NOZZLE THEORY

Alessandro Corsini,
Giovanni Delibra & Lorenzo Tieghi

Department of Mechanical & Aerospace Engineering

Sapienza University of Rome, Italy
• What is a nozzle?

A nozzle is a mechanical device that control the characteristics of a flow,

as it exits from a chamber into a receiving medium.
A nozzle is generally a duct or a pipe (rectilinear or curvilinear) with
varying sections

• Where do I find it?

Nozzles are used to accelerate fluid being part of stator cascades of
power-producing turbomachinery
Nozzles find application in every field of engineering, from engines to
atomiser to plastic extruder.

2
3
The ideal nozzle

The ideal nozzle is a quasi-one-dimensional nozzle flow under the following

assumptions:

1. Working fluid is homogeneous

2. Working fluid is a gas and free of condensed phases

3. The working substance obeys the perfect gas law

4. The flow is adiabatic, i.e. there is no heat transfer with the walls

5. There is no friction on the wall, i.e. boundary layers are neglected

6. Nozzle flow is free of shock waves or discontinuities

4
The ideal nozzle

The ideal nozzle is a quasi-one-dimensional nozzle flow under the following

assumptions:

7. Flow is stationary, i.e. time-dependent phenomena are neglected

8. Outlet velocity is perfectly axial

9. Flow quantities are circumferentially uniform, i.e. quantities vary only

along the axial direction

10. No chemical reactions

Under this assumptions, the ideal nozzle shows departure from real nozzles
of less than 6%.
5
First Law of Thermodynamics
In an adiabatic, no-shaft work process, the flow entropy does not change.

Ds = 0

Furthermore, the total stagnation enthalpy h0 is given by:

v2
h0  h   constant
2

Where v is velocity of the gas

6
Conservation of energy
The conservation of energy for isentropic flow between two sections,
hereafter denoted with the subscripts x and y, can be written as:
1 2 2
hx  hy   v y  vx   c p Tx  Ty 
2
where Cp is the specific heat at constant pressure.

An increase/decrease of enthalpy or thermal content leads to a

decrease/increase of kinetic energy

Conservation of mass
Also known as continuity equation, states that along every section the mass
flow rate remains the same

mx  my  m  Av / V

where A and V are the area of passage and the specific volume respectively.
7
Perfect gas Law

It is written as:
R
pxVx  Tx  RTx
n
with p the pressure, R’ is the universal gas constant (8314.3 J/Kg-mole-K)
and n the number of moles of the gas mixture. Further relationships to be
considered:

k  c p / cv
c p  cv  R
cP  kR /  k  1

with k the heat capacity ratio

8
Perfect gas Law

For isentropic flow process, the relations between two points x and y reads as:

Tx / Ty   px / p y   Vy / Vx 
 k 1/ k k 1

Looking at the previous expression, during an isentropic expansion Vy>Vx :

• the pressure drops substantially

• the absolute temperature also drops

• specific volume increases

9
Mach number

11
 Mach number

The velocity of sound a can be written as:

a  kRT

equal to 343.8 m/s in air at 293K. Thus, the Mach number can be defined as
the ratio between the current flow velocity and the local velocity of sound:

M  v / a  v / kRT

M<1 subsonic flow

Flow M=1 sonic flow

condition

M>1 supersonic flow

12
Isentropic flow in an ideal nozzle

13
14
15
Ideal nozzle

Example of a converging – diverging supersonic nozzle

Ac,Vc,pc
Inlet Outlet

Ay,Vy,py
Ax,Vx,px
Throat

16
Ideal nozzle

17
Isentropic flows equations

18
Isentropic flows equations

19
Isentropic flows equations

20
 Isentropic flows equations

Remark

21
 Isentropic flows equations

Remark

22
 Isentropic flows equations

Remark

23
Isentropic flows, Hugoniot eq.s

24
Isentropic flows, Hugoniot eq.s

25
Isentropic flows, Hugoniot eq.s

26
Isentropic flows, Hugoniot eq.s

27
Isentropic flows, Hugoniot eq.s

28
Isentropic flows, Saint Venant equation

29
Isentropic flows, Saint Venant equation

30
Isentropic flows, Saint Venant equation

31
Isentropic flows, Saint Venant equation

32
Throat or critical section

33
Isentropic flows, critical section quantities

34
Isentropic flows, critical section quantities

35
Isentropic flows, critical section quantities

36
Isentropic flows, critical section quantities

37
Isentropic flows, critical section quantities

38
Isentropic flows, critical section quantities

39
Isentropic flows, critical section quantities

40
Area ratio

The area ratio for a nozzle with isentropic flow can be written in term of Mach
numbers. It reads as:

( k 1)/( k 1)
Ay Mx 1   k  1 / 2  M 
2
y
  2
Ax M y 1   k  1 / 2  M x 

When the ratio between inlet and throath area A1/At is small, between 3 and 6,
the passage is convergent. In supersonic flow, the nozzle section diverges and
the ratio A2/At varies between 15 and 30 at M=4.

41
Isentropic flows, area ratio

42
Isentropic flows, area ratio

43
Isentropic flows, area ratio

44
Relationship of area ratio, pressure ratio and temperature ratio as a 45
function of Mach number, with Mx = 1.0
 Nozzle types

Convergent-divergent nozzles tare

also called De Laval nozzles

• From the continuity eq. area is inversely

proportional to the ratio specific volume
on velocity or (r c)

• A maximum in the curve (r c) because

in the convergent section velocity
increases more than the volume,
whereas in the divergent section trend
is opposite 46
 Nozzle types

Nozzle geometries

Based on the flow condition at the exit section, we find three different types
of nozzles

47
Nozzle characteristic, mass flow rate

48
Nozzle characteristic, mass flow rate

49
 Nozzle characteristic, mass flow rate

The mass flow through a subsonic nozzle is proportional to

the exit area A1 and the pressure expansion ratio p1/p0

The mass flow through a supersonic nozzle is proportional to

the throat area Ac or A* and the inlet pressure p0

50
Nozzle characteristic, Stodola cone

Const. inlet pressure p0

Const. discharge pressure p1

51
UNDER- / OVER- EXPANDED NOZZLES

(Off-design operation in nozzles)

52
Under-expanded nozzles:
• discharge the fluid at an exit pressure greater than the external pressure;
• the exit area is too small for an optimum area ratio;
• expansion of the fluid is incomplete and must take place outside.

Over-expanded nozzles:
• discharge the fluid at lower pressure than the exterior;
• the exit area is too large for optimum;
• expansion is completed in the nozzle entirely.

53
Over-expanded nozzles:

• Curve AB shows the optimum

expansion.
• AC and AD show the variation of
pressure for increasingly higher
external pressures.
• To compensate external pressure, we
find a sudden expansion (point I).
• Point I is accompanied by the
separation of the flow from the walls.
• Such phenomena are also called
compression waves.

Distribution of pressures in a
converging-diverging nozzles for
different flow conditions.
54
Four possible flow conditions:

1. external pressure is below the nozzle exit pressure: the nozzle will flow fully with
external expansion waves at the exit.
2. external pressure is slightly higher than the nozzle exit pressure: nozzle is flowing
fully. This behaviour is found for ration between external and exit pressure
between 25% and 40%. Shock waves persist outside the exit section.
3. external pressure is dramatically higher than nozzle exit pressure: the diameter of
the supersonic jet is smaller than the nozzle exit diameter. Axial-symmetric
separation occurs on the walls. The point of separation travels downstream with
decreasing external pressure. Shock waves outside the nozzle.
4. exit pressure is just below the values of the inlet pressure: the pressure ratio is
below the critical pressure ratio, subsonic low prevails throughout the entire
nozzle.

55
56
 Nozzle and Guide Vanes in Turbomachinery

e.g. in gas turbines

Combustor
NGV 1° stage
Rotor 1° stage
NGV 2° stage
Rotor 2° stage

57
Nozzle in turbomachinery

58
Nozzle in turbomachinery

59
Nozzle in turbomachinery

60
Nozzles configurations

61
 Nozzle design rules
1-D theory does not provide guidelines for the design of nozzle (either rectilinear
curvilinear)
Experimentally derived design recommendations are
here listed

Convergent section
Not influencial on performance, axial length
minimized keeping bell-mouth shape at inflow
(manufacturing limitations)

Divergent section
Length L to be minimized provided the limitations on
the opening angle q due to the occurrence of flow
separations

In case p2/p1 is small and A1/Ac is high nozzle design

typically favours short length divergent letting the fluid
expands into the receiving ambient (2) 62
 Nozzle design rules (propulsion)

Three flow conditions during operating conditions of a rocket

63
Most of the proven nozzle configurations are directly derived from rockets and propulsions.

Simplified configurations of nozzles and flow effects

64
• Nozzles and chambers are usually of circular cross section and have a converging
section and a throat in the narrowest location, followed by a diverging section;

• The converging section has low impact on the overall performances, and any radius,
cone angle, wall contour curve or nozzle inlet shape is satisfactory;

• The throat contour is also not very critical to performance, any radius or other curve
is usually acceptable. The only relevant factor is the wall friction. Wall surface needs
to be smoother as possible to minimize radial absorption and heat transfer;

Length comparison of several type of nozzles

65
Cone- and Bell-Shaped Nozzles
The conical nozzles is the oldest and simplest configuration.

Bell vs conical nozzles: design parameters

66
Cone-Shaped Nozzles
A theoretical correction factor l is usually applied to the nozzle exit
momentum of an ideal rocket with a conical nozzle exhaust.

1
 1  cos  
2

l is the ratio between the momentum of the gases in a nozzle with a finite nozzle
angle of 2a and the momentum of an ideal one-dimensional axial flow.

67
Cone-Shaped Nozzles
Small divergence angle cause most of the momentum to be axial and thus
has high thrust, but has high penalties in term of weight and complexity. A
large divergence gives short and lightweight design, but the performance
are lower.

68
Bell-Shaped Nozzles
• The length of a bell nozzle is usually given a fraction of the length of a
reference conical nozzle with a 15 deg half angle.
• An 80% bell nozzle has a length that is 20% shorter than a 15 deg cone
with the same aria ratio.

69
Bell-Shaped Nozzles
• Design of the bells can be approximated with a parabola
• Parabola is tangent qi at point I and has an exit angle qe at point E

• Design is scarcely influenced by the throat approach radius ri and the

throat expansion radius rt

70
Two-Step Nozzles
Several correction to the Bell- and Cone-
shaped nozzles, to compensate the effect of
altitude. Some of them are:
• Extendable nozzles: actuators are used to
increase the length of the nozzle during
flights. Complex and requires a power
supply.
• Droppable insert concept: a mechanism is
used to enlarge the section area.
• Dual bell nozzles: two shortened bell
nozzles are combined into one with a bump
or inflection point between them

Top to bottom: extendable nozzles, droppable

inserts, dual-bell
71
Real nozzles

72
Nozzle efficiency

73
REAL NOZZLES
• The flow is two-dimensional but axisymmetric;
• Temperatures and velocities are not uniform over any section;
• Exit velocity v2 can be seen as a function of the radius r:

2 2
v2,average 
A2 
0
v2 rdr

The 11 assumptions at the begin of the chapter cease to be universally true.

It is possible to apply them through:
1. Empirical correction factors
2. More accurate computing of the phenomena involved

74
REAL NOZZLES
Losses in real nozzles can be ascribed to:
1. Divergence of the flow at the exist section
2. Low nozzle contraction ratios A1/At cause pressure losses in the chamber
3. Wall friction reduce the performances up to 1.5%
4. Solid particles or liquid roplets reduce performance up to 5%
5. Unsteady combustion and oscillating phenomena are responsible of small
deflections
6. Transient conditions lead to lower performance
7. Erosion leads to non-optimal design of the throat section
8. Non-uniform gas composition could lead to incomplete combustion / non efficient
mixing
9. Undesired chemical reactions may change gas composition slightly
10. Real gas properties are not fixed but changed along the nozzle
11. Operations at out-of-optimum design points usually lead to lower performances
75
Boundary Layer
Real nozzles have a viscous boundary layer next to the nozzles walls.

Immediately next to the wall the flow velocity

is zero. Here the low-velocity flow is laminar
and subsonic, whereas in higher-velocity
regions the flow is supersonic and can
become turbulent.
In the boundary layer the local temperature
can be substantially higher than the core
flow, due to conversion of kinetic energy into
thermal energy. However, the layer
immediately at contact with the wall will be
cooler due to the heat transfer with the wall.
This deflection from the theoretical design
can lead to a thrust lost of 1%.
76
Multiphase Flow

• Real flows contain many small liquid droplets and solid particles that are accelerated
by the gas.
• Common examples are aluminium powders, ion oxides, propellants with beryllium or
boron.
• If the diameter of particles is small (<0.005 mm), the multiphase effect can neglected,
as particles are in thermal equilibrium with the main flow.
• If the diameter is larger, the inertia of the particles grows as the cube of the diameter,
while the drag force increases with the square.
• Larger particles do not move at the same speed as the core flow and reach higher
temperature than the smaller.

77
Multiphase Flow

• Theoretical model can be therefore be corrected to account multi-phasing.

• Governing equations can be rewritten as a function of the particle fraction b, defined
as the mass of particles divided by the total mass. Denoting with s and g subscripts
the two phases (solid and gas), we derive:

h  1    c p , g T   csT
V  Vg 1   
p  RgT / Vg
R  (1   ) Rg
1    c p , g   cs
k
1    cv, g   cs
78
Temperature ratio

Stagnation temperature can therefore be written as a function of the Mach

number:

T0  T  v 2 /  2c p 
c p  kR /  k  1

T0  T  v 2  k  1 /  2kR 

  k  1 / 2kR 
2
T0  T  M kRT
T0  T  M 2 kRT  k  1 /  2kR 
 1 
T0  T 1   k  1 M 2 
 2 

79
Pressure ratio

Thus, Mach number is also computed as:

2  T0 
M   1
k 1  T 
By looking back at the stagnation pressure equation we get:
k /  k 1
p0  p 1  v /  2c pT 
2

k /  k 1


p0  p 1  M kRT  /  2c pT 
2

 
k /  k 1
 1 
p0  p 1   k  1 M 2 
 2 

80
Isentropic flow trough nozzles

From the conservation of energy we can derive the exit velocity:

1 2 2
h1  h2   v2  v1   c p T1  T2 
2
v2  2  h1  h2   v12

By substituting the formula for enthalpy hx = cp(Tx-Ty) in the former expression:

v2  2c p T1  T2   v12

k k  T1  2
v2  2 R T1  T2   v1  2
2
RT1 1    v1
k 1 k 1  T2 

81
Isentropic flow trough nozzles

k  T1  2
v2  2 RT1 1    v1
k 1  T2 

and considering the temperature ratio:

If A1 >> At

Tx / Ty   px / p y   Vy / Vx 
 k 1/ k k 1

k   p2 
( k 1)/ k
 k   p2 
( k 1)/ k

v2  2 RT1 1      v1  2
2
RT1 1    
k 1   p1   k 1   p1  

This expression is still viable for any two points inside the nozzle. When the
inlet section is large compared to the nozzle throat section, the term v 1 can
be neglected.

82
Isentropic flow trough nozzles

The maximum theorical outlet velocity can be therefore calculated by

considering an expansion into the vacuum, i.e. p2 = 0.

k   p2 
( k 1)/ k

v2  2 RT1 1    
k 1   p1  

k
v2,max  2 RT0
k 1

83
 Nozzle characteristic, mass flow rate

The mass flow through a supersonic nozzle is proportional to

the throat area Ac or A* and pressure p1 (outlet or exit
pressure)

2 / (k  1)
( k 1)/( k 1)
At vt
m   At p1k
Vt kRT1

For a supersonic nozzle the ratio between the throat and any downstream
area x, as follows:

At Vt vx  k  1 
1/( k 1) 1/ k
 px  k 1   px 
( k 1)/ k

     1    
Ax Vx vt  2   p1  k  1   p1  


84
Nozzle flow

The minimum nozzle area is called throat area, hereafter denoted with the
subscript t. We define:

A2
nozzle area
expansion rate 
At
The maximum gas flow per unit area occurs at the throat. Here the gas
ratio is only a function of the ratio of specific heats k, obtained by setting
M=1.
k /  k 1
pt / p1   2 /  k  1
pt is also known as the critical pressure, with common values of the ratio
reported between 0.53 and 0.57.

85
Nozzle flow

At the point of critical pressure (the throat), the Mach number is equal to
one, therefore the specific volume and temperature can be written as:
1/  k 1
Vt  V1  k  1 / 2 
Tt  2T1 /  k  1
Because of M = 1, throat velocity can be immediately derived by the
definition of Mach number:

2k
vt  at  M kRT  kRT  RT1
k 1
The throat velocity is clearly the sonic velocity and it can be directly derived
from the inlet temperature T1.
86