You are on page 1of 47

Compressible Flow

Hari Prasad Neopane, PhD


Associate Professor
Department of Mechanical Engineering
Kathmandu University
hari@ku.edu.np
Compressible Flow
• Defined as that flow in which the density of the
fluid does not remain constant during flow
• Examples:
• Flow of gases through orifices and nozzles,
• Flow of gases in machines such as compressors, and
• Projectiles and airplanes flying at high altitude with
high velocities
• The change in density of a fluid is accompanied
by the changes in and pressure and temperature
and hence the thermodynamic behavior of the
fluids needs to be taken into consideration
Thermodynamic Relations
• Equation of state: The equation which gives the
relationship between the pr., temp. and sp. vol.
p
• Perfect gas pV  mRT ; pv  RT ;  RT

• Expansion and compression of perfect gas
• When the expansion or compression of a perfect
gas takes place, the pressure, temperature and
density are changed. The change in pressure,
temperature and density of a gas by two process:
• Isothermal Process, p
pV  const .; pv   const.

• Adiabatic Process,   p
pV  const.; pv  
 const.

Basic Equations of Compressible Flow

• Continuity equation
• Bernoulli’s equation or energy equation
• Momentum equation
• Equation of state
Continuity Equation
• Based on the law of conservation of mass which
states that matter cannot be created not
destroyed. Or in other words, the matter or mass
is constant. For one-dimensional steady flow,
the mass per second, AV
• As mass is constant, AV  const
or d AV   0
or d ( AV )  AVd  0
or [ AdV  VdA]  AVd  0
dividing by AV
dV dA d
  0
V A 
Bernoulli’s Equation
• As we know Euler’s equation, dp
 V  dV  g  dZ1  0

• Integrating we get,  dp   V  dV   g  d  Z1  Const.



dp V 2
Or    g  Z  Const.
 2
• Isothermal Process: p
 const.  c1   
p
 c1
2 2
p1 V p V
log e p1  1  Z1  2 log e p2  2  Z 2
1 g 2g 2 g 2g
• Adiabatic Process: p  p  p
1/ 


 const .  c2        
 c2  c2 
   p1 V1    p2 V2
2 2
    Z1      Z2
   1  1 g 2 g   1  2 g 2g
Problem 1
• A gas is flowing through a horizontal pipe at a temperature
of 4 0C. The diameter of the pipe is 8 cm at a section 1-1 in
this pipe, the pressure is 30.3 N/cm2 (gauge). The diameter
of the pipe changes from 8 cm to 4 cm at the section 2-2,
where pressure is 20.3 N/cm2 (gauge). Find the velocities of
the gas at these sections assuming an isothermal process.
Take R= 287.14 Nm/kg K, and atmospheric pressure =
10N/cm2.
• Solution Clue:
• Applying continuity equation at section 1-2 and find ratio of velocity in
terms of density ratio
p1 p2
• Use isothermal relation,    , and find V1 and V2 relation
1 2
• Use Bernoulli’s equation at 1-2 to find ratio of p1 and use p1  RT1
1 1
• ANS: V1= 40.75 m/s and V2 = 216.79 m/s
Problem 2
• A gas with a velocity of 300 m/s is flowing through a
horizontal pipe at a section where pressure is 6 X 104
N/m2 (absolute) and temperature 40 0C. The pipe
changes in diameter and at this section the pressure is 9
X104 N/m2. Find the velocity of the gas at this section if
the flow of the gas is adiabatic. Take R = 287 J/kg 0K
and Cp/Cv =1.4
• Solution Clue:
• Apply Bernoulli’s equation at section 1-2
• Use relation of adiabatic flow p1  p2
 
• Solve above two equations 1  2

p1
• Use  RT1
1
• Ans: V2 = 113 m/s
Stagnation Properties
• Recall definition of enthalpy

• Which is the sum of internal


energy ‘u’ and flow energy
P/
• For high-speed flows,
enthalpy and kinetic energy
are combined into
stagnation enthalpy h0
Stagnation Properties
• Steady adiabatic flow through
duct with no shaft/electrical
work and no change in
elevation and potential energy

• Therefore, stagnation enthalpy


remains constant during
steady-flow process
Stagnation Properties
• If a fluid were brought to a complete stop (V2 = 0)

• Therefore, h0 represents the enthalpy of a fluid when


it is brought to rest adiabatically.
• During a stagnation process, kinetic energy is
converted to enthalpy.
• Properties at this point are called stagnation
properties (which are identified by subscript 0)
Stagnation Properties
• If the process is also reversible,
the stagnation state is called the
isentropic stagnation state.
• Stagnation enthalpy is the same
for isentropic and actual
stagnation states
• Actual stagnation pressure P0,act is
lower than ‘P0’ due to increase in
entropy ‘s’ as a result of fluid
friction.
• Nonetheless, stagnation
processes are often approximated
to be isentropic, and isentropic
properties are referred to as
stagnation properties
Stagnation Properties
• For an ideal gas, h = CpT, which allows the h0 to be
rewritten

• T0 is the stagnation temperature. It represents the


temperature an ideal gas attains when it is brought to rest
adiabatically.
• V2/2Cp corresponds to the temperature rise, and is called the
dynamic temperature
• For ideal gas with constant specific heats, stagnation
pressure and density can be expressed as
Stagnation Properties
• When using stagnation enthalpies, there is no
need to explicitly use kinetic energy in the
energy balance.

• Where h01 and h02 are stagnation enthalpies at


states 1 and 2.
• If the fluid is an ideal gas with constant specific
heats
Sonic Velocity or Velocity of Sound
• Consider a one-dimentional flow through long straight rigid
pipe of uniform cross-sectional area filled with a frictionless
piston at one end as shown in Fig. The tube is filled with a
compressible fluid initially at rest

• If the piston is moved suddenly to the right with a velocity, a


pressure wave would be propagated through the fluid with a
velocity of sound wave
Velocity of Sound Contd..
• Let ‘A’ = cross-sectional area of the pipe, ‘V’ = piston velocity, ‘p’ = fluid
pressure in the pipe before the piston movement,  = fluid density
before the piston movement, ‘dt’ = a small interval of time during which
piston moves, and ‘C’ = velocity of pressure wave or sound wave
(travelling in the fluid)
• Before the movement of the piston the length ‘dL’ has an initial density 
and its total mass =   dL  A
• When the piston moves through a distance ‘dx’ the fluid density within
the compressed region of length (dL-dx) will be increased and becomes
  d  & subsequently the total mass of fluid in the compressed region =
  d  dL  dx A
   dL  A    d   dL  dx   Aby continuity
but dL  C  dt and dx  V  dt
  C  dt    d   C  V   dt or   C    d   C  V 
V
or d  C    V or C   eq.1
d
Velocity of Sound Contd..
• Further in the region of compressed fluid, the fluid particles have attained
a velocity which is apparently to ‘V’ (velocity of the piston), accompanied
by an increase in pressure ‘dp’ due to sudden motion of the piston.
Applying impulse-momentum equation for the fluid in the compressed
region during ‘dt’, we get
dp  A  dt    dL  A  V  0 
Rate of change of
dL C  dt
Force on the fluid
or dp     V    V    C V momentum
dt dt
dp
or C   eq.2
V
V dp dp
Eq.1 2, C 2   
d V d
dp
or C 
d
Velocity of Sound Contd..
• Sonic Velocity in terms of bulk modulus
K dp 1 dv d
C  K Also v  Or v  const Or  
  dv   v 
 
 v 

• Sonic Velocity for isothermal process


p
p   const. Differenting both sides ,
C  R T 

  dp  p  d dp p  d dp p
 0 Or  0    RT
2  2 d 
• Sonic velocity for adiabatic process

 p     const. Differenting both sides ,
C    R T
p      1d    dp  0 
dp p
     R T
d 
Velocity of Sound Contd..

• Since
• ‘R’ is constant
• ‘k’ is only a function of T
• Speed of sound is only a
function of temperature
Mach Number
• Second important
parameter is the
Mach number Ma
• Ratio of fluid velocity
to the speed of sound

Ma < 1 : Subsonic
Ma = 1 : Sonic • Flow regimes
Ma > 1 : Supersonic classified in terms of
Ma >> 1 : Hypersonic Ma
Ma  1 : Transonic
Mach Number Contd..
• ‘Ma’ is up to 0.2 is the incompressible flow
region
• Here the fluid flow can be treated
incompressible without introducing any
appreciable error
• ‘Ma’ is < unity- subsonic flow region
• If ‘Ma’ is higher than 0.2 and the
compressibility effects consideration
• ‘Ma’ varies between the values slightly
higher/less than unity is the transonic flow
region
• If ‘Ma’ 1 then flow is sonic
• If ‘Ma’ >1 flow is super sonic
• If ‘Ma’ > 5 flow is hypersonic
Propagation of Pressure Waves
1D C 1 1
sin     
1B V V / C M

Mach Cone: Tangent to the different circles which represent the propagated
spherical waves on both sides, cone having vortex at ‘B’
Mach angle: Defined as the half of the angle of the Mach cone
Zone of action: When ‘M’ >1, the effect of the disturbance is felt only in the
region inside the Mach cone. This region is called the zone of action
Zone of Silence: When ‘M’ >1, there is no effect of disturbance in the region
outside the Mach cone, is called zone of silence
One-Dimensional Isentropic Flow
• For flow through
nozzles, diffusers, and
turbine blade passages,
flow quantities vary
primarily in the flow
direction
• Can be approximated as
1D isentropic flow
• Consider example of
Converging-Diverging
Duct
One-Dimensional Isentropic Flow
• Example illustrates
• Ma = 1 at the location of the
smallest flow area, called the
throat
• Velocity continues to increase
past the throat, and is due to
decrease in density
• Area decreases, and then
increases. Known as a
converging - diverging
nozzle. Used to accelerate
gases to supersonic speeds.
One-Dimensional Isentropic Flow
Variation of Fluid Velocity with Flow Area
• Relationship between V, , and A are complex
• Derive relationship using continuity, Euler, and
velocity of sound equations
• Continuity

• Differentiate and divide by mass flow rate (AV)

eq. 1
Area Velocity Relationship
• The Euler’s equation for compressible fluid is given by the equation,
dp
 V  dV  g  dz  0
• Neglecting the ‘z’ terms, dp
 V  dV  0


dp d
• This equation can also be written as,   V  dV  0
 d
dp d
or   V  dV  0
d 
dp
But  C2
d
d
Hence, C 2   V  dV  0 eq. 2

dA V  dV dV dV V 2 
From eq. 1 & eq. 2  2
   2  1
A C V V C 

dA dV
A

V

M 2 1 
Area Velocity Relationship Contd..
dA dV
A

V

M 2 1
• For M < 1, the flow is sub-sonic and the right-hand side of
eq. is negative as (M2-1) is negative for the values of M < 1.
Hence dA/A > 0, dV/V < 0. This means that with the increase
of area, the velocity decreases and vice versa
• For M >1, the flow is super-sonic. The value of (M2-1) will be
positive and hence right-hand side of eq. will be positive.
Hence dA/A > 0 and also dV/V > 0. This means that with the
increase of area, velocity also increases
• For M = 1, the flow is called sonic flow. The value of (M2-1)
is zero. Hence right-hand side of eq. will be zero. Hence
dA/A =0. This means area is constant.
Flow through Nozzle & Diffuser

A diffuser is a passage of varying cross-section that serves to


achieve reduction in velocity of the flowing fluid
One-Dimensional Isentropic Flow
Variation of Fluid Velocity with Flow Area
Comparison of flow properties in subsonic and supersonic nozzles and diffusers
One-Dimensional Isentropic Flow
Property Relations for Isentropic Flow of Ideal Gases
• Relations between static properties and stagnation properties in
terms of ‘Ma’ are useful.
• The stagnation temperature for an ideal gas is,

• Using definitions, the dynamic temperature term can be expressed


in terms of ‘Ma’
One-Dimensional Isentropic Flow
Property Relations for Isentropic Flow of Ideal Gases
• Substituting T0/T ratio into P0/P and 0/
relations

• For ‘Ma’ = 1, these ratios are called critical


ratios
One-Dimensional Isentropic Flow
Property Relations for Isentropic Flow of Ideal Gases
Compressible Fluid Flow through
 k  p1
Nozzles
V  k  p2
2
V
2

   1  Z1     2  Z2
 k  1  1 g 2 g  k 1  2 g 2g
but Z1  Z 2 and also V1  0
 k  p1  k  p2 V2
2

  0  
 k  1  1 g  k  1  g 2 g
 k   p1 p2  V2
2
or    
 k  1   1 g  2 g  2 g
 k   p1 p2  V2
2

or       
 k  1   1  2  2 g
2k p p  2k p  p  
or V2   1  2    1  1  2  1   eq.1
(k  1)  1  2  ( k  1) 1   2 p1 
but for adiabatic flow, Pr. drop is small: Isothermal
k
p1 p2 p1  1  Pr. drop is large: Adiabatic
 or  
1k 2k p2   2 
1/ k
1  p1 
    eq.2
 2  p2 
Put 2 in 1,

p   p  p 
1/ k 11 / k
2k p  p  2k 
V2   1  1  2   1    1  1   2  
( k  1) 1  p1  p2   ( k  1) 1   p1  
 

p  p 
k 1 / k
2k 
  1 1   2  
(k  1) 1   p1  

p1  
k 1
p 2k
Let n 2 then, V2    1  n 
k
p1 ( k  1) 1  
Now, m   2 A2  V2 (Mass rate of flow of compressible fluid)
2k p  k 1

  2  A2   1 1  n 
k
( k  1) 1  

2 
k 1
2k p1
 A2     2 1  n k 
( k  1) 1  
1/ k
1  p 
But from eq.2, 2  1/ k
 1   2   1  n1/ k
 p1   p1 
 
 2
p
2
  2  1  n
2 2 k

2k p1  k 1

m  A2    1  n 1  n k 
2 2/ k
Now eq.
( k  1) 1  
2k  k2  
k 1 2
 A2   p1 1 n  n k k

( k  1)  
2k  k2 k 1

 A2   p1 1  n  n k 
k 1  
Value of ‘n’ or p2/p1 for max. value of
mass rate of flow
m
We have 0
n
  k 
k 1 k 1 k 1
2 k 1 k  1
2 2
1 2
n  n   0  n  n 0  n 
k k k
Or
n   k k k 1
For k  1.4, n  0.528

Value of V2 for Maximum Rate of flow of Fluid


2k p1  k  1 2k p1
V2     
k  1 1  k  1 k  1 1

Maximum rate of flow of fluid through Nozzle


mmax  0.685  A2  p1  1
Variation of mass flow rate with p2/p1
• A passage in which the sonic velocity has been
reached and thus in which the flow rate is
maximum is often said to be choked or in
choking conditions
• From the curve it is found that the flow rate
increases with the decrease of pressure ratio
and attains the maximum value at critical pr.
ratio = 0.528 for air thereafter contradiction occur
• Why: Once the velocity at throat reaches the sonic value, no
message can be transmitted upstream and so the fluid there is quite
unaware of any further reduction of pressure at the downstream end
of the nozzle
• The convergent part of the nozzle continues to behave
independently of conditions beyond the throat until these becomes
subsonic again
Velocity at outlet of Nozzle
• Velocity at outlet of Nozzle for Max. rate of flow is equal
to Sonic velocity (V2 = C2) : How ?
• As we know, 2k p1  k  1 2k p1
V2  
    eq. 1
k  1 1  k  1 k  1 1

• Use p2 p2
 n  p1 
p1 n

• Also for adiabatic flow, p1



p2
Or
p 
1   2  2 
1 / k

  2  n 1/ k
1 k
2 k
 p1 

• Put value of p1 and 1 in eq. 1, then use,


k 1
2
n k

k 1

• Then V2 = C2
Mass flow rate through Venturimeter
• Derivation Self study

 k 1

2k p1   p2   k
 1   
k  1 1   p1  
 
m   2  A2 V2   2  A2 
  A 2   p 2 / k 
1   22    2  
  A1   p1  

• The only condition for above equation is that


pressure ratio should be more than the pressure
ratio 0.528
Problem 1
• A tank contains air at a temperature of 30 0C. Air flows from the tank into
atmosphere through a convergent nozzle. The diameter at the outlet of
the nozzle is 25 mm. Assuming adiabatic flow, find the mass rate of flow
of air through the nozzle when the pressure of air in tank is 3.924 N/cm2
(gauge), 33.354 N/cm2 (gauge). Take k = 1.4, R = 287 J/kg0K and
atmospheric pressure = 10.104 N/cm2 (abs)
• Solution Clue:
• Case 1: n = p2/p1= 0.7203 which is more than pressure ratio 0.528 so
mass flow rate of air is given by 2k  k2 k 1

m  A2   p1  1 n  n k   0.146 kg / s
where k  1  
p1 p1
 RT1 Or 1   1.613 kg / m 3
1 RT1

• Case 2: n = p2/p1= 0.2325 which is less than pressure ratio 0.528 so


mass flow rate of flow will be corresponding to the pressure ratio of
0.528 (Ans: 0.494 kg/s)
Problem 2
• Find the mass rate of flow of air through a venturimenter having inlet
diameter as 300 mm and throat diameter 150 mm. The pressure at
the inlet of venturimenter is 14 N/cm2 absolute and temperature of
air at inlet is 15 0C, The pressure at the throat is given as 13 N/cm2
absolute. Take R= 287 J/kg 0K and k=1.4.
• Solution Clue:
p p
• Here n = p2/p1= 0.9285 and density of air 1  RT1 Or 1  RT1  1.66 kg / m3
1 1

• For adiabatic process, p1 p2 p 


1/ k

 Or  2  1  2   1.57 kg / m3
1 k
2 k
 p1 
• Mass flow rate through venturimeter,
 k 1

2k p1   p2  k 
 1  
k  1 1   p1  
 
m   2  A2 V2   2  A2   315 kg / s
  A 2   p 2 / k 
1   22    2  
  A1   p1  
Isentropic Flow Through Nozzles
• Converging or converging-diverging nozzles are
found in many engineering applications
• Steam and gas turbines, aircraft and spacecraft
propulsion, industrial blast nozzles, torch nozzles
• Here, we will study the effects of back pressure
(pressure at discharge) on the exit velocity,
mass flow rate, and pressure distribution along
the nozzle
Isentropic Flow Through Nozzles
Converging Nozzles
• State 1: Pb = P0, there is no
flow, and pressure is constant.
• State 2: Pb < P0, pressure along
nozzle decreases.
• State 3: Pb =P* , flow at exit is
sonic, creating maximum flow
rate called choked flow.
• State 4: Pb < P *, there is no
change in flow or pressure
distribution in comparison to
state 3
• State 5: Pb =0, same as state 4.
Isentropic Flow Through Nozzles
Converging Nozzles
• Under steady flow conditions, mass flow rate is
constant

• Substituting value of T and P, we get,

• Mass flow rate is a function of stagnation properties,


flow area, and Ma
Isentropic Flow Through Nozzles
Converging Nozzles
• The maximum mass flow rate through a nozzle
with a given throat area A* is fixed by the P0 and
T0 and occurs at Ma = 1

• This principal is important for chemical


processes, medical devices, flow meters, and
anywhere the mass flux of a gas must be known
and controlled.
Isentropic Flow Through Nozzles
Converging-Diverging Nozzles
• The highest velocity in a converging nozzle
is limited to the sonic velocity (Ma = 1),
which occurs at the exit plane (throat) of the
nozzle
• Accelerating a fluid to supersonic velocities
(Ma > 1) requires a diverging flow section
• Converging-diverging (C-D) nozzle
• Standard equipment in supersonic aircraft and
rocket propulsion
• Forcing fluid through a C-D nozzle does not
guarantee supersonic velocity
• Requires proper back pressure ‘Pb’
Isentropic Flow Through Nozzles
Converging-Diverging Nozzles
• P0 > Pb > Pc
• Flow remains subsonic, and
mass flow is less than for
choked flow. Diverging section
acts as diffuser
• Pb = PC
• Sonic flow achieved at throat.
Diverging section acts as
diffuser. Subsonic flow at exit.
Further decrease in ‘Pb’ has no
effect on flow in converging
portion of nozzle
Isentropic Flow Through Nozzles
Converging-Diverging Nozzles
• PC > Pb > PE
• Fluid is accelerated to supersonic
velocities in the diverging section as
the pressure decreases. However,
acceleration stops at location of
normal shock. Fluid decelerates and
is subsonic at outlet. As ‘Pb’ is
decreased, shock approaches nozzle
exit.
• PE > Pb > 0
• Flow in diverging section is
supersonic with no shock forming in
the nozzle. Without shock, flow in
nozzle can be treated as isentropic.

You might also like