Documente Academic
Documente Profesional
Documente Cultură
Lecture1:
Compressible Aerodynamics
NPTELIITKharagpur:Prof.K.P.Sinhamahapatra,Dept.ofAerospaceEngineering
Thermodynamics
Experimental results are the basis of any physical theory. The experimental basis of
thermodynamics is formalized in the principal laws. The law of conservation of energy is one of
these principal laws. It introduces the concept of internal energy of a system. The other principal
laws of thermodynamics introduce and define the properties and concepts of temperature and
entropy. Classical thermodynamics is concerned, at any rate as the bulk of the subject stands,
with equilibrium state of uniform matter, that is, with states in which all local mechanical,
physical and thermal quantities are virtually independent of both position and time.
Thermodynamical results may be applied directly to fluids at rest when their properties are
uniform. A very little is known of the thermodynamics of non-equilibrium states. However,
observation shows that results for equilibrium states are approximately valid for the nonequilibrium non-uniform states common in practical fluid dynamics; large through the departures
from equilibrium in a moving fluid may appear to be, they are apparently small in their effect on
thermodynamical relationships.
Fluid mechanics of perfect fluids (without viscosity and heat conductivity) is an extension of
equilibrium thermodynamics to moving fluids. In addition to internal energy, kinetic energy of the
fluid needs to be considered. The ratio of the kinetic energy per unit mass to the internal energy
per unit mass is a characteristic dimensionless quantity of the flow problem and in the simplest
cases is directly proportional to the square of Mach number.
In fluid mechanics of low speed flow, thermodynamic considerations are not needed: the heat
content of the fluid is then so large compared to the kinetic energy of the flow that the
temperature remains nearly constant even if the whole kinetic energy is transformed into heat.
The opposite can be true in high-speed blow problems.
Enclosure = a closed surface with its properties defined everywhere may or may not transmit
heat, work or mass.
NPTELIITKharagpur:Prof.K.P.Sinhamahapatra,Dept.ofAerospaceEngineering
The concepts of thermodynamics are helpful in fluid mechanics for the additional reason that in
both subjects the objective is a set of results which apply to matter as generally as possible,
without regard for the different molecular properties and mechanism at work. Additional results
may be obtained by taking into account any known molecular properties of a fluid (with the aid
of kinetic theory in case of certain gases).
It is a fact of experience that the state of a given mass of fluid in equilibrium (spatial and
temporal uniformity) under the simplest possible conditions is specified uniquely by two
parameters, which for convenience may be chosen as the specific volume = 1
and the
pressure p. All other quantities describing the state of the fluid are function of these two
parameters of state. One of the most important of these quantities is the temperature. A mass
of fluid in equilibrium has the same temperature as a test mass of fluid also in equilibrium if the
two masses remain in equilibrium when placed in thermal contact (Zeroth law). The relation
between the temperature T and the two parameters of state may be written as
f ( p, , T ) = 0
This exhibits formally the arbitrariness of the choice of the two parameters of state. The
equation is called the equation of state. Generally written as p = p( , T ) and is called thermal
equation of state. Another important quantity describing the state of the fluid is the internal
energy per unit mass e. The change in the internal energy of the system (mass of fluid) at rest
consequent on a change of state is defined by the first law of thermodynamics, as being such as
to satisfy the conservation of energy when account is taken of both heat given to the fluid and
work done on the fluid. Thus if the state of a given uniform mass of fluid is changed by a gain of
heat of amount Q per unit mass and by the performance of work on the fluid of amount W per
unit mass, then
e = Q + W
e = e(v, T ) is the caloric equation of state
The internal energy e is a function of the state parameters, and the change which may be either
infinitesimal or finite depends only on the initial and final states, but Q and W are measures of
external effects and may separately depend on the particular way in which the transition
between the two states is made. If the mass of fluid is thermally isolated from its surrounding so
that no exchange of heat can occur, Q = 0 , and the change of state of the fluid is adiabatic.
NPTELIITKharagpur:Prof.K.P.Sinhamahapatra,Dept.ofAerospaceEngineering
The most important way of performing work on the system is compression. Analytic expression
can be obtained if the change is reversible. This implies that the change is carried out so slowly
that fluid passes through a succession of equilibrium states, the direction of the change being
without effect. At each stage, the pressure is uniform p, so the work done on unit mass of fluid
for small decrease in volume is p . Thus for a reversible transition from one state to another
neighbouring state,
e = q p
The particular path by which the initial and final equilibrium states are joined is relevant here,
because p is not in general a function of alone.
Another practical quantity of some importance is the specific heat of the fluid, which is the
amount of heat given to unit mass of the fluid per unit rise in temperature in a small reversible
change. The specific heat may be written as
c = Q
This is not uniquely determined until the conditions under which the reversible changes occur
are specified: An equilibrium state is a point on a
( p, )
reversible change (p, ) starting from a point A may proceed in any direction.
adiabatic
pA
isothermal
AA
A
If the only work done on the fluid is that done by compression, the heat Q which must be
supplied to unit mass is determined as
E
E
p +
+ p
p
p
Q =
NPTELIITKharagpur:Prof.K.P.Sinhamahapatra,Dept.ofAerospaceEngineering
T = p +
p
p
The specific heat thus depends on the ratio p
change from A. Two particular well defined choices are changes parallel to the axes of the
indicator diagram giving the principal specific heats
Q
E
Cp =
= + p ;
T p
T p T p
Q E
C =
=
T T
T various sinusoidally as the point representing the final state moves round a circle of small
G
radius centered on A, being zero on the isotherm through A and maximum in a direction m
normal to the isotherm. Likewise, Q varies sinusoidally, being zero on the adiabate through A
and a maximum in a direction n normal to it. The components of the unit vectors being
(m , m ) and (n , n )
Cp =
n (Q )max
,
m (T )max
C =
n p (Q )max
m p (T )max
Since m m p and n n p are the gradients of the isothermal and adiabatic lines, the ratio of
the principal specific heats is
n
=
C n p
Cp
m
p
p
=
adiab
or =
T
p T
p a
1
is more useful.
or
p
p
This is called the coefficient of compressibility. Like specific heat, the bulk modulus or the
coefficient of compressibility takes a different value for each direction of change. Adiabatic and
NPTELIITKharagpur:Prof.K.P.Sinhamahapatra,Dept.ofAerospaceEngineering
isothermal changes correspond to two particular directions with special significance and the first
law requires the ratio of the two corresponding bulk module to equal the ratio of the principal
specific heats.
____________________________________________________________________________
___________________________________________________________________________
NPTELIITKharagpur:Prof.K.P.Sinhamahapatra,Dept.ofAerospaceEngineering
Lecture2:
Compressible Aerodynamics (Contd.)
NPTELIITKharagpur:Prof.K.P.Sinhamahapatra,Dept.ofAerospaceEngineering
Thermal equation of state for all gases for low densities approaches
p = PRT,
T in K or R,
C + 273.16 or F + 459.69
This defines a family of perfect gases, one for each value of R. Any gas at low enough density
approaches a perfect gas with a particular value of R. the caloric equation of state for a perfect
gas is e = constant T = C T
Every real gas can be liquefied. The highest temperature at which this is possible is called
critical temperature Tc, the corresponding pressure and density are called critical pressure pc
and critical density c . Critical variables are characteristics of a gas and depend on
intermolecular forces. At the critical point
p 2 p
=
= 0. An equation of state for a real gas must
v v 2
1
p = RT
1 p RT
p=
27
RTc , p 2 = 27 pc
8
e = eo (T )
= eo (T )
It is clearly possible to draw lines defining the direction of a small reversible change involving no
gain or less of heat through each point of the indicator diagram, and to regard the family of
these adiabatic lines of equal value of some new function of state. The properties of this
function are the subject of the second law. The second law implies the existence of another
extensive property of the fluid in equilibrium (even for systems with more than 2 independent
parameters of state) termed the entropy such that in a reversible transition from an equilibrium
state to another neighbouring equilibrium state, the increase in entropy is proportional to the
heat given to the fluid and that the constant of proportionality itself is a function of state,
depends only on the temperature and can be chosen as the reciprocal of the temperature. With
entropy per unit mass of a fluid s, we have
Tds = dq = the infinitesimal amount of hat given reversibly.
This is the means by which the thermodynamic or absolute scale of temperature is defined.
NPTELIITKharagpur:Prof.K.P.Sinhamahapatra,Dept.ofAerospaceEngineering
An adiabatic reversible transition takes place at constant entropy, and the process is called
isentropic. It is a consequence of the second law that in an adiabatic irreversible change the
entropy can not diminish.
Hence, for a small reversible change in which work is done on fluid only by compression
Ts = e + p
Since the equation contains only functions of state, the relation must be valid for any
infinitesimal transition in which work is done by compression, whether reversible or not. If the
transition is irreversible Ts Q , and w p
Another function of state which like internal energy and entropy proves to be convenient for use
in fluid mechanics particularly when effects of compressibility of the fluid are important is the
enthalpy or heat function. The enthalpy of unit mass of fluid is
h = e + p
h = e + pd + dp = Ts + p
The relation involves only state functions. For a reversible small change at constant pressure
h = q
Helmholtz free energy
F = E Ts per unit mass
F = E Ts sT
= p sT
Thus, the gain in free energy per unit mass in a small isothermal change, whether reversible or
not, is equal to p . When this small isothermal change is reversible the gain in free energy is
equal to the work done on the system.
Another form of free energy is Gibbs free energy defined as
G = E + pV Ts = H Ts
G = V p s T
Using and s as the 2 independent parameters
E
E
Ts = E + p
= p,
=T
s
s
NPTELIITKharagpur:Prof.K.P.Sinhamahapatra,Dept.ofAerospaceEngineering
2E
may be obtained in two different ways.
s
2E
p
= ,
s
s
2 E T
=
s s
Hence
p
T
=
s
s
Similarly,
s p
=
p s
T p
s p s
= ;
=
p T T T
may be obtained by forming the double derivative, in two different ways, of the functions
E + p , E Ts and
h
E + p Ts = H Ts = G
The four relations given above are known as Maxwells thermodynamic relations.
Alternatively, from the first relation
T p
=
p s v s
p
s
p
T
=
=
s s p
s
=
s p p s
Coefficient of thermal expansion of fluid is
1
=
T p
It plays an important role in considerations of the action of gravity on a fluid of non-uniform
temperature.
Specific heats using entropy
C=
Q
s
=T
T
T
NPTELIITKharagpur:Prof.K.P.Sinhamahapatra,Dept.ofAerospaceEngineering
s
Cp = T
,
T p
s
C = T
Using s = s (T , ),
s =
s
s
T +
T
T
s
s s
=
+
T p
T T T p
s
p
=
T
T
Since
p
C p C = T
T T p
RHS can be calculated from the equation of state.
In terms of easily measurable quantities,
p
C p C = T
T T p T p
p
p
= T
= T
T T p
T T p
2
S = S (T , p )
s
s
s =
T + p
T p
p T
C p s
s
, =
=
T p
T p T
p T
s = C p
p
T T p
= Cp
T
T
or Ts = C p T Tp = E + p
Except s and E , others are directly observable
C pT
Tp
p T
C p C T T p p
Cp
NPTELIITKharagpur:Prof.K.P.Sinhamahapatra,Dept.ofAerospaceEngineering
T
T p
p
p T
E = E (s, )
H = H (s , p )
S T
p
Conjugate variables
s = s ( .T )
s
s
ds =
dT +
d
T
T
C
dT p
+
d
T T
dT R
+ d
T
s 2 s1 = C
1
2 R
dT
+
d
1
T
s 2 s1 = C ln
T2
+ R ln 2
1
T1
Similarly, s = s ( p, T )
s
s
ds =
dT + dp
T p
p T
= Cp
dT
dp
T T p
Integrating
s2 s1 = C p ln
T2
p
R ln 2
T1
p1
NPTELIITKharagpur:Prof.K.P.Sinhamahapatra,Dept.ofAerospaceEngineering
Isentropic relations
T
p
p T
C p ln 2 = R ln 2 2 = 2
T1
p1
p1 T1
Cp R =
Cp / R
p 2 T2
=
p1 T1
Hence
Similarly,
R ln
or,
2
T
= C ln 2
T1
1
2 T2
=
1 T1
2 = 2
1 T1
1 1
C R
C
1
=
R 1
1
2 T2 1
or
=
1 T1
T
p
Hence, 2 = 2 = 2
p1 1 T1
G
G G
D
1 2
e + V = pV = p V + V p
Dt
2
G
1 D
From continuity equation V =
Dt
D
1 G G p D G
V p
e + V V =
Dt
2
Dt
NPTELIITKharagpur:Prof.K.P.Sinhamahapatra,Dept.ofAerospaceEngineering
p D Dp
D( p )
=
Dt Dt
Dt
Now
1 G G Dp G
D
p
p
V p =
e + + V V =
Dt
2
t
Dt
or
V 2 p
D
=
h +
=0
2 t
Dt
1
The time rate of change of h + V 2 following a moving fluid element is zero.
2
1
h + V 2 = constant along a path or streamline (steady, adiabatic, inviscid flow)
2
The total enthalpy ho is enthalpy at a point if the fluid element were brought to rest adiabatically
h+
V2
= ho
2
1
2
Dho
=0
Dt
The total enthalpy is constant along a streamline. If all the stream lines of the flow originate from
a common uniform free stream, then ho is constant for each line. Hence, for steady, inviscid,
adiabatic flow, the energy equation becomes
ho = constant everywhere
or, Alternatively To = constant for calorically perfect gas.
For a general non-adiabatic flow, the forms of the energy equation are not valid but the
definition of total quantities hold locally at each point of the flow. At point 1, the local static
1
2
enthalpy and velocity are h1 and V1 and the total enthalpy ho1 = h1 + V1 ; and at point 2,
2
1 2
ho 2 = V2 + h2 . But ho1 ho 2 . Only if the flow between point 1 and 2 is adiabatic ho1 = ho 2
2
The total (stagnation) pressure and density are the pressure and density if the fluid is brought to
rest adiabatically and reversibly, i.e., isentropically (temperature To ). The definition of po and
NPTELIITKharagpur:Prof.K.P.Sinhamahapatra,Dept.ofAerospaceEngineering
o involve isentropic assumption. However, the concept of total pressure and density can be
applied throughout any general non-isentropic flow. If the flow is non-isentropic between points
1 and 2, po1 po 2 , o1 o 2 . But if the flow is isentropic between points 1 and 2 then
po1 = po 2 , and o1 = o 2 . If the general flow field is isentropic throughout then both po and
NPTELIITKharagpur:Prof.K.P.Sinhamahapatra,Dept.ofAerospaceEngineering
Module2:
One-Dimensional Gas Dynamics
Lecture3:
Governing Equations
For an incompressible flow, complete information about a one-dimensional flow is obtained from
the kinematic relation: u is inversely proportional to A. The pressure is then obtained from the
Bernoullis equation. For a compressible flow the relation between velocity and area also depends
on density variation since the governing equations are interdependent.
If the flow is unsteady then the mass contained between sections 1 and 2, x distance apart,
changes at the rate
equal to the flow through 1 minus flow through 2, i.e., the net inflow
( Ax ) = ( uA) x
t
x
or
(A) + (uA) = 0
t
x
3
If the flow in the tube is steady, the continuity equation is
d
( uA) = 0
dx
This implies that the mass of fluid that passes a given section must pass all the other sections
downstream. At any two sections where conditions are uniform 1u1 A1 = 2u2 A2 .
This equation is general, since it holds even if the conditions between the sections are not
uniform. If the flow is uniform at every section, the equation can be written as
uA = m = Constant
Eulers equation or Momentum equation
u
u
1 p
+u
=
t
x
x
for steady flow, the first term is zero, and the derivatives become total derivatives.
u du +
or
dp
=0
1 2
dp
u +
= constant
2
It is often convenient to express Eulers equation in an alternative form that describes the changes
in momentum of the fluid within a fixed control space.
Multiplying Eulers equation by A and the continuity equation by u
A
u
p
u
u
+ uA = A
t
x
x
(A) + u (uA) = 0
t
x
(uA) + u 2 A = A p = ( pA) + p A
t
x
x
x
x
2
2
(
uA)dx + 2 u 2 2 A2 1u12 A1 = ( p1 A1 p2 A2 ) + pdA
1
t 1
The first integral is the momentum of the fluid enclosed between 1 and 2 and the last integral may
be evaluated by defining a mean pressure p m .
2
(uA)dx + 2u 2 2 A2 1u12 A1 = ( p1 A1 p2 A2 ) + pm ( A2 A1 )
1
t
Unsteady
change in space
The integral form of the momentum equations is more general, since it is valid even when there
are dissipative processes within the control space, provided that the reference sections are
equilibrium states. During the integration of the differential momentum equation the forces on
adjacent internal faces cancel as they are equal and opposite and only the forces and the fluxes
at the boundaries of the control space are left out. If there is a non-equilibrium region inside this
space, it does not affect the integrated result.
For steady flow in a duct of constant area, the momentum equation becomes
2 u 2 2 1u1 2 = p1 p 2
Energy equation
For a fluid flow problem the basic thermodynamic quantity is the enthalpy, rather than internal
energy due to the presence of flow work. In adiabatic flow through a resistance the total enthalpy
per unit mass upstream and downstream of the resistance is the same.
Lets select a definite portion of the flowing fluid, between sections 1 and 2 for the system.
5
is e +
1 2
u per unit mass. Comparing the energy of the system after the displacement with that
2
1 2
1 2
e2 + u 2 e1 + u1
2
2
Hence, the steady flow energy equation is
1 2
1 2
q + p1 1 p2 2 = e2 + u 2 e1 + u1
2
2
or
1 2 1 2
q = h2 h1 + u 2 u 1
2
2
1 2
1 2
h2 + u 2 = h1 + u1
2
2
These equations relate conditions at two equilibrium states. They are valid even if there are
viscous stresses, heat transfer, or other non-equilibrium conditions between the two sections
provided sections1 and 2 are equilibrium states.
If equilibrium exists all along, the equilibrium equation is valid everywhere and may be written as
1
h + u 2 = constant
2
or dh + udu = 0
For a thermally perfect gas this becomes C p dT + udu = 0
And for a thermally and calorically perfect gas C pT +
1 2
u = constant
2
1
h + u 2 = h0 = constant
2
h0 is called reservoir or stagnation enthalpy, the enthalpy of the fluid in a large reservoir where
velocity is practically zero.
If there is no heat addition to the flow between two reservoirs, then the enthalpy of both the
reservoirs is same, h0 . Since h0 = C pT0 for a perfect gas, the stagnation temperatures in the two
reservoirs are also same.
h0 = h0 , T0 = T0
From second law of thermodynamics s0 s0
s0 s0 = R ln
po
po
p0
p0
+ C p ln
T0
T0
>1
dp ,
The increase of entropy, and the corresponding decrease of stagnation pressure, represents an
irreversible process. Entropy is being produced in the flow between the reservoirs. The flow is not
in equilibrium throughout. Only if the flow is in equilibrium throughout, entropy will not be produced
and the flow will be isentropic. Only in such isentropic flow s0 = s0 ,
p0 = p0
The reservoir conditions or stagnation conditions are also called total conditions. The terms are
used to define conditions at any point in the flow. The total conditions at any point in the flow are
the conditions that would be attained if the flow there were brought to rest isentropically. For
stagnation conditions to exist it is not enough that the velocity be zero, it is also necessary that
equilibrium conditions exist.
Since the imaginary local stagnation process is isentropic, the total entropy at any point is by
definition equal to local static entropy so = s . Since To = To , then the local entropy for a perfect
gas is related to total pressure by
s s = R ln
po
po
A flow which is in equilibrium and adiabatic is isentropic. For adiabatic, non-conducting flow the
energy equation
udu + dp = 0
is applicable everywhere
dh dp = 0
ds =
1
dp
dh = 0
T
Thus an adiabatic, non-conducting, inviscid flow is isentropic. In this case, either the momentum
or the energy equation can be replaced by the equation
s = constant
For a perfect gas, using dh
= =T
po o To
dp
The conditions of equilibrium cannot be strictly attained in a real, non-uniform flow, since a fluid
particle must adjust itself continuously to the new conditions that it encounters. Thus, entropy
production is never strictly zero.
Module2:
One-dimensional gas dynamics
Lecture4:
Governing Equations(Contd.)
Speed of sound is the speed at which small disturbances or waves are propagated through a
compressible fluid or an elastic medium in general. Its relation to the compressibility of the fluid is
given by
p
a2 =
,
s
=
Ks
isentropic compressibility s =
1
p s
The disturbances (the temperature and velocity gradients) produced in a fluid by a sound wave
are so small that each fluid particle undergoes a nearly isentropic process. In a perfect gas
p = const.
a2 =
p
= RT
In a flowing fluid, the speed of sound is a significant measure of the effects of compressibility
when it is compared to the speed of the flow. This introduces the dimensionless parameter called
Mach number M =
u
a
M will vary from point to point in a flow because of change in u and a . In an adiabatic flow an
increase in u always corresponds to an increase of M. A flow is called subsonic if M < 1 and it is
called supersonic if M > 1.
Area-velocity Relations
For a steady adiabatic flow in a stream tube of varying area the continuity equation is
du dA
+
=0
u
A
For incompressible flow d = 0 and this gives the simple result that increase or decrease of
velocity is proportional to decrease or increase of area. The change in density modifies this simple
relation. Using Eulers equation for steady flow
u du =
dp
dp d
d
u du = a 2
or
= M 2
dp dp
=
d d s
du
u
At very low Mach numbers the density changes are so small compared to the velocity
changes, that they may be neglected in flow computation and it may be considered that =
constant. Hence, equivalent definitions of incompressible flow are a = or M = 0 .
The continuity equation now becomes
M2
or
du du dA
+
+
=0
u
u
A
du dA A
=
u 1 M 2
dA
du
can be finite only if
= 0 . Considering a tube in which the
u
A
velocity increases continuously from zero to supersonic speeds, it is obvious from above that the
tube must converge in the subsonic portion and diverge in the supersonic portion. Hence, the
area must be minimum at M = 1, sonic condition is reached only at the throat. The same logic
holds when the velocity decreases continuously from supersonic to subsonic. Hence M = 1 can be
attained only at a throat of the tube. The inverse is not true, that is, M is not necessarily 1 at a
throat. Since the equation shows that a throat corresponds to du = 0 ; the velocity attains a
maximum or a minimum there, depending on whether the flow is subsonic or supersonic. Near M
= 1, the flow is very sensitive to changes in the area, since the denominator (1 M2) is very small.
In adiabatic flow the energy equation for a thermally and calorically perfect gas is
1 2
u + C pT = C pTo
2
2
Now, a = RT = C p 1
or
C pT =
1
T = ( 1)C pT
a2
1
2
a
u2
a2
+
= o
2 1 1
Multiplying by
1
a2
2
ao
T
1 2
M +1 = 2 = o
2
T
a
To
1 2
M
= 1+
2
T
Hence, the isentropic relations become
( 1)
po
1 2
M
= 1 +
p
2
1 2
= 1 + 2 M
1 ( 1)
In the above equations, the values of To and ao are constant throughout the flow and can be taken
as the actual reservoir value. The values of p o and o are the local reservoir values. They are
constant throughout only if the flow is isentropic.
Instead of the reservoir, any other point in the flow can be used for evaluating the constant in the
energy equation. The throat, where M = 1, is a very useful point. The flow variables at the throat
are called sonic and are denoted by superscript *. The flow speed and sound speed are u and
u2
a2
u 2 a 2
1 + 1 2 ao
a =
+
=
+
=
2 1 2 1 2 1
1
2
a
T
2
2 =
=
+ 1 To
ao
2
Thus for a given fluid the sonic and the reservoir temperatures are in a fixed ratio, so that T* is
constant throughout in an adiabatic flow.
For air
T
= 0.833,
To
a
= 0.913
ao
p 2 1
= 0.528
=
po + 1
2
=
o + 1
= 0.643
It is not necessary that a throat actually exist in the flow for sonic values to be used as reference.
u2
a2
1 + 1 2
+
=
a
2 1 2 1
1
1
1 +1 1
+
=
2
2 ( 1)M
2 1 M 2
Or,
2 + ( 1)M 2
+1
=
2
( 1)M
( 1)M 2
or M
( + 1)M
=
2 + ( 1)M 2
2
+1
2
=
1+
M2
1
2
M2
+1
2
+ 1
M2
Alternatively
M2 =
M 2
2
=
+ 1 1 2 + 1
M
( 1)
2
2
M 2
u2
p
po
+
=
2 1 1 o
For isentropic conditions, p
p
p = o
o o
po
po p
o po
u2
po p
+
2 1 o po
po
1 o
This is the steady state Bernoullis equation for an adiabatic compressible flow.
In a compressible flow, the dynamic pressure
1 2
u (used for normalizing pressure and forces) is
2
not simply the difference between stagnation and static pressure. It depends on Mach number as
well as static pressure.
1 2
u
2
2
1
M2a
2
Hence, C p =
p p
1
U 2
2
=
=
1 p 2 1
M = pM 2
2
2
p p
1
p M 2
2
2 p
1
2
M p
Cp =
2
M 2
2
1
1
M
(
)
2
2 + ( 1) M
U 2
u2
2
,
M
=
a 2
a2
M or a can be eliminated from the above definition using the energy equation in the form
U2
a2
u2
a2
+
= +
2 1 2 1
1
2
2
1 2
u
Cp =
1
1+
M 1 2
2
M
2
U
Module2:
One-dimensional gas dynamics
Lecture5:
Governing Equations(Contd.)
The inviscid incompressible flow through a uniform duct has only one possible solution, which is
of uniform flow. However, there are two possible solutions when the flow is compressible. If there
is no change in entropy anywhere, the only possible solution is the uniform flow. An alternative
solution which contains a jump in the parameters is also possible when there is a change in
entropy or a non-equilibrium region between the two stations.
(1)
(1)
(2)
(2)
Non-equilibrium region
Considering two sections (1) and (2) where equilibrium exists but which may contain nonequilibrium region between them, all the conservation laws apply, and hence
1u1 = 2 u 2
p1 + 1u1 = p 2 + 2 u 2
2
1 2
1 2
h1 + u1 = h2 + u 2
2
2
There is no restriction on the size or details of the dissipative region as long as the reference
sections are outside it. The non-equilibrium region may be idealized by a vanishingly thin region,
across which the flow parameters may jump. The control sections (1) & (2) may be brought
arbitrarily close to it. Such a discontinuity is called shock wave. A real fluid cannot have an actual
discontinuity and this is just an idealization of the very high gradients that actually occur in a
shock wave. These severe gradients produce viscous stress and heat transfer (non-equilibrium
conditions) inside the shock.
3
The three conservation statements given above are the general equations for a steady, inviscid
adiabatic flow and hence, for a normal shock. For a thermally and calorically perfect gas, the
equations can be solved explicitly in terms of the Mach number M1. The results always apply
locally to the conditions on either side of a shock, provided it is normal to the streamline.
Both sides of the momentum equation divided by the appropriate side of the continuity equation
p1 + 1u1
p + 2u2
= 2
1u1
2u2
2
p
p
a
a
u1 u 2 = 2 1 = 2 1
2 u 2 1u1 u 2 u1
or
Using the energy equation for a perfect gas, in terms of sound speed
u12 a12
u2 a2
1 + 1 2
+
= 2 + 2 =
a
2 1 2 1 2 1
u1u 2 = a 2
This is known as Prandtl or Meyer relation.
M1
M <> = 1 corresponds to M <> = 1 . Thus, the velocity change across a normal shock must be from
supersonic to subsonic or vice-versa. There is nothing in the relation to exclude possibility.
However, from the physical ground, it can be anticipated that accelerating jump is not possible,
since increase in velocity is unlikely when dissipative processes are in action.
Using M 2 =
( + 1)M 2
2 + ( 1)M 2
( + 1) M 2 2
2 + ( 1) M 2 2
M2 =
2
If M 1 = 1,
1+
1
2
M1
2
M1
2 + ( 1) M 12
( + 1) M12
M 2 = f (M 1 )
M1 > 1 M 2 < 1
As M 1 increases above 1, the normal shock wave becomes stronger and M 2 becomes
progressively smaller.
As M 1 , M 2 = Lim
1+
1
2
M 1
M 12
M1
1
2
u1 u12
u12
=
= 2 = M 12
u2 u1u2 a
Using the continuity equation
( + 1)M 12
2 u1
2
=
= M1 =
2
1 u 2
2 + ( 1)M 1
p 2 p1 = 1u1 2 u 2 = 1u1 2 u1u 2 = 1u1 (u1 u 2 )
2
p 2 p1 1u1
=
p1
p1
Using
p1
u2
1
u1
a1
2
1 = , and relation for u1
p 2 p1 p1
2
2
=
=
M1 1
p1
p1
+1
p1
p2
p1
p1
= 1+
2
( M 2 1)
+1 1
T2 p 2 1
=
T1
p1 2
T
a
h
2( 1) M 1 + 1
2
M1 1 = 2
Also, 2 = 22 = 1 +
2
2
T1 a1
h1
( + 1) M 1
2
5
The limiting values for air with = 1.4
1
= 0.378
2
Lim M 2 =
M 1
Lim
2
1
+1
1
Lim
p2
p1
T2
M 1 T
1
M 1
M 1
Lim
= 6
All the ratios are functions of upstream Mach number for a calorically perfect gas.
The change in entropy across the shock is given by
1
p2 2 1
T2
p2
s2 s1 = C p ln R ln
= R ln
p1 1
T1
p1
or
s2 s1
2
M 12 1)
= ln 1 +
(
R
+1
1
1
( + 1) M 12 1
2
( 1) M 1 + 2
Since s2 s1 in an adiabatic process, the condition can only be satisfied if M 1 1 . If M 1 < 1 , then
s2 s1 < 0 , which is impossible according to second law. Hence, supersonic to subsonic jump is
the only possibility across a normal shock.
For a weak shock, M 12 1 is small and the change in entropy can be approximated as
M 12 1
s2 s1
2
+ 1 p1
2
R
3
12 2 p1
( + 1)
p1
. A small change
p1
in pressure, that is associated with first order change in velocity, density and temperature,
produces third order change in entropy. A weak shock produces nearly isentropic change of state.
Since the flow across a shock is adiabatic, T02 = T01 .
po1
p
po1
s2 s1
= ln o1 = ln
= ln 1 +
R
po 2
po1
po1 + po1
6
The expression for the total pressure ratio is obtained from that of s
po 2
2
M 12 1
= 1 +
po1 + 1
po1
p o1
<< 1
( + 1) M 12 1
2
( 1) M 1 + 2
hence
M 12 1
s2 s1 po1
2
=
2
3
R
po1
( + 1)
Change in entropy is directly proportional to change in total pressure. Hence, change in total
pressure is also of 3rd order in shock strength.
The shock jump relations (Rankine-Hugoniot relations) are often expressed in terms of pressure
jump instead of upstream Mach number.
Module2:
One-dimensional gas dynamics
Lecture6:
Governing Equations (Contd.)
2
Moving Shock Wave Problem
The normal shock wave relations are established above assuming the shock to be stationary and
the fluid flows through it with speed u1. Alternatively, the shock may be taken as propagating
through the fluid with speed u1.
ub
(2)
us
(a)
u(b)
p
(1)
u2
us
u1
u2
u1
Consider the fluid in front of the shock wave is at rest and the wave is moving into it with the
speed u s . The shock causes the fluid behind the wave to move with speed ub . The problem is
solved by using a transformation of the reference system. With respect to an observer moving
with the shock the fluid ahead of it is moving with speed u1 (in opposite direction) and the fluid
The jump relations across the shock may be rewritten in terms of u s and ub by using the
transformation. The Mach number of the shock is M 1 =
us
a1
a12 = p
p
For practical usefulness, all other quantities are expressed in terms of the pressure ratio 2
p1
1 + 1 p2
+
us = M 1a1 = a1
2
p1
The density ratio and temperature ratio are given by the Rankine-Hugoniot relations.
1+
+ 1 p2
1 p1
u
2
=
= 1
+ 1 p2
u2
1
+
1 p1
+ 1 p2
+
T2 p 2 1 p1
=
+ 1 p2
T1 p1
1+
1 p1
The fluid velocity behind the shock is ub = u1 u2 = us 1 u2
u1
+ 1
a p
ub = 1 2 1
p1 p2 + 1
p1 + 1
The total quantities in the two systems are not the same. It should be noted, for example, that
h0 a = ha h01 ,
However, h02 = h2 +
1
h0b = hb + ub2 h02
2
and h0b h0 a .
1 2
1
u2 = h1 + u12 = h01.
2
2
A weak shock is the one for which the normalized pressure jump is very small,
p p 2 p1
=
<< 1
p1
p1
The other disturbances are then also small and can be obtained by expanding the above
equations in series and retaining only the first order terms in
p
p1
4
This gives
1 p ub
p1 a1
+ 1 p
T 1 p
, us a1 1 +
T1
p1
4 p1
p2
p1
2
+1
1
1
T2
1 p2
T1
+ 1 p1
1
+ 1 p2 2
us a1
2 p1
ub a1
p2
2
( + 1) p1
Disturbances created in a fluid by a moving body are propagated or communicated to other parts
of the fluid. The motion of the disturbances relative to the fluid is called wave motion, and the
speed of propagation is called the wave speed. Through this mechanism, various parts of a
moving body interact with the fluid and with each other and the forces on the body are
established. The problem is unsteady and appropriate equations need to be solved. Considering
adiabatic, non-viscous motion in a constant area duct,
Continuity equation:
+
+u
=0
t
x
x
Euler/Momentum equation:
u 1
u
+u +
=0
t
x x
Since friction is neglected and external heat addition is excluded, the isentropic conditions exist
p
=
p = p( ) , for a perfect gas
p1 1
p p
=
= a2
x x
x
The disturbances or perturbations are defined relative to the fluid that is at rest, with
u = 0, = 1 . Assume the perturbed values are u and . The perturbations are not
necessarily small.
= 1 (1 + s )
Define
( 1 )
1 is called condensation.
s
u
s
u
+ 1 + s + 1u
=0
t
x
x
x
u
u
a 2 s
+u
+
=0
t
x 1 + s x
p
For a perfect gas
T
= = (1 + s )
p1
1
= (1 + s )
T1
These equations are exact for frictionless, non-conducting motion, but not easily integrable as
they are non-linear. The equations can be linearized through a small disturbance assumption.
Assuming s << 1,
The terms u
u
u
u
<<
s
can be neglected
x
x
x
s
u
and u
are of the same order and may be neglected.
x
x
d2 p
dp p 2 p
2
a =
+
( 1 ) = a1 + 1s 2
d 1 2 1
d 1
2
a 2 s
2 s
can be approximated to a1
Hence,
1 + s x
x
NPTEL IIT Kharagpur: Prof. K.P. Sinhamahapatra, Dept. of Aerospace Engineering
6
The approximate linearized equations are then
s u
+
=0
t x
u
s
+ a12
=0
t
x
These equations are called acoustic equations because the disturbances due to sound wave are
very small.
The isentropic relations for a perfect gas can be approximated as
p
= 1+ s
p1
T
= 1 + ( 1)s
T1
Differentiating the linearized continuity and momentum equations with respect to t and x
respectively
2 s 2u
+
=0
2
t
xt
2
2
u
s
+ a1 2 = 0
xt
x
Similarly,
2s
2 s
a1
=0
2
t
x 2
2
2u
2 u
a
=0
1
t 2
x 2
Both the disturbances (u, s ) satisfy the wave equation. The disturbance propagates with a definite
signal velocity or wave velocity a1 .
The general solutions of the equations are
s = F ( x a1t ) + G ( x + a1t )
u = f ( x a1t ) + g ( x + a1t ) ,
f = a1 F
g = a1G
s = F ( x a1t )
This represents a disturbance or wave which at time t = 0 had the shape
s = F (x )
7
and which at time t has exactly the same shape, but with corresponding points displaced a
distance a1t to the right. The velocity of each point in the wave and hence of the wave is a1 .
t
dx
= a1
dt
t2
t1
x
s = F ( x)
A wave in which the propagation velocity is in one direction is called a simple wave. Similarly, the
wave described by s = G( x + a1t ) is a simple wave propagating to the left with speed a1 .
t
dx
=
dt
dx
= a1
dt
t
t
x
s = G ( x)
The lines in the x t plane which trace the progress of the waves, i.e., the lines with slope
dx
dt
called the speed of sound or acoustic speed. The result is applicable to disturbances in which
velocity and temperature gradients are very small and u
a1
p
a 2 =
s
8
The amplitude of ordinary audible sound is small and the local production of entropy is negligible.
Friction and local entropy production is negligible for computing the speed of ordinary sound, but
the cumulative effect on the amplitude is not negligible.
The quantity a 2 provides a pressure-density relation and eliminates pressure from the momentum
p
= a2
. When non-isentropic processes are present pressure depends on
x
x
equation
p
p s
= a2
+
x
x s x
a 2 can be evaluated from the equation of state. For a perfect gas a 2 =
= RT
Module2:
One-dimensional gas dynamics
Lecture7:
Governing Equations (Contd.)
2
Pressure and particle velocity in a sound wave
The pressure disturbance accompanying the density wave for a perfect gas is
p
= 1+ s
p1
p p1
= s
p1
Pressure wave has the same shape as the density wave, differing by a constant factor .
As the wave progresses through the fluid, the pressure disturbance sets the fluid in motion, giving
it a velocity u which is called the particle velocity and, in general, is much smaller than the wave
speed a1 .
A simple wave s = F ( x a1t ) , propagating to the right, produces a velocity disturbance
u = a1 F ( x a1t ) = a1 s
In a left ward propagating wave u = a1G ( x + a1t ) = a1 s . The various parts of the wave are called
condensation and rarefaction depending on whether the density is higher or lower than the
undisturbed density 1 .
The effect that the wave produces on the fluid depends on the gradient of this density and
pressure distribution and on the direction of motion of the wave. Thus the portion of the wave that
increases the density as it passes is called a compression and that which decreases the density is
called an expansion. The corresponding distributions of particle velocity are
u = a1 s
for left and right propagating waves respectively. It may be seen that a compression accelerates
the fluid in the direction of wave motion, whereas an expansion decelerates it. The non-simple
wave is a superposition of two simple waves and the relation between particle velocity and density
is
u
F G
.
=
a1 s F + G
In the limit of vanishingly small disturbances the perturbation quantities may be written in
differential form with u and s replaced by du and d
3
d
1
p1
d = 1a1du
dp =
dp
d
1
=
p1
1
du = a1
Shock tube is a simple device consisting of a tube that is divided by a membrane or diaphragm
into two chambers in which pressures are different. A wave motion is set up when the diaphragm
is suddenly removed or broken. If the pressure difference is very small, the wave motion may be
approximated as isentropic and can described by the acoustic equations. The shock tube in such
a case is called acoustic or linearized.
At t = 0, immediately after the membrane is removed the wave has a shape of step distribution.
The particle velocity at this instant is zero everywhere. The wave at t = 0 is described by
s ( x, 0 ) = F ( x ) + G ( x ) = s0 ( x )
s
= 4
0
( x > 0)
x<0
u (x,0) = a1 F ( x ) a1G ( x ) = 0
1
Simultaneous solution gives F ( x ) = G ( x ) = s0 ( x )
2
1
s
= 2 4
0
x>0
x<0
s ( x, t ) =
1
1
s0 ( x a1t ) + s0 ( x + a1t )
2
2
s4
1
= s4
2
0
x > a1t
a1t < x < a1t
x < a1t
u ( x, t ) =
1
1
a1s0 ( x a1t ) a1s0 ( x + a1t )
2
2
0
1
= a1s4
2
0
a1
s = s4
1
s2 = s4
2
s =0
x > a1t
a1
t = t1
A compression wave is propagated to low pressure side and an expansion wave of equal strength
to the high-pressure side.
The simple acoustic waves as seen above have constant wave velocity and permanent shape.
These properties are due to the linearization of the equations achieved through the assumption of
infinitesimal amplitudes and gradients. When such an assumption is not possible, the conditions
at a given point in the wave cannot be approximated by those in the undisturbed fluid. The wave
velocity varies from point to point and the simple wave becomes distorted as it propagates. To
describe such finite disturbance, the complete non-linear equations need to be solved.
Consider a right moving plane compression wave. At two adjacent points, as shown in the figure
below, the fluid properties differ in magnitude by du , dp, d , da, etc. The respective parts of the
5
wave passing through these same points differ in wave speed by the amount dc . Since a finite
wave may be thought of as a succession of infinitesimal pressure pulses, each element of the
wave may be analyzed as an acoustic wave. As long as the velocity and temperature gradients
are moderate, the viscous and heat conduction effects are negligible. Hence, each part of the
wave travels at the local speed of sound with respect to the fluid in which it is propagating. The
propagation velocity of a part of the wave with respect to fixed coordinates is then,
c=u+a
The propagation velocity of an adjacent part of the wave is
c + dc = u + du + a + da
dc = du + da
dc du da
=
+
dp dp dp
du
1
,
=
dp a
dc da 1
=
+
dp dp a
6
Since the entire fluid was originally at rest with uniform pressure and temperature and each
particle of fluid undergoes isentropic changes, the increments in pressure and density between
adjacent fluid-particles obey
a2 =
dp
da d dp d d dp
2a
=
=
d
dp dp d dp d d
d dp
1 d d
dc
=
1+
dp a 2 dp
d
d
d d
d
1 d
=
= 2
= 2
d d d
d
d
d2p
d
dc
s
=
2a dp
dp
d s
2
( )
( dp d )
2
of d p
dp
concave downwards. Hence, higher-pressure parts of the wave overtake the lower-pressure parts
when
dc
is positive. Consequently, a compression wave steepens as it progresses and an
dp
dc
is negative.
dp
(i)
d2p
> 0 , or the
Compression waves steepen and expansion waves flatten when
2
d s
isentrope is concave upward. This is the usual case for all real fluids.
(ii)
d2 p
< 0, or the
2
d s
For a gas with straight line isentrope waves of finite amplitude propagate through the gas with
unchanged shape.
d 2 p ( + 1) p
=
> 0 . Hence, compression waves
2
2
d s
From the point of view of an observer moving with the local particle velocity the acoustic theory
applies locally. Relative to an observer moving with the local fluid velocity, the wave at that point
dp
propagates with the local acoustic speed a =
1
2
reference in the undisturbed fluid, it propagates with the speed c = a + u . Considering both left
and right moving waves, the local wave speed at any point is given by
c = a + u
The wave speed is no longer constant since a is a variable and
To evaluate these in terms of the density, the acoustic theory is applied locally. Using the
a = a1
1
p1
to eliminate p from a 2 =
1
2
d
du = a
This becomes integrable if a is replaced by above
8
1
2a1 2
2
1 =
(a a1 )
u = a
=
1
1 1
1
or
a = a1
c = a1 +
+1
2
+1 2
1
or c = a1 1 +
1 1
where a1 is the speed of sound in the undisturbed fluid and 1 is the density of undisturbed fluid.
Module2:
One-dimensional gas dynamics
Lecture5:
Governing Equations (Contd.)
2
Propagation of Finite waves
Consider the propagation of a simple wave of finite amplitude. Assume that the initial density
distribution is as shown in the figure (t = t 0 = 0 )
1
+ 1 2
1
For the right ward propagating wave c = a1 1 +
1 1
a1
in regions rarefaction. Thus the wave distorts as it propagates, the regions of higher condensation
tending to overtake those in regions of lower condensation. In regions of higher condensation, the
characteristic lines are inclined more, since the slope is inversely proportional to the wave speed.
t
t3
t2
t1
t0 = 0
~
s ( x,0)
In terms of the compression and expansion regions the net effect is to steepen compression
regions and to flatten expansion regions in which the characteristic lines converge and diverge
respectively. In a compression region, the characteristics lines would eventually cross leading to
the situation t = t3 . But this would be physically impossible, for it implies three values of density at
a given point. Actually, well before this happens, the velocity and temperature gradients in the
compression regions become so large that friction and heat transfer effects become important.
These have a diffusive action which counteracts the steepening tendency. The two opposing
effects achieve a balance and the compression portion of the wave become stationary, in the
sense that it propagates without further distortion. It is then a shock wave.
3
In compression regions, the isentropic relations are valid until friction and heat transfer become
important. When a stationary balance between the diffusive and steepening (non-linear) terms
has been reached, the conditions across the wave front are given by the shock wave relations.
The intermediate unsteady, non-isentropic states can be treated only with the full unsteady
equations including viscous and heat transfer terms.
An expansion wave always remains isentropic as it tends to flatten and so reduce the velocity and
temperature gradients further. It never achieves stationary condition, corresponding to the fact
that there are no expansion shocks.
Consider a duct containing fluid enclosed by a piston. If the piston is withdrawn an expansion
wave is produced. If the piston starts impulsively, with speed u p , the distribution of particle
velocity in the first instant is a step. However, the expansion wave begins to flatten as soon as the
wave starts propagating. At some later time t1 the particle velocity has a linear distribution and
the pressure has a corresponding distribution.
p4
p3
u=0
u3 = - |up|
a4
up
(3)
t
(4)
+1
x = c3 t = a 4
u p t
2
Piston path
x = -|up| t
(3)
x = a4 t
(4)
Expansion path
x
The front of any isentropic wave propagates at the speed of sound of the undisturbed fluid. Thus,
the front of the wave propagates with speed a4 into the undisturbed fluid that is in the opposite
direction of the piston and fluid motion. The wave speed in the portions of the wave behind the
front is given by
c = a4 +
+1
2
Wave speed c decreases continuously through the wave since u < 0 . The fan of straight lines are
lines of constant c and thus of constant u and . These lines are the characteristics. With
increasing time the fan becomes wider and the wave becomes flatter and the gradients of
velocity, density, temperature become smaller. Thus the wave remains isentropic. The terminating
characteristic is given by
whether a4 <>
+1
2
u p . Between the terminating characteristics and the piston, the fluid properties
have the uniform values 3 , p3 , T3 , etc. For a perfect gas they are given in terms of the isentropic
relations
3 1 u p
= 1
2 a4
4
p3 1 u p
= 1
p4
2 a4
This pressure ratio defines the strength of the expansion wave. The maximum expansion is
obtained when u p =
2a4
giving 3 = T3 = 0 . This implies that all the fluid energy is converted into
1
kinetic energy of flow. If the piston velocity is higher than this limiting value, it produces no further
5
effect on the flow. The wave produced by an impulsive withdrawal of piston is called a centered
expansion wave after the fan like set of characteristic lines in the x t plane.
p4
p1
The diaphragm at x = 0 separates the low pressure chamber (1) from the high pressure chamber
(4). The chambers are also called respectively, the expansion and compression chamber. The
characteristic parameter of a shock tube is the diaphragm pressure ratio
p4
p1
. The two
chambers may be at different temperature T1 and T4 and may contain different gases with gas
constants R1 and R4 .
When the diaphragm is burst ( t = 0 ) , the pressure distribution is ideally a step. This splits into a
shock wave which propagates into the expansion chamber with speed u s and an expansion wave
which propagates into the compression chamber with the speed a4 at its front. The condition of
the fluid which is traversed by the shock wave is denoted by (2) and that of the fluid traversed by
the expansion wave is denoted by (3). The interface between these two regions is called the
contact surface. It marks the boundary between the fluids which were initially on either side of the
diaphragm. The region behind the shock is different from the one behind the expansion and they
have different values of entropy. Neglecting diffusion, the fluids in the two regions do not mix, but
are permanently separated by the contact surface.
t = t1
us
u2
p2
p1
T2
T1
T4
T3
On either side of the contact surface, the temperature and density may be different, but it is
necessary that the pressure and fluid velocity be the same,
p 2 = p3 ,
u 2 = u3
Thus u 2 = u 3 must be the velocity of the contact surface. Using shock wave and expansion wave
relations
u2 = a1 2 1
p1
2 1
( 1 + 1) p2 p + ( 1 1)
1
( 4 1)
2 4
2a 4 p3
1
u3 =
4 1
p4
7
Since u 2 = u 3 and p 2 = p3 , the basic shock tube relation is
4 1) a1 a p2 p 1
(
p4 p2
4
1
1
=
p1 p1
p
2 1 2 1 + ( 1 + 1) 2 1
p1
p2
p1
2 4
( 4 1)
p4
p1
. The
p2
p3 p3 p1
p1
=
=
since p3 = p2
p4 p1 p4 p4
p1
The values of u and p calculated across the shock and expansion are identical. However, and
T are usually different. The temperature T3 behind the expansion wave is given by the isentropic
relation
T3 p3
=
T4 p 4
( 4 1)
p2
p1
=
p 4
p1
( 4 1)
1 1
1 +1
T2
=
1
T1
1+ 1
1 +1
1+
p2
p1
p1
p2
The duration of the flow is limited by the lengths of the expansion and compression chambers,
since the shock wave and expansion wave reflect from the ends of the chambers and eventually
interact with each other.
Module3:
Waves in Supersonic Flow
Lecture9:
Waves in Supersonic Flow
2
The motion of a body in a fluid at rest creates disturbance in the fluid. The disturbances, in general,
may not be small. The disturbances in the fluid close to the body are transmitted to other parts of the
body and also to the other parts of the fluid through propagation of the waves. The wave motion is
compatible with the motion of the body. This wave motion determines the pressures on the body as
well as the complete flow field around the body. When the flow is subsonic, it is not essential to
consider the wave motion. Particularly, if the motion is steady it is easier to study the motion from a
reference system where the body is at rest and the fluid flows over it. However, if the relative wind is
supersonic, the waves can not propagate ahead of the immediate vicinity of the body. Consequently,
the wave system travels with the body. Thus, the wave system is stationary in the reference system
that moves with the body. Limited upstream influence allows the flow to be analyzed or constructed
step by step.
u1 . The flow
velocity vector ahead of the shock has an inclination to the shock given by = tan 1
associated with the tangential component does not cross the shock and is not affected by the shock
and, hence 1 = 2 = . The normal component undergoes a jump and u2 u1 . Hence, the inclination
of the flow after the shock is different from that ahead. The flow turns abruptly at the shock.
Since u2 < u1 , the turn is always toward the shock and the angle of deflection is positive.
3
w2
2 =1
w2
u1
w1
u2
1
w1
Since, the superposition of a uniform velocity ( 1 = 2 = ) does not affect the static pressures and
other static parameters, the relations for the conditions before and after the shock can be obtained
from normal shock relations. The modification is due to initial Mach number since M 1 =
u1 = w1 sin
u1
= M 1 sin
a1
In the normal shock relations
u1
a1
or
M 1 is to be replaced by M 1 sin
( + 1)M 1 2 sin 2
2
=
1 ( 1)M 1 2 sin 2 + 2
p2
2
M 12 sin 2 1
= 1+
p1
+1
w1
a1
T2 a 2
2( 1) M 1 sin 2 1
2
2
= 2 = 1+
M
sin
+1
1
2
2
T1 a1
( + 1) M 1 sin 2
2
s2 s1
2
= ln 1 +
M 12 sin 2 1)
(
R
+1
( 1)
( + 1) M 12 sin 2
2
2
+
M
1
sin
2
(
)
1
( 1)
= ln
p01
p02
The ratios depend only on the normal component of the velocity and this component has to be
supersonic i.e. M 1 sin 1 . This sets a minimum wave inclination for a given Mach number. The
maximum wave inclination is that for a normal shock =
sin 1
M1
2
M 2 2 sin 2 ( ) =
Also,
tan =
u1
1+
1
2
a2
and
u2
a2
M 12 sin 2
M 12 sin 2
, tan ( ) =
w2
u2
1
2
u2
tan ( ) u2 1 ( 1) M 12 sin 2 + 2
= =
=
tan
u1 2
( + 1) M12 sin 2
The trigonometric relation can be shown to be
= M 2 sin ( ) .
M 12 sin 2 1
tan = 2 cot
M 12 ( + cos 2 ) + 2
= 0 at =
1 1
sin
and
M1
max
M2 = 1
Within this range is positive and must therefore have a maximum. For each value of M 1 there is a
maximum value of . For < max , each value of and M corresponds to two possible solutions,
having different values of . The larger value of gives a stronger shock. In the solution with strong
shock the flow becomes subsonic. With weak shock the flow remains supersonic except for a small
range of value of slightly smaller than max .
or
1
M 1 sin 2
2
M 12 sin 2 1 =
+ 1 tan( ) 1
2
tan
2
+1
2
M 12
sin sin
cos ( )
+1 2
M 1 tan
M 12 sin 2 1
2
Module3:
Waves in Supersonic Flow
Lecture10:
Waves in Supersonic Flow (Contd.)S
2
Supersonic Flow over a wedge
Any streamline in inviscid flow can be replaced by a solid boundary. Thus the oblique shock flow
provides the solution to supersonic flow in a corner. For a given values of M 1 and , the values of
and M 2 are determined. Using symmetry the flow over a wedge of nose angle 2 is also obtained.
The flow on each side of the wedge is determined only by the inclination of the surface on that side.
Thus, the wedge need not be symmetric. When the shock waves are attached to the nose, the upper
and lower surfaces are independent since there is no influence on the flow upstream of the waves.
M2
M2
M1
M1
M1
M1
M2
Mach Lines
Assuming that downstream flow remain supersonic ( M 2 > 1 ), the wave angle decreases with
decrease in the wedge angle. When decreases to zero, decreases to the limiting value , given
by
M 1 sin 2 1 = 0
2
or
= sin 1
The jump in the flow quantities is then zero and, hence the strength of the wave is zero. The flow is
continuous without any disturbance. There is nothing unique about the point where this wave
originates; it might be any point in the flow. The angle is simply a characteristic angle associated
3
with M 1 . It is called the Mach angle. The lines of inclination which may be drawn at any point in
the flow field are called Mach lines or Mach waves.
()
()
M1
M1
M1
(+)
If the flow is nonuniform varies with M and the Mach lines are curved. At any point P in a 2-D flow
field, there are always two lines which intersect the streamline at the angle . In 3-D flow, the Mach
lines or characteristics define a conical surface with vertex at P. A 2-D supersonic flow is always
associated with two families of Mach lines denoted by the labels (+) and (). Those in the (+) set run
to the right of the streamlines and those in the () set run to the left. They are also called
characteristics from the mathematical theory of hyperbolic PDEs. These are analogous to the two
families of characteristics that trace the propagation of 1-D waves in the x-t plane. Like the
characteristics in the x-t plane, Mach lines have a distinguished direction, the direction of flow or the
direction of increasing time. This is related to the fact that there is no upstream influence in
supersonic flow.
+1 2
2
M 1 sin 2 1
M 1 tan
2
M sin 1
2
1
+1
2
M 12
M 1
2
1
, as tan tan =
1
2
M1 1
p 2 p1 p
=
p1
p
M 12
2
M1 1
The changes in other flow quantities are also proportional to the deflection angle ' ' . The change of
entropy is proportional to the third power of the shock strength and hence to third power of deflection
angle
s 3
The difference between the wave angle and the Mach angle , to first order accuracy, can be
found as follows,
Let = + , <<
Hence,
By definition
sin = 1
Hence,
M 1 sin 1 + M 12 1
, cot = M 1 1
2
M1
M 12 sin 2 1 + 2 M 12 1
M sin 1 2 M 1
2
1
or
+ 1 M 12
4 M 12 1
2
1
+1
2
M 12
M 1
2
1
5
Hence, for a finite deflection angle , the direction of the wave differs from the Mach direction by an
amount , which is of the same order as .
The change in flow speed can be obtained as
u2
2
2
2
w2
u2 +
= 2
=
2
2
w1
u1 +
u1
M 12 1
2
cos = 1 sin =
1
2
2
M1
M 1 1
Now,
+1
tan 2 ( ) + 1
cos 2
=
=
2
tan 2 + 1
cos 2 ( )
+1
w2
1
w1
M 12 1
or
=
2
w1
M1 1
Module3:
Waves in Supersonic Flow
Lecture11:
Waves in Supersonic Flow (Contd.)
2
Supersonic compression by turning
A shock wave passing through a fluid increases the pressure and density of the fluid. Hence, shock
waves can be used to compress a flow. A simple method for compressing a supersonic flow is to turn
it through an oblique shock by deflecting the wall through an angle . The turn may be subdivided
into several segments which make smaller corners of angle so that compression occurs through
successive weaker oblique shocks. These shocks divide the field near the wall into segments of
uniform flow. Away from the wall the shocks tend to intersect each other since they are convergent. In
the near wall region each segment of the flow is independent of the next one and may be constructed
step by step proceeding downstream. This property of limited upstream influence exists as long as
the deflection does not become so great that the flow becomes subsonic.
For each wave in the multiple shock p and s ( ) .
3
pk p1 n ~
sk s1 n ( ) ~ ( n )( ) ~ ( )
3
Thus, when the compression is achieved through a large number of weak shocks, the entropy
increase can be reduced significantly compared to a single shock giving the same net deflection. It
decreases as
1
. By continuing the process of subdivision, the segments can be made vanishingly
n2
small ( 0 ), and in the limit, the smooth turn or isentropic compression is obtained.
When the shocks become vanishingly weak, they are almost straight Mach lines. Each segment of
uniform flow becomes vanishingly narrow and finally coincides with a Mach line. Thus, the flow
inclination and Mach number are constant on each Mach line. Thus, in the limit of smooth flow, the
velocities and flow inclination are continuous, but their derivatives may still be discontinuous. The
approximate expression for the change of speed across a very weak shock
=
2
w
M1 1
becomes the differential equation
dw
d
=
= (M )
2
w
M1 1
M2
M1
M1
M2
Due to the convergence of the Mach lines, the change form M 1 to M 2 on the streamline b occurs in a
shorter distance than on the streamline a. Hence, the gradients of velocity and temperature on b are
higher than those on a. An intersection of Mach lines would imply an infinitely high gradient for there
would be two values of M at one point. However, this cannot occur since in the region where Mach
lines converge and the gradients become very high the conditions are no longer isentropic. Before the
Mach lines cross a shock wave is developed. Far from the corner, there would be a simple oblique
shock for M 1 and . The convergence of Mach lines in a compression is a typical nonlinear effect:
decreasing Mach number and increasing flow inclination both tend to make successive Mach lines
steeper.
If a wall is placed along one of the streamlines, say b, where the gradients are still small enough for
the flow to be isentropic; then an isentropic compression in a curved channel is obtained. Since this
flow is isentropic, it may be reversed without violating the second law of thermodynamics.
Flow round a concave turn, that is turns in which the wall is deflected in to the flow, undergoes
compression through shock wave/Mach lines. Expansion takes place in a flow over a convex corner.
In this case a turn through a single oblique wave is not possible.
u1
w2
w1
Since 1 = 2 , u2 must be greater than u1 decrease in entropy. Hence, expansion shocks are not
possible.
The non-linear mechanism that steepens a compression produces the opposite effect in expansion.
Instead of being convergent, the Mach lines are divergent.
M1
M2
1
1
2
M1
M2
d = M 2 1
dw
w
+ K = M 2 1
Now
w = aM and
dw
= (M )
w
a0 2
1 2
=
1
+
M
2
a2
dw dM da dM
1
=
+
=
w
M
a
M 1 2
M
1+
2
Hence
M 2 1 dM
(M ) =
1 2 M
1+
M
2
=
+ 1 1 1 2
tan
M 1) tan 1 M 2 1
(
1
+1
This function is known as the Prandtl-Meyer function. The constant of integration is chosen arbitrarily
so that = 0 corresponds to M = 1 . The corresponding values of the flow properties are obtained
from isentropic relations.
A supersonic Mach number M is always associated with a definite value of the function . As M
varies from 1 to , increases monotonically form 0 to max , where
max =
+1
1 2.27685 rad = 130.454D for = 1.4
2 1
7
value of then gives the corresponding value of M . Usually, the value of 1 is set to 0, since only
the deflection matters.
M , ,
M 1 , 1
1
= 1 1
For compression
M 1 , 1 ,1
M , ,
= 1 + 1
For expansion
Module3:
Waves in Supersonic Flow
Lecture12:
Waves in Supersonic Flow (Contd.)
Simple wave
Simple wave ()
3
The deflections produced by the two shocks are equal in magnitude but the pressure ratios are not,
since M 2 < M 1 . The strength of the reflection is defined by the overall pressure ratio which equals the
product of the individual shock strengths.
p3 p3 p2
=
p1 p2 p1
M2
M1
M3
p3
p2
p1
On streamline
p2
p1
On wall
Usually, the reflection is not specular, i.e., the inclination of the reflected shock is not the same as
the inclination of the incident shock. The shock angles are different since both Mach number and
flow inclination ahead of the second shock are smaller than those ahead of the first shock. The two
effects are opposite and the result depends on the particular values of M 1 and . An explicit relation
cannot be found but the values can be found easily.
The wall streamline, in the reflection case, may also be identified with the central streamline of the
symmetric flow in the intersection of two shocks of equal strength but of opposite families. The shocks
pass through each other but are slightly bent in the process. The flow downstream of the shock
system is parallel to the initial flow.
M1
M2
M3
Flow symmetric about the
M2
M3
central streamline
If the intersecting shocks are of unequal strengths, the flow experiences different changes in
traversing the shock wave system. The streamline through the intersection point divides the flow into
two portions. The two portions have the same pressure and the same flow direction. The direction is
not necessarily that of the free stream. These two requirements determine the final direction and the
final pressure p3 . All other parameters are then determined, but they do not have the same values on
the two sides of the dividing streamline. A slip stream or shear layer develops since the magnitudes
of the velocity on either side of it are different. It is also called a contact surface, because the
temperature and density on either side are different. These differences are related to the net entropy
changes experienced by the fluid on the two sides of the intersection.
M3
M2
p3
M1
p3
, M2
Dividing streamline
or slip stream
M3
Module3:
Waves in Supersonic Flow
Lecture13:
Waves in Supersonic Flow(Contd.)
d
Slip stream
o
Weak reflection
e
a
The flow on either side of the intersection point, o, experiences different entropy changes and a
slipstream is produced. An additional wave oe , of the opposite family, is needed to equalize the
pressures on the two sides of the slipstream. This may be either a compression or an expansion
wave depending on the particular configuration and Mach number. However, it is very much weaker
than the primary waves. If the second shock bo is much weaker than the first one ao , then oe is
usually a compression. In this case the second shock is partly transmitted along oc , thus
augmenting the first one and partly reflected along oe .
In the interaction of an expansion wave with a shock wave of the same family, the main effect is an
attenuation of the shock, but there is also a partial reflection of the expansion along Mach lines of the
opposite family. These reflected waves are always very mach weaker than the primary ones and may
be neglected in all but the strongest interactions. Instead of the single slipstream, there is a whole
region of vorticity, that is, an entropy field downstream of the interaction.
Weak reflections
Detached shocks
For a given supersonic stream if the wall deflection is > max , the flow cannot negotiate the turn
through an attached oblique shock. The observed flow configurations are, as example,
c
b
M>1
M<1
M1
M=M1
b
=max
4
d
/2
M>1
M<1
The flow is compressed through a curved shock, detached from the body and stands at some
distance ahead of it. The shape of the shock and the detachment distance depend on the geometry
and flow Mach number. On the central streamline, the shock is normal and the flow behind the shock
is subsonic. On the nearby streamlines, the shock is nearly normal and the flow is compressed to
subsonic conditions. Further out, the shock becomes weaker and less steep, approaching
asymptotically to the Mach angle. Thus conditions along the detached shock wave contain the whole
range of the oblique shock solution for the given Mach number. In such configurations, shock
inclination corresponding to strong solution is found. When the flow behind the shock is subsonic, the
shock is no longer independent of the downstream conditions. A change in geometry or pressure in
the subsonic portion affects the entire flow up to the shock and the shock needs to adjust itself to the
new conditions. In the case of a blunt-nosed body, the shock wave is detached at all Mach numbers.
A wedge of half-angle > max is a blunt-nosed body so far as the oncoming flow is concerned.
The sequence of events in the flow over a wedge with after-body with decreasing Mach number is as
follows: (1) when M 1 is sufficiently high the shock wave is attached to the nose, the straight portion is
5
independent of the shoulder and after-body. The shock angle increases as M 1 decreases. (2) At a
certain reduced Mach number, the flow after the shock becomes subsonic. The shoulder now affects
the whole shock, which may be curved, even though still attached. These conditions correspond to
the region between the lines M 2 = 1 and = max in the M relationship. (3) At the Mach
number corresponding to max , the shock wave starts to detach. This is called detachment Mach
number. (4) With further decrease of M 1 , the detached shock moves upstream of the nose.
A similar sequence of events occurs in flow over a cone with cylindrical after-body. The detachment
Mach numbers are lower than for wedges.
Module3:
Waves in Supersonic Flow
Lecture14:
Waves in Supersonic Flow
(Contd.)
Mach Reflection:
The appearance of subsonic regions in the flow complicates the problem. The complications are also
encountered in shock reflections, when they are too strong to give the simple or regular reflections. If
M 2 after the incident shock is lower than the detachment Mach number for , then no solution with
simple oblique wave is possible. A three-shock Mach reflection appears that satisfies the downstream
conditions.
M2>1
M1
M2<1
Slip stream
A normal, or, nearly normal, shock that appears near the wall forms with the incident and reflected
shocks a triple intersection point at O. Due to the difference in entropy on streamlines above and
below the triple point, the streamline that extends downstream from the triple point is a slipstream.
The nearly normal shock is termed shock stem.
The subsonic region behind the shock stem makes a local description of the configuration impossible.
The triple point solution that occurs in a particular problem and the location of the triple point are
determined by the downstream conditions which influence the subsonic part of the flow.
Shock-Expansion Theory
3
Oblique shock wave and simple isentropic wave relations can be used to analyze many 2-D
supersonic flow problems, particularly for geometries with straight segments.
(1) Diamond-section airfoil: Consider a diamond section or double-wedge section airfoil with
semi-vertex angle . Assume the semi-vertex angle to be sufficiently smaller than
max associated with the free stream Mach number M 1 . An attached oblique shock appears at
the nose that compresses the
p2
2
M1, p1
3
4
t
2
M1>1
p1
p4
p3
flow to pressure p2 .On the straight portion, downstream of the shock the flow remains uniform at M 2 .
The centered expansion at the shoulder expands the flow to pressure p3 and the trailing edge shock
recompresses it to nearly the free stream pressure ( p 4 p1 ). Hence, an overpressure acts on the
forward face and an under-pressure acts on the rearward face. Since the pressure on the two straight
portions is unequal, a drag force acts on the airfoil. This drag force is given by
t is the section thickness at the shoulder. Pressure values p 2 and p3 can be obtained using the
shock and expansion relations. This drag exists only in supersonic flow and is called supersonic
wave drag.
4
(2)
Flat plate at incidence: Consider a flat plate of chord c set at an angle of attack . Due to
no upstream influence, the streamlines ahead of the leading edge are straight and the upper surface
flow is independent of lower surface. The flow on the upper surface turns at the nose through a
centered expansion by the angle whereas on the lower side the flow is turned through a
compression angle by an oblique shock. The reverse happens at the trailing edge.
p3
M1>1
M1, p1
2
Slip stream
p1
p2
From the uniform pressures on the two sides, the lift and drag forces are
L = ( p3 p2 ) c cos
D = ( p3 p2 ) c sin
The shock on the lower surface at the nose is weaker than the shock at the trailing edge on the upper
surface (shock at higher Mach number). Hence, the increase in entropy for flow on the two sides is
not same and consequently the streamline from the trailing edge is a slipstream inclined at a small
angle relative to the free stream.
If a larger portion of the flow field is considered, then the shocks and expansion waves will interact.
The expansion fans attenuate the oblique shocks, making them weak and curved. At large distances
they approach asymptotically the free-stream Mach lines. Due to the interaction the waves will reflect.
The reflected wave system will alter the flow field. In shock-expansion theory, the reflected waves are
neglected. For a diamond airfoil and a lifting flat plate, the reflected waves do not intercept the airfoil
at all. Hence, the shock-expansion results are not affected.
Module3:
Waves in Supersonic Flow
Lecture15:
Waves in Supersonic Flow
(Contd.)
The shock-expansion theory provides numerical solutions as long as the shocks are attached. The
method is quite simple and general for computing lift and drag. The results, in general, cannot be
expressed in a concise analytic form. However, for a thin airfoil at small angle of attack all the flow
deflections are small and the shock-expansion theory can be approximated by the approximate
relations for weak shocks and expansions. The basic approximate expression for computing pressure
change is
M2
M 1
2
M 12
M1 1
2
If all pressure changes are referenced to free-stream pressure p1 and all deflections to the free-stream
direction, then
p p1
p1
M 12
M1 1
2
where
stream
Cp =
p p1
2 p p1
=
2
1
2
p
M
1
1
1u1
2
2
M1 1
2
Thus for a flat plate at a small angle of attack , the pressure coefficients on the upper and lower
surfaces are
Cp =
2
M 12 1
CL =
CD
( pl pu ) c cos
1
1u12c
2
( p pu ) c sin
= l
1
1u12 c
2
= C pl C pu cos
4
M 12 1
= C pl C pu sin
4 2
M 12 1
CD 1
2
=
M
1
1
CL 2 4
Cp =
2
M 12 1
p2 p3 =
1
2
u
1 1
M 12 1 2
4
D = ( p2 p3 ) t = ( p2 p3 ) c =
Hence,
CD =
4 2
M 1
2
1
4 2
1u12 c
M 12 1 2
t
M 12 1 c
4
4
For a general airfoil that has thickness, camber and angle of attack, the pressure coefficient on the
upper and lower surfaces are
dyU
2
dyL
, C pL =
2
2
M 1 1 dx
M 1 1 dx
2
C pU =
The profile can be resolved into a symmetric thickness distribution t ( x ) and a camber line of zero
thickness c ( x ) at an incidence
dyU
d d
= c + t
dx
dx
dx
d d
dyL
= c t
dx
dx
dx
Hence,
1
c
L = 1u12 C pL C pU
2
o
4 1u12
2
c + dc
dx =
dx
M 12 1 o
1
c
dy
D = 1u12 C pL L
2
o
dx
dx
dyU
+
C
dx
pU
dx
2 1u12
2
2
c dy
2
dyU
L
=
+ dx
M 12 1 o dx dx
4 1u12
2
2
c
dc dt
2
=
+
+
dx
2
o
dx
dx
M1 1
CL =
CD =
dc
1 c
+
dx
M 12 1 c o
4
dx =
4
M 12 1
2 1 c d 2 d 2
+ c + t dx
2
c 0 dx dx
M 1 1
4
The lift coefficient depends only on angle of attack but the drag coefficient also depends on camber
and thickness. The drag splits into three parts: a drag due to lift, a drag due to thickness and a drag
due to camber.
Module4:
Fiber in Ducts
Lecture 16:
Flow in ducts, (Nozzles and diffusers)
and wind tunnels
2
The flow can be assumed to be one-dimensional, that is, conditions across each section are uniform.
The conditions at any two sections in a steady flow are related by the equation
1u1 A1 = 2 u 2 A2
Using the sonic condition as reference
uA = u A
When the flow is purely subsonic, A is a fictitious area that does not occur in the flow. But, if sonic
and supersonic conditions are attained in the flow, then A = At = area of the actual throat
Since u = a ,
A a o a
=
=
u o u
A
We have
2
=
o + 1
+ 1
2
a
= M
u +1
1
2
o 1 2
M
= 1 +
2
1
1
1 2 1 2
A
M
1 +
= 2
1
2
+
A
M
( +1)
( 1)
Area-pressure relation
1
1
2
p
1 p
p
po
A
u
=
=
1 +1
1
A u
2
1 2 2 1
2 +1
m
p
pu
u=
= u =
A
RT
RT
pM
=
To
1+
1
2
To
T
1
To
M2
m =
m =
m To 1
A p
W
M 1+
1
2
M2
m
=
A
po
R
To
M
1 2
M
1 +
2
+1
2 ( 1)
Hence, for a given Mach number, the flow rate is proportional to the stagnation pressure and
inversely proportional to the square root of stagnation temperature.
m To
po
is used as a non-
m
2
m
= =
A
R
1
+
A
max
+1
1
po
To
Hence, for a given gas, the maximum flow per unit area depends on
po
To
passage, the maximum flow that can pass is relatively large for gases of high molecular weight and
small for gases of low molecular weight.
The fact that the curve of mass flow rate per unit area has a maximum is connected with the
interesting and important effect called choking.
M1
,
o 1
chart or
isentropic relations
A is constant .
(A A )
(A A ) = AA (A A )
Hence,
chart
( )
T , A
T
A
o 1
or
isentropic relations
M2,
,
o 2
T
T
o 2
T
p
p
T2 To 2
o 2
p2 =
=
,
p1 p
T1 T
T
p
o 1
o 1
A2
A1
5
A2 /A1 = 2
A2 /A1 = 1
A2 /A1 = 0.8
M2
No throat
--- Throat
A2 /A1 = 1
A2 /A1 = 0.8
A2 /A1 = 1.2
M1
(1) For a given initial Mach number M 1 and a given area ratio
A2
A1
final M 2 or none at all. When there are two solutions, one is subsonic and the other is supersonic.
Which one of the two occurs depends, the part, on whether a throat exists between sections (1) and
(2), since in order to change the regime the flow must pass a throat at M = 1 .
For example if M 1 is subsonic and the passage is converging, then M 2 must be subsonic. But if the
passage is converging-diverging and has a throat between (1) and (2), the flow at section (2) may be
either subsonic (venturi) or supersonic (nozzle) depending on the pressures imposed at the inlet and
exit.
(ii) No solution (M 2 ) for the chosen values of M 1 and
A2
A1
mathematical sense can occur only if A2 is smaller than A1 . Physically, this result signifies that for a
given flow at section 1, there is a maximum contraction which is possible. The maximum contraction
corresponds to sonic velocity at 2. If conditions at section (1) are specified, the mass flow is fixed and
there is then a minimum cross-sectional area required to pass this flow. This phenomenon is called
choking. For a given area reduction, in subsonic flow there is a maximum initial Mach number which
can be maintained steadily; and in supersonic flow a minimum initial Mach number which can be
maintained steadily. At either of these limiting conditions, the flow at section (2) is sonic and is said to
be choked.
A2 will produce certain effects at 2 and will comprise an increase in M 2 and decrease in p 2 and T2 .
This slight reduction in A2 without a change in conditions at 1 must be accompanied by a reduction in
the back pressure p 2 . Further reduction in A2 may be made in the same way until M 2 reaches unity.
After this point is reached, there is no way of reducing the area further without simultaneous change
in the steady state conditions at section 1. If, for example, the pressure and temperature at 1 are held
constant, a reduction in
A2
A1
propagation, result in a reduced steady-state M 1 reducing the mass flow rate. The maximum possible
value of M 1 (also, maximum flow rate) is obtained when M 2 = 1 . To obtain this limiting flow, the back
pressure p 2 must of course be adjusted accordingly.
Module4:
Fiber in Ducts
Lecture 17:
Flow in ducts, (Nozzles and diffusers)
and wind tunnels (Contd.)
2
Converging-Diverging Nozzles
pB
V0 = 0
P0, T0
Constant
pB
pe
T
pB
p*
p0
p
p0
In regime I the flow is entirely subsonic, and the duct behaves like a conventional venturi tube. The
flow rate is sensitive to changes in back pressure (conditions 1a, 1b).
3
p*
p0
pe
p0
pB
p0
p*
p0
pB
p0
pB
p0
m T0
At p0
p*
p0
pB
p0
At condition 2, which forms the dividing-line between regimes I and II, a normal shock appears
downstream of the throat, and the process aft of the shock comprises subsonic deceleration. As the
back pressure is lowered, the shock moves down the nozzle. At condition 4, it appears in the exit
plane of the nozzle. Both in regime I & II the exit pressure pe is virtually identical with the back
pressure p B . But the flow rate in regime II, unlike regime I, is constant and is unaffected by the back
pressure (conditions at the throat are sonic and unaltered).
In regime III, as for condition 5, the flow within the entire nozzle is supersonic, and the pressure in the
exit plane is lower than the back pressure over expanded nozzle. The subsequent compression
occurs outside the nozzle by oblique shocks and their reflections.
Condition 6 is the design condition for the nozzle under supersonic conditions. The exit pressure is
identical with the back pressure. A reduction in the back pressure below that corresponding to
condition 6 has no effect on the flow pattern within the nozzle. In regime IV the expansion from
pe to p B occurs outside the nozzle in the form of expansion waves (underexpanded nozzle). In both
regimes III and IV the flow within the nozzle is independent of p B and corresponds to the flow pattern
for the design condition. Adjustment to the back pressure is made in the jet.
5
For subsonic flow there are an infinite number of possible pressure-distance curve. For the
supersonic region of flow, the pressure-distance curve is unique. Since, in subsonic flow the pressure
ratio does not depend solely on area ratio, but in supersonic flow the pressure ratio depends solely on
area ratio.
Module4:
Fiber in Ducts
Lecture 18:
Flow in ducts, (Nozzles and diffusers)
and wind tunnels(Contd.)
2
Laval nozzle as a supersonic wind tunnel
As the exit pressure is described the shock moves downstream, finally reaching the exit, pressure
there then reaches a value of p 4 . If pressure at the exit decreases further, the flow in the nozzle is not
affected; the pressure adjustment being made through system of oblique shock waves. For exit
pressure lower than p 4 , flow up to the exit is completely supersonic. Thus a Laval nozzle may be
used as supersonic wind tunnel provided
pe p 4
This is the principle of the open-circuit type of supersonic wind tunnel, operating from a high-pressure
reservoir or into a vacuum receiver or both. Continuous flow may be obtained if enough power is
available; otherwise, it is used as an intermittent or blow-down wind tunnel.
If the nozzle discharges directly into the receiver, the minimum pressure ratio for full supersonic flow
in the test section is
po = po
p
p4
e
But if a diffuser is attached to the exit, operation at a lower pressure ratio is possible, since the
subsonic flow downstream of the shock may be decelerated isentropically to the stagnation pressure
po in principle.
Test
section
M1
P1
P0
M2
P 2 = p4
P0
The pressure ratio required then is the ratio of stagnation pressures across a normal shock at the test
section Mach number M 1 , i.e.,
2
2
M1 1
s = o = 1 +
po + 1
( 1)M 1 + 2
2
(
)
1
M
+
1
( 1)
( 1)
Practically, this type of diffuser does not give the expected recovery, the interaction of shock wave
and boundary layer develops a flow different from this model.
However, a long constant area duct ahead of the subsonic diffuser nearly realizes the normal shock
recovery. Such a duct, provided it is long enough, gives nearly the same recompression as a normal
shock. The compression occurs through a system of shocks interacting with the thickened boundary
layer. Through this recovery through a dissipative system is not the most efficient, but it is often most
practical. It is quite stable with respect to variations of inlet conditions.
The normal shock pressure recovery is an ideal, convenient reference or standard for comparison of
performance.
If the supersonic flow at the test section could be isentropically compressed to sonic conditions at a
second throat, if could then be decelerated sub-sonically in the diffuser. Ideally, then it is possible to
operate at even a lower pressure ratio than the one for normal shock recovery. In the idealized case,
no shock complete recovery, po = po .
A2 = A1
A2
A1
po1 po
=
= s
po
po 2
A1
A2
max
A
A A
A 1
= 1 = 1 1 = 1 = f (M 1 )
A2
A1 A2
A1 s
With minimum area for the second throat and with larger ones, the shock wave jump from the test
section to the downstream side of the diffuser throat swallowing the shock test section is
supersonic, but also the second throat and part of the diffuser. The second throat area can be
reduced after the flow has started to move the shock toward the start. A2 = A1 will now make it ideal,
but not possible to achieve the reduction, but some contraction is possible.
Stationary power plants, aircraft propulsion, high-vacuum technology, fluid transport in chemical
process plants, natural gas transport in long pipe.
Wall friction is the chief factor, with the assumption that no special attempt is made to transfer heat to
or from the stream. When the ducts are reasonably short, the flow is approximately adiabatic, but for
extremely long ducts; there is sufficient area for heat transfer to make the flow non-adiabatic and
approximately isothermal.
1
h + u 2 = ho
2
m
u = = G ,
2
G2
h = ho
2 2
a relation between
constant)
h0
All possible states of the fluid for a given adiabatic, constant area flow lie on one of these lines.
Since for a pure substance s = s ( h, ) , the curves may be transferred to the enthalpy-entropy
diagram. The curves are called Fanno curves or Fanno lines.
For all substances, the Fanno curves have the general shape
6
ha
p0a*=p0b*
p0a=p0b
ho
pa
pa*=pb*
Large G
pb
Small G
b
s
Fanno curves
Module4:
Fiber in Ducts
Lecture 19:
Flow in ducts, (Nozzles and diffusers)
and wind tunnels(Contd.)
2
The upper branch of each Fanno curve corresponds to the subsonic flow, and the lower branch
corresponds to supersonic flow, and the Mach number is unity at the point of maximum entropy on
each Fanno curve.
Since the flow is adiabatic, the second law of thermodynamics states that entropy can not decrease;
thus the path of states along any Fanno curves must be towards the right. Consequently, if the flow at
a point in the duct is subsonic (a), the effects of friction will be to increase the velocity and Mach
number and to decrease the pressure and enthalpy. If the flow is initially supersonic (b), the friction
will decrease the velocity and Mach number and will increase the enthalpy and pressure. A subsonic
flow, therefore, will never become supersonic and a supersonic flow will not become subsonic, unless
a discontinuity is present.
The limiting pressure, beyond which the entropy would suffer a decrease, occurs at Mach number
unity and is denoted by p . p denotes the state where M = 1 for the adiabatic flow at constant area.
Referring to a state 'a ' the value of p a will be different for an isentropic flow as compared with the
value for an adiabatic constant area flow.
The isentropic stagnation pressure is reduced as a result of friction, irrespective to whether the flow is
subsonic or supersonic.
Choking due to friction Consider the stagnation enthalpy, flow per unit area and length of duct are
such that Mach number unity is reached at the end of duct. If the duct length is increased, it is evident
from the foregoing considerations that some sort of adjustment in the flow is necessary. When the
flow is subsonic, this adjustment is in the form of a reduction in the flow rate, that is, the flow is
chocked. When the flow is supersonic, the adjustment at first involves the appearance of shock
waves, and for sufficiently large increase in duct length, involves ultimately a choking of the flow.
Adiabatic, Constant-Area Flow of a Perfect Gas
+ d, M + dM
w
dx
p = RT
dp d dT
=
+
p
--- (1)
u2
dM 2 du 2 dT
a = RT , M =
= 2
RT
M2
u
T
2
Energy:-
--- (2)
( 2 )= 0
2
u
C p dT + d
dT 1 2 du 2
+
M
=0
2
T
2
u
--- (3)
Mass conservation
G=
m
= u
A
1 du 2
+
=0
2
2 u
--- (4)
Momentum Conservation:
f =
w
1 2
u
2
The hydraulic diameter is defined as 4 times the ratio of c.s. area to wetted perimeter,
D=
A
4A
=4
dx
dAw
dAw
dx
dp 4 f
u 2 dx
2 D
du
m
du = u 2
u
A
dp M 2
dx M 2 du 2
+
4f
+
=0
p
2
D
2 u2
NPTEL IIT Kharagpur: Prof. K.P. Sinhamahapatra, Dept. of Aerospace Engineering
-- (5)
5
From the definition of isentropic stagnation pressure
1 2 1
p o = p 1 +
M
2
dp
dp
o =
+
po
p
M2
2
2 dM
1 2 M 2
M
1+
2
--- (6)
F = pA + Au 2 = pA 1 + M 2
dF dp
M 2 dM 2
=
+
F
p 1+ M 2 M 2
--- (7)
We have now seven simultaneous linear algebraic relations involving eight differential variables
dp , d , dT , dM 2
du 2 2 , dpo
2,
p
T
po
M
u
dF
and
dx
or 4
f dx
The physical phenomenon causing changes in state is viscous friction. Hence, the variable 4 f
the physically independent variable.
From (1) and (3)
dp d 1 2 dk 2
=
p
2
k2
Introducing (4) into it
D.
dx
is
D
dp
1 + ( 1)M 2 du 2
=
p
2
u2
dp
M 2 1 + ( 1)M 2
dx
4
=
f
p
D
2 1 M 2
-- (8)
Similarly,
dM
=
M2
1 2
M
2
4 f dx
D
1 M 2
M 2 1 +
M2
du
dx
=
4
f
u 2 (1 M 2 )
D
--- (10)
dT 1 da
dx
( 1)M 4
4
=
=
f
2 a
T
D
2 1 M 2
M2
2 1 M 2
4f
-- (9)
--- (11)
dx
D
-- (12)
dpo
dx
M2
=
4f
2
p
D
M2
dF
dx
=
4
f
F
D
2 1+ M 2
- (13)
-- (14)
To 2
Also,
To1
s
1 dpo
ds
= ln
1
po
cp
cp
p
o2
po1
ds 1 2
dx
M 4f
=
cp
D
2
--- (15)
Module4:
Fiber in Ducts
Lecture 20:
Flow in ducts, (Nozzles and diffusers)
and wind tunnels(Contd.)
2
Since entropy can not decrease in an adiabatic process, (15) tells that ' f ' must be positive, thus the
shearing stress must always act on the stream in a direction opposite to the direction of flow, as
assumed.
The isentropic stagnation pressure and impulse function must decrease if friction is present both in
the subsonic or supersonic flow. Wall friction reduces the effectiveness of all types of flow machinery
and also reduces the thrust obtainable from jet propulsion devices.
Subsonic
Supersonic
decreases
increases
increases
decreases
increases
decreases
decreases
increases
decreases
increases
po , F
decreases
decreases
Summary
The Mach number always tends toward unity. Continuous transition from one regime to the other is
impossible. For given conditions at an initial section of the duct, the maximum possible duct length
which can be employed without attiring the given initial conditions and without introducing
discontinuities is that length for which the exit Mach number is exactly unity.
Lmax
1
dx
= 2
4f
M
D
1 M 2
dM 2
1 2
M
M 2 1 +
2
f =
1
Lmax
Lmax
f dx ;
we have
Lmax 1 M 2 + 1
( + 1) M
4f
=
+
ln
D
2
M2
1 2
M
2 1 +
2
max value of 4 f L
for given M
Hence, the length of duct L required for the flow to pass from M 1 to M 2 can be found from
4f
L
L
L
= 4 f max 4 f max
D
D M1
D M2
dp
=
p
1 + ( 1)M 2
dM 2
1 2
2 M 2 1 +
M
2
1
M
+1
1
21 +
M2
2
v
Similarly,
=M
( + 1)
1 2
M
21 +
2
T
a2
+1
=
=
T a 2
1 2
21 +
M
2
u
1
=
=
u M
po
po
F
1
M
1 2
M
21 +
2
+1
1 2 + 1
21 + 2 M 1
+1
1+ M 2
1 2
M 2( + 1)1 +
M
2
s s
ln M 2
cp
+1
2M 2 1 + 1 M 2
+1
Hence,
M 2
p2 = p
p1 p
M 1
p
and so on.
xxx---
Module4:
Fiber in Ducts
Lecture 21:
Flow in ducts, (Nozzles and diffusers)
and wind tunnels(Contd.)
2
Isothermal flow in long ducts
Isothermal flow with friction is of interest in connection with pipe lines for transporting gas over long
distances. Mach number is usually quite low, but substantial pressure changes because of friction
over great lengths.
Energy equation: -
v2
dQ = c p dT + d = c p dTo
2
--- (1)
(
1)M 2 dM 2
1 2 dTo
=
To = T 1 +
M
2
To
1 2 M 2
21 +
M
2
--- (2)
dp
p
Also
= d
--- (3)
dM 2
du
=
2
M2
u
--- (4)
The continuity, momentum and definition of stagnation pressure (local) remain unaltered (e.g. 4, 5, 6
earlier)
Solving the algebraic equations
M2
dp d
du
1 dM 2
dx
=
=
=
=
4
f
p
u
2 M2
D
2 1 M 2
+1
M2
dpo
2
4 f dx
=
po
D
1 2
M
2 M 2 1 1 +
2
M 2 1
dTo
=
To
( 1) M 2
1
2
2 1 M 1 +
M2
2
4f
dx
D
In this case, the direction of change depends not on M alone, but or M 2 . Since 4 f
dx
is always
D
(subsonic)
decreases
increases
decreases
increases
increases
decreases
increases
decreases
To
increases
decreases
po
decreases
increases
M<
Again
Lmax
dx
4f
D
L
4 f max
D
Denoting properties at M =
2
Since M 2 = u
M2
1 M 2 2
M
4
M
1 M 2
=
+ lu M 2
2
M
1
by symbols V t ,
2
p t , it can be written as M
u
V
Also
= M
u t
1
=
*t =
u
M
1
p
=
=
p t t
M
From the formula for isentropic stagnation pressure
(V t )
po
po
1 2 1
M
1 +
2
= t =
1 1 1
1 +
M2
1+
1 2 1
2
M
3 1
1+
M2
To
T
2 1 2
2
M
= t =
=
1 +
t
1 1 3 1
T
2
To
1+
2
In long commercial pipelines, the Mach numbers employed are so low that the loss in stagnation
pressure is virtually identical with the loss in static pressure.
L 1 M1 1 M 2
L
M1
L
= 4 f max 4 f max =
+
4f
l
n
2
D 2
D 1
D
M 12
M 22
M2
2
Since
p
p1 M 2
=
M 2 = M1 1
p2
p2 M 1
2
p
2
1 2
p
L
p
4f
= 2 1 ln 1
D
M1
p2
6
Since M 2 cannot exceed 1
p
, it follows form the pressure ratio relation that 2
p2
p1
>M 2
1
p1
( p1 p2 )
p1
1 + M 12 p1 p2
L
2 p1 p2
2
1 M1
4f
p
D M 12 p1
2
1
The conventional pressure-drop formula for incompressible flow is similar to above except the square
bracket on the rhs is unity. Solving the quadratic equation
2
2
1 M 12
p1 p2 1 M 1
M1
L
=
f
4
2
2
2
D
p1
1+ M1
1+ M1 1+ M1
It is convenient to use
2
M1
m RT1
=
A P1
For a given value of M 1 , there is a maximum length for continuous isothermal flow, hence, it follows
that choking effects may occur in similar fashion to those for adiabatic flow.
Module4:
Fiber in Ducts
Lecture 22:
Flow in ducts, (Nozzles and diffusers)
and wind tunnels(Contd.)
2
Flow in ducts with heating or cooling factors tending to produce continuous changes in the state of a
flowing stream are (i) changes in cross-sectional area, (ii) wall friction and (iii) energy effects such as
external heat exchange, combustion, or moisture condensation. Simple To change is difficult to
achieve in practice. If To is changed through external heat exchange, the connection between the
mechanisms of friction and of heat transfer assure that frictional effects will be present. Combustion
change in mass rate, chemical composition Simple To change is an ideal case.
With constant area and no friction, the momentum equation is p + u 2 = F
Continuity u =
Combining p +
= constant
m
= constant = G
A
G2
F
A
For fixed mass flow rate per unit area and constant impulse function per unit area, the above equation
defines a unique relation between p and called the Rayleigh line. Since both enthalpy and
entropy are functions of p and , the above equation can be used for representing the Rayleigh line
on the h s diagram. All fluids have Rayleigh curves of the general form.
p01
isentrope
Fanno
p*
p1
3
The relation above, in the differential form, becomes
dp G 2
=
= u 2 u = dp
d
d 2
dp represents the local velocity of sound only for a special circumstances, namely, when the
d
infinitesimal variation of pressure with density is such that there is no change of entropy. This
condition is fulfilled at the point of maximum entropy on the Rayleigh line. This point represents the
state of Mach number of unity for the process of simple To - change.
Beginning with state 1, Mach number unity might be reached in several way (isentropically,
adiabatically at constant area, etc), and it is only for simple heating the * point will correspond to
Mach number unity. The branch of the Rayleigh curve about the point of maximum entropy generally
corresponds to subsonic flow. Since the process of simple heating is thermodynamically reversible,
heat addition must corresponds to an entropy increase and heat rejection must corresponds to an
entropy decrease. Thus at subsonic speeds the Mach number is increased by heating and decreased
by cooling. The reverse happens in case of supersonic flow. Hence, heat addition, like friction, always
tends to make the Mach number approach unity. Cooling causes the Mach number to change always
in the direction away from unity.
For heat addition at either subsonic or supersonic speeds, the amount of heat input can not be
greater than that for which the leaving Mach number is unity. If the heat addition is too great, the flow
will be choked, the initial Mach number will be reduced to a magnitude that is consistent with the
amount of heat thermal choking.
p1, T1
M1, T01
Mass Conservation
p2, T2
M2, T02
2 u1
=
1 u 2
4
Momentum Equation
Using
p1 p 2 =
m
(u 2 u1 )
A
m
= u and u 2 = pM the momentum equation can be arranged to give
A
p2 1 + M 1
=
p1 1 + M 2 2
2
Equation of state:
p2 2T2
=
p1 1T1
Impulse function
F2
F1
M2
M1
T2 p2 1 p2 u2
=
=
T1 p1 2 p1 u1
or
u2 a1 u2
=
a2 u1 u1
(
(
T1
T2
)
)
p2 1 + M 2
=1
2
p1 1 + M 1
2
1 2 1
1+
M2
po 2 p2
2
po1 p1
1 2 1
M1
1 +
2
Change in entropy
T2
s2 s1
T1
= ln
cp
p2 1
p
When the process involves heat exchange, the change in stagnation temperature is a direct measure
of the amount of heat transfer. Form the energy equation
Q = C p (T2 T1 ) +
u 2 u1
= C p (To 2 To1 )
2
2
5
temperatures would be made respectively identical for the real process and for the equivalent
process. For a Rayleigh process, the change in stream properties are due primarily to changes in
stagnation temperature, u , the rate of change of stream properties along the Rayleigh line is a
function of the rate of change of stagnation temperature.
Now
To
1 2
= 1+
M
T
2
T
T
o2 = 2
To1 T1
1+
1+
M2
M2
T2 1 + M 1 u 2
=
T1 1 + M 2 2 u1
2
u2
from the definition of Mach number
Using
u1
(
(
)
)
2 2
T2 M 2 1 + M 1
=
2
T1
M1 1+ M 22
(1 + M )
(1 + M )
2 2
To 2 M 2
=
2
To1
M1
1 ,
p2
1+
2 2
p1
u2
u1
1+
1
2
1
2
M2
M1
normalize the equation by setting the Mach number equal to unity at one of the sections, say at 1.
( + 1) M 2
T
=
2
T
1+ M 2
2
u
( + 1)M 2
=
=
u
1+ M 2
p
+1
=
p 1+ M 2
po
po
1 2 1
21 +
M
+1
2
=
2
+1
1+ M
+1
s s
= ln M 2
2
cp
1+ M
+1
The ratio of properties at two sections where the Mach numbers are M 1 and M 2 are found using
these normalized expressions
To 2
To1
To
M
To 1
and so on...
=
T o
M 2
To
Module5:
Multi - Dimensional
Problems
Lecture 23:
Linearized Compressible Flow
Stokes theorem:
Green-Gauss theorem:
Or
A dr
A dS
A ndA
AdV
A ndA
gn j dA
g
dV
x j
Continuity equation
u j 0 or
t x j
u j
D
D
u
0
Dt
Dt
x j
Momentum equation
p
ui uiu j fi
t
x j
xi
Assuming steady flow and no body forces, integrating the equation over an enclosing control surface,
the force of the fluid on the body is obtained as
Fi
A1
ui uk nk dA A
A1
A1
pni dA
u1 uk nk dA A
pn1dA
u2 uk nk dA A
pn2 dA
3
Energy equation
1 2
1 2
p ui
e u
e u u j q fi ui
t
xi
2
2
x j
or
De
D 1 2
p ui
u q f i ui
Dt
Dt 2
xi
u
De
q p i
Dt
xi
Or
[using continuity]
[using momentum]
D 1
De
p q
Dt
Dt
Introducing entropy
De
D1
DS
p T
Dt
Dt
Dt
Introducing enthalpy
D
1 2
1 p
h
u
q
f
u
i i
2
Dt
t
If the flow is inviscid, non-conducting, adiabatic and steady both entropy and total enthalpy are conserved
along streamlines. Using the natural co-ordinate system (streamline s and normal n ), the steady flow
equations of motion are
u n const .
u
u
p
s
s
- continuity
- s momentum
u2
p
u 2
R
n
s
1
h u 2 ho
2
TdS dh
- energy
1
udu dp dho
S
u 1 p
u
0
s
s s
u 1 p dho
u u dho
S
u
u
n
n n dn
n R dn
dho
u
dn
ho constant throughout
And
0 throughout
TdS dh
dp
T S h
p f DV
Dt
V V
V
DV V
V V
V V
Dt
t
t
2
V
1 2
V V
t
2
V
1
T S h f
V 2 V
t
2
1
V
V T S h V 2
2
t
T S ho
V ho TS
Circulation and rotation
u dl u n dA
c
Vorticity =
u 2 angular velocity
ij
u j
xi
ui
xj
----xxx---
Module5:
Multi - Dimensional
Problems
Lecture 24:
Linearized Compressible Flow
(Contd.)
Irrotational flow
ui 1 ui u j
x j 2 x j xi
1 ui u j
2
x
j xi
u , u i
xi
1
eij ijk k
2
Kelvins theorem:
In an inviscid flow of a homogeneous fluid without body forces the circulation along a closed fluid line
remains constant with respect to time.
Irrotational Flow
Consider adiabatic, irrotational flow. From Croccos theorem, these flows are isentropic. The equations
are
u j 0
t x j
ui
u
p
u j i
x i
t
xj
po o
or,
S constant
Auxiliary conditions
ui u j
0
x j xi
ui
x i
1 2
u1 u2 2 u32 hO ; for a perfect gas
2
a
a 2 u1 u 2 u3
1 1 2
O
a
1
2
1 2 1
2
Consider steady flow. Pressure can be eliminated from the momentum equation using
p p
2
xi s xi
xi
Scalar multiplying the momentum equation by ui
ui
ui p
a 2
ui
ui u j
xi
xi
xj
a2
u k
xk
Expanding
2
1
a2
ux u
a2
ux u
2
a2
ux
u u
u1u2 1 2
x2 x1
u u
u u
u2u3 2 3 u3u1 3 1 0
x3 x2
x1 x3
In terms of the velocity potential, the equation becomes
1 2
2
a 2 xi x j xi x j x j x j
NPTEL IIT Kharagpur: Prof. K.P. Sinhamahapatra, Dept. of Aerospace Engineering
4
or
y2
x2
z2
1 2 xx 1 2 yy 1 2
a
a
a
x y
a
xy 2
y z
a
yz 2
z x
a2
zz
xz 0
u 2 1 u
0
1 2
a u s n
Free stream
U along x1 , p , , T a
M U
u1 U , u 2 0, u3 0
Small perturbation
u1 U u , u 2 , u 3 w
u
w
,
,
1
U
U
U
The equation of motion can then be written as
u w
w
2 u
2
w2
a2
U u
x1
x2
x3
x1 x2 x3
w
w u
u
U u
U u w
w
x
x
x
x
x
1
2
2
3
1 x3
From the energy equation
U u 2 2 w 2
2
a 2 a
2
or
1
2
2U
a
U
a2
1 1 2
u u 2 2 w 2
1 M xu x
2
x3
u 1 u 2 1 2 w 2 u
1
2
U
2
2 U 2 x1
U
u 1 2 1 w 2 u 2
1
2
2
U
2
2
U
U
x2
u 1 w 2 1 u 2 2 w
1
2
2
U
2
2
U
U
x3
v
u u w
u u w w w
M 2
U
U
x
x
U
U
x
x
U
x
x3
2
1
3
1
2
----xxx---
Module5:
Multi - Dimensional
Problems
Lecture 25:
Linearized Compressible Flow
(Contd.)
2
The equation is still exact. Neglecting the terms containing squares (second degree) of perturbation
velocities, the equation becomes
1 M xu x
2
u u
u w
w
2
M 1
M 1
U x1
U x2 x3
x3
w u w
M 2
U x2 x1
U x3 x1
If the entire right hand side is neglected the equation becomes linear
1 M xu x
2
1 M
2
or
w
0
x3
2 2 2
0
x1 x2 x3
u
on the lhs becomes very small and it is then not
x1
,
x2
( is perturbation potential)
M 1
w
and can be neglected on the rhs. So, the small-perturbation equation valid for subsonic,
x3
1 M xu x
2
1 M
2
w
u u
2
M 1
U x1
x3
2 2 2 1M 2
2
2
2
U
x1 x12
x1 x2
x3
2
For flows outside the transonic range, the rhs can be neglected.
2
For M very large, hypersonic-flow, certain terms on the rhs must be retained as the product of some of
2
Pressure coefficient
Cp
p p
2 p p
2
1
M
p
2
U
2
po
p
2 p
2
2
1
1
Cp
2
M 2 p M 2 po
M
p
2 1
2
1
2
2 1M
V2
U2
2
M 2 , M 2
a
a
2
where
U 2 a 2
V2
a2
2 1
2
1
a a
2
Cp
or
a 2
2
M 2
Cp
1
2
1
2
1
2
V2
1 2
U
V2
1 2
U
2
1
1 2
V
M 1 2 1
1
2
U
2
M 2
2
2
2 1
1 2 U u w
1
M 1
2
U
2
M 2
1
2
2
2
u w
1
2 2u
M
1
1
2
2
U
U
u
u 2 2 w2
2
2
1 M
2
U
U
U 2
neglecting
higher power
For 2-D and planar flows C p , consistent with the firstorder perturbation equations, is
C p 2
u
U
For flow over axially symmetric or elongated bodies, the consistent approximation is
2u 2 w 2
Cp
2
U
U
(Ackerets problem)
Consider the flow past a boundary of sinusoidal shape, the so called wavy wall. The boundary is given
by
y h sin x 0
h denotes the amplitude of the waves of the wall, and l 2
is the wavelength
5
y
d
x
For plane subsonic or supersonic flow, the linear equation may be used
1 M
2
2 2
2 0
2
y
x
,
x y
y
U
y y 0
x wall
x, 0
and
U h cos x
Consider 1 M 0 , i.e., a subsonic flow. The equation is elliptic
2
2 1 2
0
x2 2 y2
Try
x , y F x G y
F G FG 0
2
or
F 1 G
0
F 2 G
6
----xxx---
Module5:
Multi - Dimensional
Problems
Lecture 26:
Linearized Compressible Flow
(Contd.)
2
The first term is function of x only, and the second one is functions of y only. The equation holds for any
x, y , hence,
F
constant
F
k 2 , &
1 G
k2
2
G
F A1 sin kx A2 coskx
and
G B1e k y B2 e k y
B2 0
From the wall boundary condition
dG
F x
U h cos x
y
dy
y 0
y 0
The condition is satisfied if
A1 0, k , and
x, y
Uh
U h
1 M
A2 B1 k U h
e y cos x =
2 y
he
cos
e y sin x
U h e y cos x
C p 2
u
2h y
e
sin x
2 x
l
p wall
C p y 0
2 h
sin x
There is no drag force, as the pressure is in phase with the wall and hence symmetrical about the wavy
wall. The pressure coefficient increases with Mach number, proportionally to 1
the
Prandtl-
Glauert factor. Also, the attenuation of the perturbation away from the wall becomes weaker as the Mach
number is increased.
Cp
f x , y
The relation is between 3 variables, instead of six a modified pressure coefficient and a modified
coordinate system. The factors that make this reduction possible incorporate the effects of Mach number,
wave amplitude and wavelength in such a way that one relation is valid for all combination of these three
variables.
1. since, as per our assumption
h
1 M
,
1, it is evident from the solution that
U
U
1 M M
2
1 M
2
u
U
M 1 h
2
1 M
M 1 h
2
1 M
or
M 1 h
2
1 M
2
2
The transonic parameter include and 3 as exponent of 1 M
3.
h e y cos x
h e h sin x cos x
boundary
2
h cos x 1 h 1 M sin x H .O.T .
The second term is always smaller than or at most equal to h ; it should be negligible
compared to the first, for the small perturbation theory to be valid. This approximation becomes
even better as the Mach number is increased. The amplitude of the permissible perturbation, for
linearized theory to be accurate, is given by
h 1
or in terms of the maximum inclination, , of the wall
1
For supersonic flow, M 1 0 and the equation becomes hyperbolic
2
2 2
2 0
2
x
y
2
This is the simple wave equation with the general solution being the sum of two arbitrary functions
f x y and g x y
NPTEL IIT Kharagpur: Prof. K.P. Sinhamahapatra, Dept. of Aerospace Engineering
x, y f x y g x y
Boundary conditions are identical.
Only the function f is needed and g is set equal to zero. This choice has to do with the direction of
flow or the distinction between upstream and downstream region of flow.
f x y y 0 f x U h cos x
y 0
f x
hsin x
f x, y f x y
h sin x y
2
sin x y M 1
M 1
Uh
Uh
cos x y
U h cos x y
Cp
2 h
M 1
2
cos x y
----xxx---
Module5:
Multi - Dimensional
Problems
Lecture 27:
Linearized Compressible Flow
(Contd.)
2
1. The solution does not contain an exponential attenuation factor like the subsonic solution and
hence the perturbation does not decrease with y . Instead, the same value of the perturbation
exists all along the straight lines
x y constant
These lines are inclined at the Mach angle with respect to the undisturbed flow. They are the
Mach lines or characteristics. The existence of these characteristics is independent of the specific
boundary conditions, being contained in the form of the solution. f = constant along
x y constant, and g constant along lines x y constant. The former characteristics are
inclined upstream; they originate at infinity and so carry no perturbation. Hence, g 0 when the
fluid above the wall is unlimited.
2. C p on the wall is C p
2 h
cos x
Comparing with the subsonic case, it can be seen that the maxima and minima of the pressure
are now shifted by the phase 2 from the maxima and minima of the wall coordinate y . Hence,
the pressure distribution is now anti-symmetrical around the crests and troughs of the wall and a
drag force exists. The magnitude of the drag coefficient per wavelength is
dy
sin
tan
]
[
dx w
l1
dy
C D C p dx
ol
dx
This can be evaluated by replacing the sine of the slope of the wall by the tangent, valid within the
frame work of small perturbation
C p can be written as
CD
Cp
dy
2
M 1 dx
2
dy
,
2
M 1 dx
2
dy 1 l dy
o dx
dx l dx
3. The range of validity of the approximations follows exactly the same line as in the subsonic case
Force
(M<1)
M>1
x, y f x y ,
x , y g x y ,
yo
yo
dy
U
f x
dx U y y 0
NPTEL IIT Kharagpur: Prof. K.P. Sinhamahapatra, Dept. of Aerospace Engineering
f x
U y
u
x U
Similarly,
g x
Hence,
Or
U y
x L
2
U f x upper surface
Cp
2 g x lower surface
U
C pU
2
dy
dy
;
C
pL
2
2
dx
M 1
M 1 dx L
U
2
----xxx---
Module6:
Bodies of Revolution
Lecture 28:
Slender Body Theory
2
Different classes of small perturbation flow
1) Two-dimensional flow
2) Planar System 2D flow is a special class
3) Elongated body Slender body is a special case, which is so elongated that bcs may be
applied on the axis.
4) Interference Problems
Cylindrical co-ordinates are more convenient for elongated bodies and bodies of revolution
z
xx
locates the meridian plane, x r , relative to some convenient reference, say x z plane. The
velocity components are given by
u1 = U + u =
1
,1 = =
, w1 = w =
x
r
r
is total potential
Mass conservation:
x2
x3
x1
Volume = x1 x2 x3
r
x
Volume = x r r
or
( u1 ) + ( u 2 ) + ( u 3 ) = 0
x1
x 2
x3
( u1 x r ) x + ( 1 r x ) x + ( w xr ) = 0
x
r
Also,
1
1
( u1 ) +
( r ) +
( w) = 0
x
r
w
( u1 du1 + d + wdw ) = dp = a 2 d
Hence,
( u1 ) = u1 + u1
x1
x1
x1
=
u
1
u
u
u1 2 u1 1 + u 2 2 + u 3 3
x1
x1
x1
a x1
(1 M ) + + = 0
2
x1
x 2
x3
is perturbation potential
Cylindrical system contains one additional term from
( r ) = ( ) +
r r
r
r
Module6:
Bodies of Revolution
Lecture 29:
Slender Body Theory
(Contd.)
2
Hence
(1 M
2 2 1 1 2
) x2 + r 2 + r r + r 2 x2 = 0
Boundary conditions
2D flow,
u (0 )
~
2
U
dx 2
x1
u2
u 2 (0)
dR
=
dx U + u R
Approximation similar to 2D or planar case is not applicable
Consider longitudinal section of either 2D/planar body or a body of revolution and radial velocity near
the surface
x2 =
r=R
2D/planar
Body of revolution
x2 = 0
r =0
4
This velocity field may be obtained by a suitable source distribution on x 2 = 0 or r = 0 . In 2D/planer
flow, the velocity near the axis is nearly the same as at the boundary. In axially symmetric flow the
radial velocity at the axis must be infinite if it is to be finite on the boundary.
Using power series expansion, for 2D case
u1 ( x1 , x 2 ) = U + u ( x1 ,0) + a1 x 2 + a 2 x 2 + ...
2
1
( r ) . However, this term must be of the same order as the
r r
other terms
1
u
( r ) ~
r r
x
or
u
( r ) ~ r
r
x
( r ) ~ 0
r
as
r 0 r = a0 ( x )
1
r
a0
+ a1 + a2 r + ...
r
The correct form for an approximate boundary condition on the axis in case of an elongated body is
dR r ( r )0
=
dx U + u R U
5
Using irrotationality,
u
=
r x
a0
+ a1 + ...
r
Cp =
2
M 2
( 1)
2
2
2
w
1 2 2u u
M + 2 + 2 + 2
1
1
2
u u U U
2
u
2 + w 2
2 u
+ 1 M
+
C p = 2
2
2
U
U
U
For 2-D case, it is possible to neglect all but the first term in the first order theory. But for elongated
axi-symmetric body, the radial component is of different order form , near the axis, as shown
above. Since is very large near r = 0 , the pressure coefficient for axially symmetric case to the first
order accuracy is
Cp = 2
U U
Module6:
Bodies of Revolution
Lecture 30:
Slender Body Theory
(Contd.)
A
x2 + r 2
- A for source
If the source is at x = ,
r=0
( x, r ) =
Since,
the
governing
equation
(x )
is
A0
+
( x, r ) =
x2 + r 2
+ r2
linear,
superposition
is
allowed
+
+ ...
2
2
2
2
( x 1 ) + r
( x 2 ) + r
A1
A2
( x, r ) =
f ( )
is
obtained
by
f ( ) d
(x )
satisfying
the
+ r2
boundary
conditions.
Solution is most often obtained through numerical method with finite number of sources and sinks.
Subsonic flow
2 = 1 M 2 > 0
The potential equation is
1 2
1 2
+
+
=0
2 r 2 2 r r x 2
or
2 1 2 1 2
+
+
=0
x 2 2 y 2 2 z 2
x = x, r = r
or
x = x ,
y = y, z = z
2 1 2
2+
+
=0
r r r x2
or
2 2 2
+
+
=0
x 2 y 2 z 2
( x, r ) =
( x, y , z ) =
(x )
+ r
2 2
(x )
+ 2 y2 + 2 z2
( x, r ) =
f ( ) d
(x )
+ 2r 2
Supersonic flow
2 = M 2 1 > 0
The equation becomes
2
2 1
2
+
=0
r 2 r r
x2
(wave equation)
By analogy,
( x, r ) =
(x )
2r 2
The equation is satisfied. Representing flows is problematic, since denominator may become zero or
negative, resulting infinite or imaginary .
( x, r ) =
x r
f ( ) d
(x )
2r 2
5
For suitable distribution of sources, solution may be obtained that have no singularities off the axis.
Even though the sources are distributed along the x axis from 0 to 1, the value of at
( x, r )
include
Module6:
Bodies of Revolution
Lecture 31:
Slender Body Theory
(Contd.)
Mach cone
from origin
( x, r )
The upper limit implies that a source has no influence ahead of its Mach cone.
In linearized 2D supersonic flow a disturbance has no effect either upstream or downstream of its
Mach lines. In the axially symmetric case, there is an effect over the whole region downstream of the
Mach cone. Thus ( x,r ) is affected by all the disturbances ahead of = x r .
at1
at2
Ut1
Ut2
Ut2
Ut1
Wave propagation from a moving a
source
U<<a
The integrand is undefined at the upper limit. Need to be treated specially (Hadamard rule).
Differentiation
must
follow
Leibnitz
rule.
g ( s ) F
d g (s )
dg
(
)
:
Leibnitz
rule
F
s
d
d + F {g (s ); s}
=
0
ds 0
ds
s
Substituting = x r cosh
d = r sinh d =
= 0 = cosh 1
(x )
x
r
= x r = cosh 1 (1) = 0
2 r 2 d
5
Hence
( x, r ) =
cosh 1
u ( x, r ) =
f ( x r cosh ) d
cosh 1
1
r
=
f ( x r cosh ) d f ( 0 )
2 2
0
2
x
x r
( x, r ) =
cosh 1 ( x )
x
r
f ( x r cosh )( cosh ) d + f ( 0 )
=
2
0
r x 2r 2
r
f=
f
( x r cosh )
u =
1 x r
r 0
f ( )
x r
(x )
f ( )( x )
(x )
d ,
2 2
2 2
Module6:
Bodies of Revolution
Lecture 32:
Slender Body Theory
(Contd.)
f ( ) = a
f (0) = 0,
f ( ) = a
( x, r ) =
( r ) a ( x r cosh ) d
cosh 1 x
1 x
= ax cosh
1
r
x
x
x
u = a cosh
, = a
1
r
r
1
( x r),
x
= constant (radial lines or rays from the
r
origin) A conical flow field. On a ray which his on the Mach cone from the vertex, the perturbation
velocities are zero; the flow is still at free stream condition. Higher perturbation velocities near the axis,
flow direction changes. One streamline along the ray. The cone that contains the ray can be
represented as a solid cone in supersonic flow. Internal flow than of no consequence
Vertex angle of the cone depends on a. Hence for a given cone of semi-vertex angle , it is required
to determine a by satisfying the boundary condition
= tan
+
U
u
cone
On the cone surface
u = a cosh 1
cot
= a cot 2 2
NPTEL IIT Kharagpur: Prof. K.P. Sinhamahapatra, Dept. of Aerospace Engineering
U tan
cot
cot 2 2 + tan cosh 1
cot >>
then
cot
cosh 1
~ 2
log
2
tan log
and
a=
Since, for such a slender cone r
is also small
U
= U 2
cot
U 2 x log
2x
1
r
u
2x
,
= 2 log
U
r
=2
x
r
, and hence
u
2
= 2 log
U cone
=
cone
C p = 2
U U
for a wedge C p =
2 1
= 2 2 log
Pressure rise on the cone is much less than on the wedge three-dimensional relieving effect.
Wedge
cp
Cone
Uc
Ua
Uc
r ,
w
6
z
Since the equation is linear, the solution can be written as sum of two solutions.
( x, r , ) = a ( x, r ) + c ( x, r , )
a is an axial flow and c is a cross flow.
For the axial flow, the free stream velocity is U a = U cos and for the cross flow it is U c = U sin .
The axial problem is already discussed and can be solved similarly provided the boundary condition
can also be split.
The cross-flow problem must satisfy the full equation,
2
2c 1 c 1 2c
2 c
+
+
=0
r 2 r r r 2 2
x2
Module6:
Bodies of Revolution
Lecture 33:
Slender Body Theory
(Contd.)
2
The axial flow satisfies
2
2a 1 a
2 a
=0
r 2 r r
x2
2 a
r 2 r
1 a
+
r r r
2
1 a
2 a
2
x 2 r
r r
=0
a
a
is replaced by cos
r
r
2
a 1
a cos a
2
2
cos
cos
cos a = 0
2
2
2
r
r r r
r
r
x
r
r
or
2
a 1
a 1 2
a
2
2
+
+
cos
cos
cos
cos a = 0
2
2
2
2
r
r r r
r r
r
x
r
cos
or
a
r
c ( x, r , ) = cos
a 1 x r
=
r
r 0
a a
=
r
z
f ( ) d
(x )
2r 2
x r
f ( ) d
a
= 2r
3
0
r
( x ) 2 2 r 2 2
1
+
r
x r
f ( )( x )
2
( x ) 2 r 2 0
For a pointed body the last term at the lower limit gives zero, but be comes infinite at the upper limit.
Hence, only the finite part of the first integral should be considered.
3
The result can formally be written as
a
= 2r
r
f ( ) d
x r
( x ) 2 2 r 2
The notation denotes the finite part of the integral. The undetermined functions f ( ),
f ( ) are to be
determined by the cross flow boundary conditions. The cross flow solution is represented by these two
forms, which can be written as either
c ( x, r , ) =
or
cos
r
c ( x, r , ) = 2 r
x r
x r
f ( )( x )
(x )
2 2
f ( ) d
( x ) 2 2 r 2
Boundary conditions
Radial velocity in any cross section is
U c cos +
and the axial velocity is
Ua +
dR
=
dx U a + u body
dR
+ U c cos
=
Ua +
x body
r
body dx
or
and
dR
a c
Ua + a + c
=
r + r + U c cos
x
x body
body dx
dR
U
a
=
+
a
x body
r body dx
dR c
c
+
=
cos
U
c
r
dx x body
body
For slender body the rhs of the exact cross flow be is lower order and may be neglected
c + U c cos 0
r body
Drag on slender bodies of arbitrary profile in axially symmetric flow
( x, r ) =
f ( ) d
x r
(x )
2r 2
For slender bodies, the integral need to be evaluated for small values of r
f ( ) d
x r
(x )
2r 2
x r
x r
f ( ) d
(x )
= ( I 1 + I 2 )
Since, the first integrand is not singular; it can be expanded in power series of 2 r 2
f ( )
(x )
2r 2
f ( ) 1 f ( ) 2 2
r + ...
+
x 2 ( x )3
2r 2
I1 = f (0)log x f ( x )log +
f ( )log(x )d + f (x )log
0
f ( x r cosh ) d
I2 =
= f ( x)
r +
cosh 1
cosh 1
r +
r
d r
f ( x ) cosh d + H .O.T . in r
For r 0 ,
I 2 = f ( x ) log
2
+ f ( x ) log f ( x ) + ...
r
= I1 I 2 = f ( x ) log
f ( x )
=
r
r
Apply bc:
r=R
or
x
2
f ( ) log ( x ) d
r 0
r = f ( x)
f ( x ) = body R
f (x ) = U R
dR U dA
=
dx 2 dx
A = R 2 is the cross-sectional area of the body at x . Hence, source strength is proportional only to the
local rate of change of area of the body. The portions of the body that are far away do not influence
the local conditions. The rate at which the fluid is pushed outward locally depends entirely on the local
rate of area change.
6
Thus, the solution for axial flow over or slender body of revolution with closed nose and arbitrary
smooth meridional section is
U
2 U
A ( x ) log
r 2
2
( x, r ) =
(C )
A ( ) log ( x ) d
A ( x ) R dR
=
2 r
r dx
p body
A( x )
2
1 d x
dR
(
)
(
)
A
x
d
log
log
R dx 0
dx
dR
R+
M ,
pB
dx
dR
dx
dx
Projected area
dA (x) = 2R dR
p at an arbitrary section x is uniform over the cross-section. The pressure acts on the projected
area 2RdR .
Thus the drag is
CD =
D
1
U 2 AB
2
1
AB
Cp
dA
dx + C pB =C D11 C p1 + C pB
dx
AB C D1 = C p A( x )dx =
L
A( x )
1
2
I1 =
L dR
2
A( x )A( x )log
A( x )dx
R(x ) 0 dx
d x
A( )log( x )d dx = I1 I 2 + I 3
dx 0
log
R
2
2
d [A( x )]
L
R
1
dR
2
(
)
A
A
log
=
+
dx
2
2 0 0 dx
L
L
x
1
1 L
2 x
I3 =
A( x ) A( )log( x )d A( x ) A( )log( x )d dx
0 0
0
0
A(L ) A( )log(L )d
L
AB C D1 =
A( )A( )log( x )d dx
1
2
A(L ) L
+
A( )log(L )d
R(L )
0
A = 2 R R A(L ) = 0
base
or
In such a case
AB C D1 =
A( )A(x )log(x )d dx
1
= 0, = x, x = L
Making the integrand symmetric,
AB C D1 =
1
2
A( )A( )log x d dx
L
----xxx---
Module6:
Bodies of Revolution
Lecture 34:
Slender Body Theory
(Contd.)
2
Lift of slender bodies of revolution
Full solution = a + c
Velocity at any point can be obtained from this. Since C p contains square of velocity, it cannot be split
in two parts. Now
V 2 = U a +
+ U c cos +
+ U c sin
r
x
r
2
+
= +U a + 2U a
U a +
x body
x body x body
2
U 1
2
+ 2U
body
body
dR
U c cos +
U
dx
r body
c = cos =
( x)
a f ( x )
=
cos =
cos
r
r
r
U c cos + c 0
r body
(x ) = U c R 2 =
c = U c
A( x )
Uc
R2
R2
cos = U
sin cos
r
r
Hence
2
2
= (2U c sin ) (2U sin ) 2
U c sin
R body
To second order,
V 2 M
Cp 1 2 +
2
U
U
C pbody
2
=
U
2
dR
2
2
2
+
4 sin +
2
U x body
dx
x body
2
C pa
and
2
=
U
C pc =
2 c
U x
dR
body dx
+ 1 4 sin 2 2
body
C pc = 4
dR
cos + 1 4 sin 2 2
dx
dx
1
U 2 C pc Rd dx
2
N=
L 2
1
U 2 C pc ( cos )R d dx
0 0
2
= 4 q
cos 2 R
2
q = 2 U ,
= 2 q AB
L 2
dR
d dx 2 q cos 1 4 sin 2 Rd dx
0 0
dx
AB = A(L ) = R 2 (L )
C N = 2
Cross flow contribution to axial force is
A2 = q
R ( L ) 2
C pc R d dR = q [R(L ) ] = q A(L ) 2
2
C 2 = 2
NPTEL IIT Kharagpur: Prof. K.P. Sinhamahapatra, Dept. of Aerospace Engineering
A1 = q A(L ) C D1 + C pB
A = A1 + A2 ,
L = N cos A sin
D = N sin + A cos
2
C D1 + C pB + C 2
C L C N 1
2
C D C N + C D1 + C pB
Retaining only the dominant terms, assuming
C L = 2 ,
+ C 2 1
2
C D1 + C pB << 2 ,
C D = C D1 + C pB + 2
(referred to base area)
The drag increment due to angle of attack is called the induced drag (due to trailing vortices).
Induced drag coefficient is
C Di
CL
C Di = 2
Di 2
=
L
2
CDi
CL
CL
]
AR
The force vector due to angle of attack is midway between the normal to the body axis and the
normal to the flight path.
Module7:
Similarity Rules
Lecture35:
Similarity Rules
Cp
For supersonic flow over slender cones
Cp 1 M 2
= f M 2 1
= f x, y M 2 1
These results show that the parameters could be arranged into functional groups so that a single
curve represents the solution for a whole family of shapes and range of Mach numbers. These are
examples of similarity relations.
Since the solution of the equations of motion can be found in these special cases, the functions are
known explicitly. If the solutions are not easily obtainable, as in the non-linear cases of transonic or
hypersonic flow, the similarity analysis is especially useful. The similarity analysis is also useful when
solutions are known. The steady 2-D or axially symmetric flow of a perfect gas over a body of chord
length c and maximum thickness t is characterized by the following parameter,
cp = Cp x , M , , t
c
c
This result is obtained from dimensional analysis. The problem now is to find the functional form
which will represent C p in such a way that the five dimensionless variables are grouped into a smaller
a numbers of similarity parameters. The pressure coefficient parameter at a given station x can
then be represented by a single curve for all Mach numbers and gases and whole family of shapes.
Dimensional analysis lists the dimensionless parameters that are involved. Only certain amount of
book keeping is needed only the variables involved need to be known or guessed. Similarity
analysis shows how to group these dimensionless quantities in such a way as to reduce the number
of independent variables. The differential equations and boundary conditions are required. Some
integral relations may also be required. Set of experiments may also give similarity rules.
3
Most useful for non-linear problems
(x, y ) is the perturbation potential in a plane, steady flow with free stream mach number M 1
2
1 2
2+
=0
x 1 M 12 y 2
Boundary shape is given in the form
x
y = t1 f = 1 c
c
x
f , c is chord
c
1 = t1 c
or
is thickness ratio
y
x
= 1 f
c
c
( )
dy
= U 1
= U 1 1 f x
c
dx body
y y =0
Pressure coefficient C p1 on the boundary is
C p1 =
U 1 x y =0
Consider potential function ( , ) of a second flow in the ( , ) system and related to by the
relation
4
2
U1
U 1
1 M1
(x, y ) = A ( , ) = A x,
2
U2
U2
M
1
2
= x, =
1 M1
1 M 2
2
1 2
+
=0
2 1 M 1 2 2
if is solution corresponding to Mach number M 1 , then is a solution corresponding to M 2
Boundary condition gives
2
U1 1 M 1
= A
y
U 2 1 M 22
y =0
= U 1 1 f
c
=0
= U 1 1 f
c
=0
f is the same function in both cases, and
1 M 2
1
2 = A
1 M 22
A = A1
A2
Since f is same in both flows, only bodies of the same family can be compared.
Now
C p1 =
2
2
=
U 1 x y =0
U2
A
=0
C p2 =
2
U 2 =0
A = A1
A2
C p1 = AC p2
Two members of a family of shapes characterized by the thickness ratios 1 and 2 have pressure
distributions given by coefficients C p1 and C p2 . If the Mach numbers of the flows are M 1 and M 2 ,
then C p1 = AC p2 , provided that
1 = A
1 M1
1 M 2
= f
2
A
A 1 M
Cp
A is arbitrary, since the linearized equation is homogeneous in , can be multiplied by any constant
factor without changing the equation.
1.
2.
A =1 Cp = f
2
1 M
A=
1
1 M
Cp =
1
1 M
f ( )
[Prandtl-Glauert rule]
2
A = C p = f 1 M
3.
A=
4.
1
1
1 M 2
C
f
=
p
2
2
1 M
1 M
2) C p increases with M as
(1 M )
1
2 2
[Gothert rule]
1 M 2
constant
Mach number as 1 M
2 1
(1 M )
(1 M )
1
2 2
is replaced by
(M
2
= f 2
A 1 M 2
A
Cp
in
Example:
(C )
p 0.75
= (C p )0.40
1 0.40 2
= 1.38(C p )0.40
2
1 0.75
Module7:
Similarity Rules
Lecture36:
Similarity Rules (Contd.)
2
1 2 ( 1 + 1)M 1 1 2
+
=
2
U x x 2
x 2 1 M 12 y 2
1 M1
2
The equation is written for a flow with free stream Mach number M 1 and velocity U 1 in a gas with 1 .
Introducing ( , ) as before, it can be seen that ( , ) satisfies
1 2 ( 1 + 1)M 1 A 2
2
=
+
2
2
2 1 M 12 2
1 M 1 U 2
2
if is to satisfy the TSD equation for a flow with free stream speed M 2 ,U 2 in a gas with 2 , then
( 1 + 1)M 12 A = ( 2 + 1)M 2 2
1 M1
1 M 2
2 + 1 M 2 2 1 M 12
A=
1 + 1 M 12 1 M 2 2
Consideration of the boundary condition gives, as before
1 M1
A
2 = 1
2
1 M 2
2
and
C p1 = AC p2
3
Thus the similarity rule becomes
C p ( + 1)M
1 M
( + 1)M 2
= f
3
1 M 2 2
1 M 2
( + 1) M
2
C p ( + 1) M 2
1 M 2
= f
2
( + 1) M 2 3
= f ( )
Since A is not arbitrary, it is impossible to compare the same body at different Mach number or
different bodies at a given Mach number. Comparison is possible for bodies of different thickness
ratios at different Mach numbers in different gases so that
1 = 2 or
Flow is transonic if
1 M 12
1 ( 1 + 1) M
2
1
1 M 2
2
( 2 + 1)M 2
1 1
4
Linearized Axially Symmetric Flow
1 2 1
2
+
+
=0
2
2
2
r r
x 1 M 1 r
( x, r ) may be related to a second flow ( , R ) through the transformation
U1
1 M12
( x , r ) = A x, r
1 M 22
U 2
=
U
f
1 1
symmetric body, then the exact boundary condition is
r body
c
Since the bc cannot be applied at r = 0 , it is necessary to use the exact bc.
( )
= U1 1 f x
c
r r =1cf ( x c )
Now
U1 1 M 12
=A
r
U 2 1 M 2 2 R R =
r =1cf ( x c )
1 M12
1 M 22
1c f
c
= U 2 2 F
c
R R = 2c F c
1 M1
=2
2
1 M 2
2
( c)
1 f x
or
A =
1 M1
1 M1
x
A
f
1
2
2
1 M 2
1 M 2
c
2
1 M 2
1 M1
2
1
2
C p1 =
U 1 x 1c f x =0 U 1 1cf x =0
c
In terms of ,
2
C p1 =
U2
A 2 1 M 1
2
2
1 M 12
x
R
1
U
M
2
2
1c f = 0
2
1 M 2
C p1 = AC p2
Hence, the similarity law may be expressed as
2
1 M 12
1 M 2
cf
2 1
x
=0
c
= f
A 1 M 2
Cp
A
Since A = 1 M
2 1
2
2
C p1 1 M = f 1 M
Module7:
Similarity Rules
Lecture37:
Similarity Rules (Contd.)
2
Planar Flow
Planar boundary
x y
z = 1cf ,
c b
2
2
U1
1 M1
1 M1
,z
( x, y , z ) = A x, y
2
2
U2
1 M 2
1 M 2
2
2
1
1 M1
y
x
y 1 M1
=A
,
= =
2
2
b1 b2 b2 1 M 2 2
1 M 2
b2
1 M1
=
2
b1
1 M 2
2
Cp
A
= f
, b 1 M
A
1
M
Cp
A
= f
,
A 1 M 2
AR 1 M
is
C p ( + 1) M 2
1 M 2
,
=P 2
2
3
( + 1) M 2 3
AR 1 M
Cl = C p cos C p .
Cl = C p cos C p
C d = C p sin C p C p
CL ( + 1) M 2
CD ( + 1) M 2
1 M 2
=L 2
; AR 1 M 2
2
3
( + 1) M 2 3
1 M 2
;
=D 2
2
3
( + 1) M 2 3
AR M
Hypersonic Similarity
The potential equations and associated pressure coefficients looses their validity at sufficiently high
Mach numbers. The approximations used in deriving the TSP equation imply that the shock waves on
the body are weak. In supersonic flow, these shock waves lie close to a characteristic, Mach line,
4
whereas in transonic flow they are nearly normal. At very high Mach numbers, the Mach angle be of
the same order or less than the maximum deflection angle, , of the body. Since sin =
or
1
, this
M
M 1 .
This regime is thus characterized by the parameter M , or rather M . This parameter is called the
hypersonic similarity parameter K = M .
For large values of the hypersonic parameter, the difference between the shock angle and the
Mach angle is especially important. This is also true in case of transonic cases. In both cases, the
equations are non-linear but the non-linearity comes from different sources, since the nature of flow in
the two regimes are radically different. In the transonic case, the lateral extent of the field is large and
the changes along the flow direction are of main importance. In the hypersonic regime, the flow field
is narrow, since the shock is very close to the body, and, hence the changes normal to the direction
of flow are the most important ones. Furthermore, in hypersonic flow it is not possible, in general, to
assume irrotational flow, with a corresponding velocity potential. The shock waves are strong even for
thin bodies and the entropy gradients in the wakes of shocks are usually not negligible.
The relation between, M , , for an oblique shock is
M 2 sin 2 1 =
+1
2
M2
sin sin
cos( )
For small values of and large values of M such that M > 1, must also be small allowing the
approximations
M 2 2 1 =
+1
2
M 2
+1 +1
1
=
+
+
2
4
4 (M )
2
p 2 p1
2
M 2 2 1 M 2
p
+1
or
2 p 2 p1
1
2 +1 +1
= 2 = 2
Cp =
+
2
2
M p
4 4 (M )
C p = 2 f (M )
Cp
and the similarity rule is
= f (M )
Cp
= f (M )
+ 1 1 1
1
2
2
M 1 tan 1
M 2 1 tan 1 M 1 tan 1 M 2 1
tan
1
+1
+1
tan 1 x =
1
+ ...
x
Using
1
2 1
1 M
= 1+
M
1 M M
M
2
Cp =
2
M 2
2
( 1)
2 + ( 1)M
2
(
)
+
2
1
Cp =
or
Dividing by
2
M 2
CpM
1
M 1
M
2 1
1
=
M 1
1 +
M 2
Cp
= f (M )
Cp
or
= f (M )
Cp
= P(M )
CD
= D (M ) ,
C D based on chord
= f
,
2
A
A
M
1
Cp
For
A = 2,
Cp
1
=
= f
M
1
2
f M 1
Since, in a hypersonic flow M 1 M , the above rule is valid for the whole supersonic2
Cp
For a planar flow
= P M 1,
AR M 2 1
----xxx---
Module8:
Method of Characteristics
Lecture 38:
Method of Characteristics
Method of characteristics
In many cases, the accuracy of small perturbation theory, which is valid for slender bodies, is not
sufficient. It is necessary to include higher order approximation or to solve the exact equations. It is
rarely possible to solve the exact equations analytically.
The full, non-linear equations of motion for 2-D inviscid, irrotational flow are
(u
a2
) ux + u uy + x + (
a2
) y = 0
=0
x y
These can be extended to flow with vorticity by including the appropriate term in the right-hand side of
the second equation.
If
If
(u
+ 2
a2
(u
+ 2
a2
) <1
) >1
, the equations are hyperbolic and the numerical solution is obtained by the method
of characteristics. Transonic flow is a mixed case, and even numerical methods present problems
since the boundary between two regions is not known a-priori.
1. An equation is hyperbolic if a certain relation is satisfied by the coefficients of its highest order
(u
+ 2
a2
) >1
2. There are certain characteristic directions or lines in the (x y ) plane on which the normal
derivatives of the dependent variables (u , ) may be discontinuous. These lines are called
characteristics. In the present case the characteristics are the Mach lines. Because of the
nonlinearity, the characteristic network is not known a-priori.
Because the normal derivatives of the velocity on the Mach lines may be discontinuous, it
is possible to patch different flows together at these lines. The only restriction is that velocity
itself must be continuous.
3. On the characteristics, or Mach lines, the dependent variables satisfy a certain relation, known
as compatibility relations. This provides the key to the computation.
cot 2
=0
s n
1
=0
n s
The characteristics or Mach direction is explicitly introduced here by the expression
cot 2 = M 2 1
The form of these equations lends itself naturally to introduction of the Prandtl-Meyer function ,
which is a dimensionless measure of the speed defined by the relation
cot
dv
v
or
d = cot
dv
v
tan
=0
s
n
tan
=0
n s
We rewrite the equations in a coordinate system ( , ) which consists of the network of Mach lines.
The two sets of coordinates are related by the fact that the Mach lines are inclined at the angles to
the streamlines.
f =
Also,
f =
f f n
f
f
s
s + n = +
s
n
s n s
Comparison yields
f f f n
=
+
s s n s
NPTEL IIT Kharagpur: Prof. K.P. Sinhamahapatra, Dept. of Aerospace Engineering
or,
sec
f f
f
=
+ tan
s
n
Similarly
sec
f f
f
=
tan
n
s
The flow equations in (s, n ) coordinates may be written, by adding and subtracting, as
( ) + tan ( ) = 0
s
n
( + ) tan ( + ) = 0
s
n
( ) = 0
and
( + ) = 0
Hence,
Method of computation
Solutions
are
M , ,
,p
obtained
po
in
terms
of and .
The
values
of can
be
converted
into
The data, or, boundary conditions, are specified on a data curve. This must not be a characteristic. It
is now required to find the conditions at an arbitrary point P . Through P there are two characteristics,
one of type and the other of type. These two characteristics intersect the data curve at A and B. At
A and B and are known, and, hence Q and R on the characteristic curves. At P , the values of
Q and R are then obtained by choosing the appropriate one for each of the characteristics.
Q3 = Q1 ,
R3 = R2
or
3 + 3 = 1 + 1 , 3 3 = 2 2
Hence,
3 =
1
1
(1 + 2 ) + (1 2 )
2
2
3 =
1
1
(1 2 ) + (1 + 2 )
2
2
or,
1
1
(Q + R ) , = (Q R )
2
2
Thus, the solution for any arbitrary point P is very simple and elegant. But the solution is not yet
complete, for the location of the characteristics is not known a-priori.
Hence, the solution is essentially numerical one. To locate the characteristics a step-by-step
procedure must be used. By subdividing the region into a characteristic network, the mesh spacing
may be made small enough so that the mesh sides may be approximated by straight line segments.
For example, point 5 is located by using the known Mach angles and flow directions at 1 and 4 to
draw the characteristic segments. Then conditions at 5 are determined from the data at 1 and 4.
Similarly point 7 is found and then point 8 is found from 5 and 7. The computation proceeds outward
from the data curve.
The accuracy depends on the mesh size. In the 3rd figure a constructed mesh element is compared to
the true mesh which is defined by the true mesh of curved characteristics. The computed values
5 , 5 apply to point 5, but the solution scheme gives them for 5 . Iteration can be used to improve.
However, finer mesh is the best option.
p
whereas at a free boundary, as at the edge of a jet, the pressure ratio
Interior point
1
3
Q1 ,3
Q1 , 3
2
3
1
R is obtained from point 1. The other quantity is obtained from the shock equations(R-H relations) as
a relation between 3 and 3 . Thus, the condition at point 3 may be solved. These then determined
the shock wave angle , which can be used to draw the next shock segment.
Module8:
Method of Characteristics
Lecture 39:
Method of Characteristics
(Contd.)
p
whereas at a free boundary, as at the edge of a jet, the pressure ratio
Interior point
1
3
Q1 ,3
Q1 , 3
2
3
1
R is obtained from point 1. The other quantity is obtained from
the shock equations(R-H relations) as
a relation between 3 and 3 . Thus, the condition at point 3 may be solved. These then determined
the shock wave angle , which can be used to draw the next shock segment.
cot 2 v sin
sin
tan
= tan
v s n
r
s
n
r
1 v
= 0 tan
=0
n s
v n s
NPTEL IIT Kharagpur: Prof. K.P. Sinhamahapatra, Dept. of Aerospace Engineering
Compatibility relations
sin
( ) = sin
sin
( + ) = sin
r
3
sin
d
(
)
2
2 sin r
3
d ;
3
sin
d ( + ) = sin
1
r
(3 3 ) (2 2 ) = sin 2
sin 2
23
r2
(3 + 3 ) (1 + 1 ) = sin 1
sin 1
13
r1
or
3 =
sin
sin 2
1
1
1
23
(1 + 2 ) + (1 2 ) + sin 1 1 13 + sin 2
2
2
2
r1
r2
3 =
sin
sin 2
1
1
1
23
(1 2 ) + (1 + 2 ) + sin 1 1 13 sin 2
2
2
2
r1
r2
cot 2 v sin
=
v s n
r
T dS 1 dho
1 v
= 2
+ 2
v n s
dn dn
( ) = sin
r
dn dn
( + ) = sin
dn dn
r
3 3 = 2 2 + sin 2
sin 2
cot 2
T2 ( S3 S 2 ) ho3 ho2
23
2
2
r2
3 + 3 = 1 + 1 + sin 1
sin 1
cot 1
T ( S S ) ho3 ho1
13
r1
12 1 3 1
S 3 and ho 3 are needed for the computation. Once 3 is located, the streamline through it may
approximately be located by drawing a line with slope 3 =
1 + 2
2
S and ho are invariant along a stream line and their values at 3 are the same as at 3 .
Plane flow theorems
We have for plane supersonic flow
+ = Q
on
= R
on
These relations are independent of specific flow geometry. These lead to three kinds of flow field (1)
general or non-simple region, (2) simple region, or simple wave, (3) uniform region or constant state.
In the general region the characteristics are curved and each one corresponds to one value of
Q+R
QR
, =
2
2
Along an -characteristic R is constant and so the changes in and depend only on changes in Q
owing to crossing of -characteristic.
1
Q =
2
1
R =
2
In a simple region or simple wave, the one of the invariants is constant throughout the region. Then
and are individually constant along the other characteristic and it must therefore be straight line1
with uniform conditions on each one. The flow changes in crossing the straight characteristics are
related by
=
A constant state or uniform flow is one in which both the invariants are constant throughout. and
are uniform and both sets of characteristics consist of straight lines forming a parallel network.
The usual convention is to omit the Mach lines in the uniform region, the show only the straight lines
in a simple wave and both sets in the non-simple region.
The uniform region does not adjoin the non-simple region except at one point. This is a general
theorem.
Module8:
Method of Characteristics
Lecture 40:
Method of Characteristics
(Contd.)
n
n
(a) The wall and flow are continuous. Several characteristics of the straight family are shown. As
many as can be drawn from arbitrary points on the wall. Other family is not shown.
(b) The wall is approximated by a series of straight line segments, meeting at finite angles. The
expansion fans or waves which originate at the corner divide the field into segments of
uniform flow. Since, these are simple waves, the flow changes across them are related by the
same relations.
=
The positive sign is for waves from the upper wall and negative sign is for waves from the
lower wall. being considered positive when counterclockwise.
(c) Further approximation to replace each fan by a single line, the central one. The changes at
the waves are now discontinuous, but this approximation is no greater than in representing
the wall by straight segments.
If the wall is concave, it gives a continuous compression, which may be approximated by a series of
weak shocks. The above equations for the flow changes apply without modification. The position of
each wave is midway between the characteristics of the uniform flow fields which it separates. This is
the same rule as for the expansion waves above.
In summary, the method of weak waves is to replace a continuously varying flow by a series of
uniform segments with steps between them. These steps occur at the positions of the waves. To
extend them to non-simple flows the rules for intersection of waves of opposite family are needed.
Interaction of waves
b. Approximate form
The use of finite weak waves for the construction of plane flows depends on the following theorem
The strength of a weak wave is not affected by intersection with other waves.
The strength here is defined as the flow deflection which the wave produces.
(a)
Shows the intersection of two simple expansions, for which the theorem is easily proved by
noting that the flow crosses the same characteristics on either side of the intersection. In (b) the
expansion waves are represented by single lines.
With these it is possible to set up a single, systematic computing method. We first consider an
example with expansion waves. These waves (represented as a single wave) divide the flow into a
number of cells, in each of which the flow is uniform. It is required to calculate and in each cell.
3, 0
2, 0
3, 1
1, 0
2, 1
3, 2
2, 2
1, 1
1
3, 3
2, 3
1, 2
0, 1
1, 3
0, 2
0, 3
Expansion waves from the upper wall deflect the flow upward (+ve) and increase . Those from the
lower wall deflect the flow downward 1and also increase . To reach a cell the flow crosses m waves
of - type and n waves of type. Thus the condition in (m, n) cell are given by
= 1 + i
i =1
i =1
= 1 + + i
i =1
i =1
1 = m n
1 = m + n
Conditions in a cell are obtained by simply adding up the number of waves of each kind that are
crossed.
The compression waves can also be included in the result. If the flow crosses k compression waves
from the upper wall and l compression waves from the lower wall then
1 = m n k + l
1 = m + n k l
Complete construction of the flow requires rule for reflection and cancellation of waves. On reflection
from a solid wall a wave of - type is changed to one of - type. The turning strength of the reflected
wave is the same as that of the incident one since the flow must return to the original direction parallel
to the wall.
Reflection
Cancellation
By accommodating the wall to the flow direction after the incident wave, the reflected wave may be
cancelled. The wall deflection should be equal to the strength of the wave.
In designing supersonic nozzle, all the waves produced in the upstream part of the nozzle must be
cancelled before they reach the test section.
Initial
Expansion
1
A*
2
Cancellation
AT
= 1.283
A*
T=16
s li
o n
n e
i
c
1
AT
2
MT=1.639
Reflection
There are four such waves, the corresponding maximum value of the wall deflection
being max = 8D . The inclination of each wave is midway between the Mach lines on either side of
it. The inclination of the first wave, measured relative to the centerline is
1
[(90 0) + (62.0 2)] = 75
2
2. The waves are reflected from the centerline of the symmetrical nozzle. (Alternatively, the
centerline might be the wall of an asymmetrical nozzle).
3. In the region of cancellation the reflected waves are canceled by deflecting the wall 2 at each
wave intersection. After the last wave the wall is parallel to the centerline and the flow is uniform.
It has the value of T = 16D having crossed eight expansion waves of 2 each.
1
Rule: Value of the initial expansion (maximum) wall deflection is given by max = T
2
4. The final test section height must agree with the value
T = 16 AT
AT
A Corresponding
= 1.283
to
a) The length of the initial expansion contour is arbitrary. The shortest nozzle is obtained by
having zero length for this region a Prandtl-Meyer expansion to max . But such rapid
expansion gives troublesome effects in the side wall boundary layer.
b) The construction has been started at the narrowest part, i.e., at throat assuming the sonic line
to be straight. But it is actually curved and meets the wall somewhat upstream of the throat,
the details depending on the throat shape. For greater accuracy, it may be necessary to use
transonic solution for the throat region to start the supersonic solution correctly.
c) The problem above may be avoided by starting the solution well downstream of the throat, in
the initial expansion, where the flow may be assumed radial. This is reasonable approximation
if the wall curvature is not too great. The Mach number in the radial flow is obtained from the
area relation A
, with A being the area of the curved cross section over which M is uniform.
1
T is valid only if all the waves are cancelled after one reflection form the
2
centerline. But if they are reflected back to the wall and reflected there once more, then
1
4
max = T as the flow then cross the expansion four times. Partial cancellation may also be
used. In these cases, the nozzle will be longer.
e) Another approach is to specify the centerline Mach distribution. It should fair smoothly
into M T and its slope should be consistent with the actual transonic throat flow that depends
on the throat shape. This approach is particularly suitable if the method of characteristic,
instead of method of waves, is used. Convenient for axially symmetric nozzle.
f)
The boundary layer on the nozzle and side walls has a displacing effect which reduces the
effective height and width of the nozzle. Allowance for this is to be made by adding a
correction for boundary layer. The side walls should also diverge to allow for their boundary
layers.
2. Computation with waves is convenient only in plane flow, since it depends on the theorem that
the strength of a wave does not change after intersections and reflections. In axially
symmetric flow and in general 3-D flow the strength of a wave varies continuously.
3. The wave method is more intuitive in plane flow than the characteristic method and is usually
preferred. In some problem it is more convenient because of the idea of wave cancellation
to determine a boundary shape.
----xxx---