr    97.9 m / s
2
2A r 2 2.77  0.0332
Taking C r 2  97.7 m / s , the angle of the diffuser vane leading edge for zero
incidence should be 1 C r 2 1 97.9
tan  tan  16
Cw 2 342
b) The throat width of the diffuser channels may be found by a similar
calculation for the flow at the assumed throat radius of 0.33 m
C w 2  r1 C w1 / r2  0.25  410 / 0.33  311 m / s Try C r 2  83 m / s
2
C2
2 2
C  Cw 2
2
3.11  0.83
2 2 C2
 r2  T
 51.5 K 2  To2   488  51 .5  436 .5 K
2c p 2c p 0.201 2c p
p2 436.5 3.5 po2
( )  1.5823.5 ( p 2 / p o1 )  ( p 2 / p o 2 )(p o 2 / p o1 )
po2 488 p o1
p2 436.5 3.5 p 2  3.37 1.1  3.71 bar
 (1.582  )  3.37 p 3.71100
2  2   2.96 kg / m 3
p o1 488 RT2 0.287  436.5
 University of Tehran, Faculty of Engineering

Solution
At the first approximation we may neglect the thickness of the diffuser
vanes, so that the area of flow in the radial direction
Check on C r 2 Area r  2 π  0. 33  0 .0176  0 . 0365 m 2

m 9
Cr 2    83.3 m / s
 2 A r 2 2.96  0.0365
Direction of the flow
Cr 2 83.3
tan 1  tan 1  15
Cw 2 342
The total throat area can be found from the fact that:
A r 2C r 2
  2 A r 2Cr 2  2 A 2C2  A 2 
m  0.0365 sin 15  0.0094 m 2
C2
With 12 diffuser vanes, the width of the throat in each passage of depth
0.0176 m is therefore
A2 0.0094
width    0.044 m or 4.40 cm
12 depth 12  0.0176
 University of Tehran, Faculty of Engineering

Compressibility Effects
When an effort is made to obtain the maximum possible mass flow from the
smallest possible compressor, as is done most particularly in the design of
aircraft gas turbines, the air speeds are very high. It is of the utmost that the
Mach numbers at certain points in the flow do not exceed the value beyond
which losses increase rapidly due to formation of the shock waves. The critical
Mach number is usually less than unity when calculated on the basis of the
mean velocity of the fluid relative to the boundary. As a general role, unless
actual tests indicate otherwise, the Mach numbers are restricted to about 0.8.
At the inlet, the air is deflected through a certain
angle before it passes into the radial channels on the
impeller. There is always a tendency for the air to
break away from the convex face of the impeller
vane. Here then is a point at which the Mach
number will be extremely important; a shock wave
might occur. If we assumed that the axial velocity
is uniform from the root to the tip of the eye, the
velocity of the air relative to the vane, V1 , will
reach a maximum at the eye tip where the vane M  V1
speed is the greatest. RT1
 University of Tehran, Faculty of Engineering

Even though this Mach number be satisfactory under ground level atmospheric
conditions, if the compressor is part of an aircraft gas turbine the Mach number
may well be too high at altitude. It is possible to reduce the relative velocity, and
hence the Mach number, by introducing pre-whirl at the intake. This is achieved
by allowing the air to be drawn into the compressor
eye over curved Inlet Guide Vanes (IGV) attached to the
compressor casing. An additional advantage of
using pre-whirl is that the curvature of the inlet
impeller vanes at the inlet is reduced, i.e. the inlet
angle  is increased.
This method of reducing the
Mach number unfortunately reduces the work
capacity of the compressor. The air now has an initial
Component C w1 , so that the rate of change of angular
momentum per unit mass flow of air is: C w 2 r2  C w1r1
If C w1 is constant over the eye of the impeller, then the initial angular momentum
will increase from root to tip of the eye. The amount of work done on each
kilogram of air will therefore depend upon the radius at which it enters the eye of
the impeller. A mean value of the work done per kg can be found by using the
initial angular momentum of the air at the mean radius of the eye.
University of Tehran, Faculty of Engineering

Pre-whirl
 University of Tehran, Faculty of Engineering

Example 4.3
Check the relative Mach numbers in the compressor dealt with
in previous example:
ψ =1.04, σ = 0.90
N = 290.0 rev/s,
D = 0.5 m
Deye,tip = 0.3 m, Deye,root = 0.15 m
m = 9.0 kg/s
T01 = 295 K
P01 = 1.1 bar
ηc = 0.78
Radial width of vaneless space 5 cm
Approximate mean radius of diffuser throat 0.33 m
Depth of diffuser passage 1.76 cm
Number of diffuser vanes 12
 University of Tehran, Faculty of Engineering

Solution
Consider first the inlet Mach number for the tip radius of the impeller eye.
Inlet velocity  143 m / s and is axial
Eye tip speed  273m / s
Relative velocity at tip  1432  2732  308 m / s
Velocity of sound  1.4  0.287  284.5  10 3  338 m / s
Max. Mach number at inlet  308 / 338  0.91
This would not be considered satisfactory even if it were actually the
maximum value likely to be reached. But if the compressor is part of an
aircraft engine required to operate at an altitude of 11000 m, where the
atmospheric temperature is only 217 K, we must calculate the Mach
number under these conditions. We will take 90m/s as being the minimum
speed likely to be reached at high altitude.
Temperature equivalent of forward speed  4K
Inlet stagnation temperature  217  4  221 K
Temp. equ. of axial inlet velocity from the first Example  10.5K
Inlet static temperature at altitude  210.5K
 0.91284.5 / 210.5  1.06
1/ 2
Inlet Mach number
 University of Tehran, Faculty of Engineering
Solution
This is clearly too high and we will find the Mach number when 30 degrees of
pre-whirl are introduced. In this case the absolute inlet velocity will be slightly
higher than before, so that the inlet static temperature will be slightly lower. A new
value for the axial velocity must be found by the usual trial and error process.
Reverting to the original sea-level case:
Try
C a1  150 m / s Ca1
C1   173.22 m / s
C1
2
cos 30 C
 14.9 K T01  295 K T 1  T o 1  1  280 . 1 K
2c p 2c p
po1 p1
p1   /(1)
 0.918 bar    1.21 kg / m 3
(To1 / T1) 1
RT1
Check on Ca1 m
C a1   149 m / s
1A1
Whirl velocity at inlet, Cw1  149 tan 30  86 m / s
Max. relative velocity V1  149 2  (273  86) 2  239m / s
239
Hence maximum Mach number when  0.71
1.4  0.287  280.110 3
Under altitude conditions this would rise to little more than 0.8 and 30 degrees of
Pre-whirl can be regarded as adequate.
 Solution
University of Tehran, Faculty of Engineering

To show the effect of 30 degrees pre-whirl on the pressure ratio, we will take
the worth case and assume that the pre-whirl is constant over the eye of the
impeller.
273  136.5
Speed of impeller eye at mean radius, Ue   204.8 m / s
Actual temperature rise, 2

T03  T01  (U 2  C w1U e )  1.04  (0.9  455.52  86  204.5) / 1.005  103  175 K
cp
po 3  ηc (To 3  T01 )   0.78  175 
3. 5

 1    1   3.79
p o1  To1   295 
This pressure ratio may be compared with the value of 4.23 obtained with no
pre-whirl.
Mach number in the diffuser
At the impeller tip C 2
T2  394.2 K , 2  93.8 K  C2  434m / s M  434
 1.09
2c p 2
1.4  0.287  394.2  10 3
At the leading edge of the diffuser vanes
C r 2  97.9 m / s, C w 2  342 m / s  C 2  356 m / s, T2  425.1K 356
M2   0.86
1.4  0.287  425.1 10 3

As long as the radial velocity component is subsonic , Mach numbers greater

than unity can be used at the impeller tip without loss in the Efficiency.
University of Tehran, Faculty of Engineering
Radial compressors
 Volute
University of Tehran, Faculty of Engineering
 University of Tehran, Faculty of Engineering

Non-dimensional numbers - maps

• We state that:
P02  f1 (m , P01 , T01 , n,  , R, , design, D)
c  f 2 (m , P01 , T01 , n,  , R, , design, D)
n  rotational speed
  kinematic viscosity
D  linear scale of design
based on the observation that we can not think of
any more variables on which P02 and ηc depends.
 University of Tehran, Faculty of Engineering

Non-dimensional parameters

• Nine independent parameters

• Four primary variables
– mass, length, time and temperature
• 9 - 4 = 5 independent non-dimensional
parameters
– According to pi theorem.
 University of Tehran, Faculty of Engineering

Non-dimensional numbers
• Several ways to form non-dimensional numbers
exist. The following is the most frequently used
formulation:
P02 m
  RT nD nD 2
 f1 ( 01
, , ,  , design )
P01 P01 D 2
RT01 
m RT01 nD nD 2
c  f 2 ( , , ,  , design )
P01 D 2
RT01 

( P02 / P01 )( 1) /   1

c 
(T02 / T01 )  1
 University of Tehran, Faculty of Engineering

Non-dimensional numbers
For a given design and working fluid we obtain:

P02 m
  RT nD nD 2
 f1 ( 01
, , )
P01 P01 D 2
RT01  
Re number

m RT01 nD nD 2
c  f 2 ( , , )
P01 D 2
RT01  
Re number

Compressors normally operate at such high Reynolds

numbers that they become independent of Re!!!
 University of Tehran, Faculty of Engineering

Compressor maps

( P02 / P01 ) (  1) /   1

c 
(T02 / T01 )  1

• Data is usually collected in

diagrams called
compressor maps
– What is meant by
surge?
– What happens at
right-hand extremities of
rotational speed lines?
University of Tehran, Faculty of Engineering
University of Tehran, Faculty of Engineering
 University of Tehran, Faculty of Engineering

Surge
• What will happen in point D if mass
flow drops infinitesimally
– Delivery pressure drops
– If pressure of air downstream of
compressor does not drop quickly
enough flow may reverse its
direction
– Thus, onset of surge depends on
characteristics of compressor and
components downstream
• Surge can lead to mechanical failure
 University of Tehran, Faculty of Engineering

Choke
• What happens for
increasing mass flow?
– Increasing mass flow
– Decreasing density
– Eventually M = 1 in some
section in impeller
(frequently throat of diffuser
 University of Tehran, Faculty of Engineering

Rotating stall
• Different kind of oscillatory
instability
– Compressor may still be in
stable operation (vibrations)
• Assume a breakdown of
flow in B
– Enforces breakdown in cell
A and B can recover since it
is now operating at
increased incidence
University of Tehran, Faculty of Engineering