QUASI-ONE-DIMENSIONAL

FLOW

Chapter 5
 5.1 Introduction
We will be dealing with nozzles and diffusers.
Convergent Divergent Nozzle: designed to smoothly
acclerate flow to supersonic
Convergent Divergent Diffuser: designed to slow
supersonic airstream to subsonic (thru multiple
shocks) with minimum pressure loss

We will assume that flow properties only vary

axially along the length of a duct. Changes in the
axial flow properties result from a cross-
sectional area change in the duct.
Previously, we dealt with a constant cross-sectional
area duct and it was Normal shock, friction or heat
addition which caused change in properties
We relax the constant cross-sectional area
constraint by allowing the stream-tube area, A, to
vary with distance, x.
 A=A(x)
p=p(x)
ρ=ρ(x)
T=T(x)
V=u=u(x)

Is the flow in the stream-tube truly one-dimensional?

No, Quasi 1-D flow is an approximation and it is in reality 3D and
require high order methods. But for most engineering problems,
the quasi-one-dimensional results are usually sufficient (ex. flow
through wind tunnels, rocket engines, converging-diverging
nozzles) for analysis as well as to understand Compressible flows.
 5.2 Governing Equations

The quasi-one-dimension assumption is actually three-dimensional, and

the exact solution requires a three-dimensional numerical simulation.
Quasi 1D means that properties are uniform across any cross section –
which is an approximation i.e. instead of being dependent on x,y,z ;
properties are only dependent on x
The quasi-one-dimensional assumption is an approximation to the actual
flow physics; however, the governing equations will exactly enforce the
mass conservation, Newton's second law (momentum), and the first law
of thermodynamics (energy) for the quasi-one-dimensional flow.
This is accomplished by utilizing the integral forms of the conservation
equations.
Let us consider a control volume for a quasi-one-dimensional flow.

■ For a steady,quasi-1D flow

  
The continuity equation :  v ds  
 d
s
t 

1u1 A1  2u2 A2 …… 5.1)

Hence the differential form of mass

conservation equation is

Figure 5-6: Finite Control Volume for

Quasi One Dimensional Flow
Why do we not include the integral taken over the sides of the control
volume between locations 1 and 2? The velocity is oriented along the
surface; no flow goes over the surface or V.dS=0 along the side .
 The integral form of momentum conservation equation for inviscid flow is,
 
  ( v ) 

s
( v d s ) v  
 t
d  

f d  
s
p ds

A2 
X-dir p1 A1   u A1  ( A
2
1 1 pdA) x  p2 A2   2u22 A2
1

This is momentum eqn for steady quasi 1D flow.

It is not an algebraic eqn because of integral term which represents the
force on the sides of the CS between locations 1 and 2
The energy equation
     V2 V2 


q  d  
pv ds  
s
 ( f v )d [  (e  )]d  
  t 2
 (e  )v ds
s
2
Assuming adiabatic, steady flow with no body forces. Using Continuity eq

h e  p

u12 u 22 total enthalpy is
h1  h2  h0 const constant along the flow
2 2
Actually, the total enthalpy is constant along a streamline in any adiabatic steady flow
Continuity eqn In differential forms uA const
Momentum eqn for small CV  d ( uA) 0 P P +dP
pA  u 2 A  pdA  A A +dA
u u +du
( p  dp )( A  dA)  (   d )(u  du ) 2 ( A  dA) ρ ρ+dρ

Dropping 2nd order terms

Adp  Au 2 d  u 2 dA  2 uAdu 0 (1) dx

d ( uA) 0Multiplying by u u 2 dA  uAdu  Au 2 d 0 (2)

(1) - (2) = Adp  uAdu 0
dp()  du( )
𝑑𝑝=− 𝜌 𝑢𝑑𝑢 Euler’s equation
dp( )  du()

Differential form of Energy eqn

u2 u2
 h  const d (h  ) 0
2 2

dh  udu 0
 5.3 Area-Velocity Relation

d ( uA) 0

From Euler eqn

udA  Adu  Aud 0
dP dP d
uA   udu
 d 
∵ adiabatic & inviscid
∴ no dissipation mechanism
d du dA
  0
 u A → isentropic
 P  2
  a
dA 2 du    S
( M  1)
A u d udu u 2 du 2 du
 2  2  M
Area Velocity Relation  a au u
Important information dA 2 du
( M  1)
1. M→0 incompressible flow, A u
Au=const consistent with the familiar continuity eq for incompressible flow.
2. 0≦M ＜ 1 subsonic flow,
an increase in velocity (du+) is associated with a decrease in area (dA ，－ )
and vice versa.
3. M>1 supersonic flow
an increase in velocity is associated with an increase in area , and vice versa.
4. M=1 sonic flow →dA/A=0; a minimum or maximum in the area. Min area is
realistic!!!
dA du
( M 2  1)
A u
Euler eqn
dp=-

Area, Pressure, M relation

  Subsonic flow through area INCREASE

M< 𝑑𝐴
𝑑𝑃 ( 1− 𝑀 ) =𝜌 𝑢
2 2
𝐴
1 PRESSURE INCREASES
VELOCITY DECREASES

 Subsonic flow through area DECREASE

𝑑𝐴
𝑑𝑃 ( 1− 𝑀 ) =𝜌 𝑢
2 2
M<1 𝐴
PRESSURE DECREASES
VELOCITY INCREASES
 Supersonic flow through area INCREASE

𝑑𝐴
𝑑𝑃 ( 1− 𝑀 ) =𝜌 𝑢
2 2
M>1
𝐴
PRESSURE DECREASES
VELOCITY INCREASES

 Supersonic flow through area DECREASE

𝑑𝐴
𝑑𝑃 ( 1− 𝑀 ) =𝜌 𝑢
2 2

M>1 𝐴
PRESSURE INCREASES
VELOCITY DECREASES
For a gas to expand isentropically from subsonic to supersonic,
it must flow thru Convergent Divergent duct & vice versa
At min area (THROAT), flow is SONIC
A subsonic flow is to be
accelerated isentropically
from subsonic to supersonic
In Converging: A, u, p 
In Diverging: A , u, p 

Supersonic flow is to be
decelerated isentropically from
supersonic to subsonic
In Converging: A, u , p 
In Diverging: A , u , p 
 Application of area-velocity relation
1.Rocket engines
Expand the exhaust gas to high velocity
supersonic speed.

2.Ideal supersonic wind tunnel

Supersonic Wind tunnel designed to expand stagnant gas to supersonic for testing,
compress supersonic stream to subsonic before exhausting to atmosphere
Diffuser is to slow down the flow in the convergent duct to sonic flow at the second throat,
and then further slowed to low subsonic speeds in the divergent duct; finally being
exhausted to the atmosphere for a blow-down wind tunnel
■ The above relation utilized the Conservation eqns
■ It holds true for real and chemical reacting gases, as well as
for perfect gas – as long as flow is isentropic
■ The Area velocity relation is in differential form and in order
to make quantitative use of it, integration is required
■ Primary importance of Area Velocity relationship is the
information it provides – as discussed earlier
■ Now we use fundamental governing eqns as well as Area
velocity relation to solve flow thru Nozzle

1u1 A1  2u2 A2
A2 
p1 A1   u A1  ( A
2
1 1 pdA) x  p2 A2   2u22 A2
1

u12 u22
h1  h2  h0 const
2 2
 5.4 Nozzles
■ Analysis of flow thru variable area ducts in general require
numerical solutions
■ However we can obtain closed-form results for case of
calorifically perfect gas
■ Discussion in 2 parts
– Purely isentropic subsonic-supersonic flow of a perfect gas
Isentropicthrough
Flow of nozzles
a Calorically Perfect Gas
through
– Variable-Area Duct:ratios across nozzles
Effect of pressure
 A  * 0 a* is Stagnation density
Since u*=a*, therefore  (constant throughout an
A*
0  u isentropic flow)

0 r  1 2 r 1 1 (1)
(1  M )
 2
Using relations at sonic conditions
0 r  1 r1 1 r  1 r 1 1
(1  ) ( ) (2)
 *
2 2

¿
See Eqn 3.37
(3)

Squaring the top eqn

r 1 Area – Mach
A 2 1 2 r  1 2 r 1
( *)  2 [ (1  M )] Number
A M r 1 2 Relation
 r 1
A 2 1 2 r  1 2 r 1 M at any location is function of
( *)  2 [ (1  M )]
A M r 1 2 local duct area to sonic throat are

M  f (A * )
A
dA du
( M 2  1)
A u

A must be greater or equal to A*;

case where A<A* is physically not
possible in an isentropic flow

There are two values of M which

correspond to a given A/A* >1 , a
subsonic & a supersonic value

Value of A/A* and M is found from

table A.1 for supersonic and
subsonic flow
• Consider CD nozzle where Ainlet/A* is
tending to infinity and M is approx. 0
(& not zero else no mass flow)
• Assume CD nozzle expands flow
isentropically to supersonic speed at
exit
• For given nozzle, A 2there
1 is
2 only r  1 one r 1
2 r 1
( *)  2 [ (1  M )]
isentropic solution
A given
M rby1 2

• In convergent portion, subsonic flow

is accelerated with local M dictated
by A/A*
At throat At=A* and M=1
• In divergent portion, flow expands
supersonically and local M depends
on A/A*
• If variation of M is known in nozzle,
then variation of static pressure,
static temp etc. can be found (eqn
3.28, 3.30, 3.31)
P, T, density decrease in nozzle
Pe/Po, Te/To, rhoe/rhoo depend on
Effect of Different Pressure Ratio Across a Given Nozzle

■ If CD nozzle is placed
stationary, nothing happens

■ To accelerate gas thru CD

nozzle, pressure difference
mandatory – exit pressure
must be lower than Inlet
pressure pe/po<1

■ For isentropic Ss flow (shock

free), pe must conform to
 What happens if pe/po does not conform to
above eqn or Backpressure downstream
of the nozzle exit is independently
P =P at the beginning ∴there is no flow exists
•governed?
e 0
in the nozzle
• Minutely reduce Pe ， this small pressure
difference will cause a small wind to blow
through the duct at low subsonic speeds
(curve 1). Remember A* is sonic throat area
and here At > A*. In diverging part, M
decreases.
• Further reduce Pe ， this large pressure
difference will cause more wind to blow thru
the duct at low subsonic speeds (curve 2).
• Further reduce Pe, sonic conditions are reached
at throat (curve 3). Now At = A*.
• All cases sketched are subsonic flows and there
are infinite isentropic solutions where both P e
/P0 & A/At are the controlling factors for the
local flow properties at any given section
• In Ss flow, where only 1 isentropic condition
Mass flow increases as pe decreases.

.
Calculating mass flow at throat m =tAtut, now when pe is reduced to

.
pe3 then m =*A*a*

If pe is reduced further, mach number at throat cannot increase beyond

M=1 as

dA . du
( M 2  1)
A u
Hence flow properties thru entire subsonic section of duct become
frozen when pe<pe3 – mass flow becomes constant ----this condition

after sonic flow is obtained at throat is Choked Flow

Due to divergent duct downstream of throat, p e/po required to attain
sonic flow at throat is larger than 0.528 (though isentropic relation
dictate that for sonic flow at throat , this ratio is 0.528

r
p* r  1 r 1
(1  ) 0.528
p0 2
What happens when Pe is further reduced below Pe3 ?

In convergent section, nothing happens – flow

properties remain as given by subsonic portion
of curve 3

No isentropic soln occurs in divergent until pe is

adequately reduced to specified low value as per

For values of exit pressure above this but below

pe3, a Normal shock exists inside divergent duct
Let exit pressure be pe4. Normal shock forms in
divergent section resulting in decrease in M at
exit and pressure increases to pe4 at exit
As exit pressure is reduced, shock moves
closer to exit.
Shock exactly at exit when pe=pe5 where
pe5 is static pressure behind shock at
design exit M.
When backpressure pB is reduced further
(pe6<pe<pe5); flow is Ss & isentropic in
nozzle.
Increase to pB takes place across Oblique
shock attached to nozzle exit – outside
duct. Nozzle is OVEREXPANDED
because exit pressure has expanded
below back pressure or pe6<pB.

If pB is further reduced below pe6,

equilibration of flow takes across
expansion wave outside duct. Nozzle is
UNDEREXPANDED since exit pressure
is higher than back pressure pe6>pB and
flow is capable of additional expansion
after leaving nozzle.
Note: quasi-1D consideration does not tell us much about how to design the
contour of a nozzle – essentially for ensuring a shock free supersonic nozzle -
Method of characteristics
 Flow Through C – D Nozzle
 We revise the
effects of Back
Pressure on the flow
through a
converging-diverging
nozzle.

 Flow
configurations for
various back
pressures and
corresponding
pressure and M No
distributions are
 Flow Through C – D Nozzle (Subsonic)
 Curve a.
 Back Pressure is
equal to the reservoir
pressure, pb= p0.
 No flow through
nozzle.
 Curve b.
 Back Pressure
slightly reduced, pb< p0
and flow is initiated in
the nozzle, but the
condition at throat is
still subsonic.
 Flow Through C – D Nozzle (Sonic)
 Curve c.
 Back Pressure is low enough
to make the flow reach sonic
conditions at the throat, pb = pc.
 Flow in diverging section is
still subsonic as back pressure
is still high.
 Nozzle is choked now. This
pressure is the highest pressure
to choke the C-D Nozzle
 The back-pressure at which
throat pressure first drops to
critical value is called Pb,crit
Flow Through C – D Nozzle (Shock wave
 Curve d.
 When Back Pressure is Pd,
flow follows the supersonic
path, but Back Pressure is
higher than Pi.
 This region of supersonic flow is
terminated by what is effectively a
normal shock wave.
 The flow downstream of the
shock wave then decelerates
subsonically until the
pressure on the exit plane Pe
is equal to the back-pressure,
P b.
 Flow Through C – D Nozzle (Shock
 Curve d. wave)
Decrease in Back
Pressure strengthens
shock and moves it
downstream.
 The flow in the nozzle under these
circumstances is as shown
Flow Through C – D Nozzle (Shock wave)
 Point f.
 One can think of
a Back Pressure Pf,
when the shock
formed is found at
the exit plane.
 Pf/P0 is the
smallest pressure
ratio required for
subsonic flow at
nozzle exit.
 Flow Through C – D Nozzle (Over-
expanded)
 Point g, h.
 As Pb is decreased below Pf,
the conditions at all sections of
the nozzle remain unchanged
and the pressure on the exit
plane, Pe, remains unchanged.
This results in the shocks
being formed outside of the
nozzle.(Pb>Pe)
 Flow Through C – D Nozzle (Over-

expanded)
Point g, h.
These are not Normal
Shocks. These are Oblique
Shocks.
Implication is that the flow
has reduced the pressure to low
values.
Additional shocks are required
to compress the flow further.
 Such a nozzle is termed as
Over-expanded
Flow Through C – D Nozzle (Design)
 Curve i.
 As Pb is further decreased the
oblique shock waves in the discharge
flow become weaker and weaker and
the difference between Pe and Pb
becomes smaller and smaller.
We can now think of a
back Pressure, pb = pi ,
which is small enough to
render the flow in the
diverging section
supersonic.
 For this Back Pressure,
the flow is everywhere
Flow Through C – D Nozzle (Design)
 Curve i.
It corresponds to design
conditions and flow is
perfectly expanded in the
nozzle.
 There are no waves inside or
outside the nozzle.
 The design back-pressure, i.e., the
back-pressure is given by
 Flow Through C – D Nozzle (Under-expanded)
 Point j.
 The other interesting situation
is If Pb is further reduced, it
becomes less than the exit plane
pressure, Pe, or Back Pressure is
less than pi.
 Even now the flow adjustment
(expansion from Pe to Pb) takes place
outside of the nozzle, not
through shocks, but through
Series of Expansion Waves.
 Flow Through C – D Nozzle (Under-
 Point j. expanded)
 Here implication is that
the flow could not
expand to reach the
back Pressure.
It required further
expansion to finish the
job.
 Such a nozzle is
termed Under-expanded.
Flow Through C – D Nozzle Summary
Location of Shock or
A/A ?
■ The first approach is an iterative solution.
t
■ Assume the location of the shock in the nozzle, i.e.,
assume the value of A/At for the shock.
■ Then calculate the pressure ratio pe/po that would
correspond to the shock in this assumed location.
■ Check to see if pe/po from this calculation agrees with the
specified value of pe/po
■ If not, assume another location of the shock, and calculate
the new value of pe/po corresponding to this new shock
location.
■ Repeat this iterative process until the proper shock
location is found that will yield a calculated pe/po that
agrees with the specified value.
 5.5 Diffusers
 Assume that we want to design (a) Exhaust the nozzle directly to the
atmosphere
a supersonic wind tunnel with a In order to avoid shock or expansion waves
test section M=3: in the test region downstream of the exit, the
exit pressure pe, must be equal to the
from A-1 surrounding atmospheric pressure, i.e., pe, =
1 atm. Since po/pe = 36.7, the driving
Ae/A*=4.23
reservoir pressure for this case must be 36.7
P0/Pe=36.7 atm.

■ Assume the wind tunnel exhausts

to the atmosphere.
■ What value of total pressure po,
must be provided by the reservoir
to drive the tunnel? There are
several possible alternatives.
 Normal Shock Diffuser

(b) Exhaust the nozzle into a constant area duct which

serves as the test section.
In this case, because the testing area is inside the duct, shock waves
from the duct exit will not affect the test section. Therefore, assume a
normal shock stands at the duct exit. The static pressure behind the
normal shock is p2, and because the flow is subsonic behind the shock,
p2 = p ∞ = 1 atm.

Where p2/pe is obtained at Mach 3 from A-

2

∴The reservoir pressure required to drive the wind tunnel

has markedly dropped from 36.7 to 3.55 atm
 Normal Shock Diffuser

(c) Add a divergent duct behind the normal the normal shock to even
slow down the already subsonic flow to a lower velocity. At the duct exit,
the Mach number is a very low subsonic value, and for all practical
purposes the local total and static pressure are the same. Moreover,
assuming an isentropic flow in the divergent duct behind the shock, the
total pressure at the duct exit is equal to the total pressure behind the
normal shock.
po2 = p ∞ = 1 atm

P0 Pe P2 1  1 
P0  P (36.7)( ) 1 3.04atm
Pe P2 P02 10.33  1.17 
Note: from Table A-2
M 3 P02
0.328
P01
P01 1
  3.04 Same as previous
P02 0.328
Thus, from this experiment, we infer that the reservoir pressure required
to drive a supersonic wind tunnel (and hence the power required from
the compressors) is considerably reduced by the creation of a normal
shock and subsequent isentropic diffusion to M ～ 0 at the tunnel exit,
and that this pressure is simply determined by the total pressure
loss across a normal shock wave at the test section Mach number.
The normal shock and divergent exhaust duct are acting as a specific
mechanism to slow the air to low subsonic speeds before exhausting to the
atmosphere.
Such mechanisms are called diffusers, and their function is to slow the flow
with as small a loss of total pressure as possible. Of course, the ideal diffuser
would compress the flow isentropically, hence with no loss of total pressure.
 Perfect Diffuser?
■ Of course, the ideal diffuser would compress the flow isentropically,
hence with no loss of total pressure.
■ After isentropically expanding through the supersonic nozzle and
passing through the test section, conceptually the supersonic flow
could be isentropically compressed by the convergent part of the
diffuser to sonic velocity at the second throat, and then further
isentropically compressed to low velocity in the divergent section
downstream of the throat.
■ This would take place with no loss in total pressure, and hence the
pressure ratio required to drive the tunnel would be unity-a
perpetual motion machine! Obviously, something is wrong.
■ When the convergent part of the diffuser changes the direction of
the supersonic flow at the wall, it is extremely difficult to prevent
oblique shock
■ Moreover, evenwaves from occurring inside the duct.
without
shocks, the real-life effects
of friction between the flow
and the diffuser surfaces
cause a loss of total
pressure.
■ Therefore the design of a
perfect isentropic diffuser is
 Oblique Shock Diffuser

 Accepting the fact that a perfect diffuser cannot be built. Can we still hope to
do better than the normal shock diffuser ?
 The answer is yes, because it can easily be shown that the total pressure loss
across a series of oblique shocks and a terminating weak normal shock is less
than that across a single strong normal shock at the same upstream Mach
number.
Wise to replace the normal shock diffuser with an oblique
shock diffuser.
Here, the test section flow at Mach number Me and static pressure pe is slowed
down through a series of oblique shock waves initiated by a compression corner at
the inlet of the diffuser.
Further slowed by a weak normal shock wave at the end of the constant-area
section, and then subsonically compressed by a divergent section which exhausts
to the atmosphere.
At the diffuser exit, the static pressure is pd, which for subsonic flow at the exit is

equal to p∞.
 Reality
However, in practice. The interaction of the shock waves with the viscous
boundary layer on the diffuser walls creates an additional total pressure loss
which tends to partially mitigate the advantages of an oblique shock diffuser.
 Diffuser Efficiency
For wind tunnel work, the most common definition of diffuser efficiency is to compare
the actual total pressure ratio across the diffuser with the total pressure ratio across a
hypothetical normal shock wave at the test section Mach number,
Let d denote diffuser efficiency. Then
P
( d0 ) actual Actual total pressure across diffuser
P0
 D Total pressure ratio across hypothetical normal shock at test section M
P02
( )
P01

If, d = 1, then the actual diffuser is performing as if it were a normal shock diffuser.

For low supersonic test section Mach numbers, diffusers in practice usually perform
slightly better than normal shock (d > 1);
 •In wind tunnel nomenclature, the nozzle throat is called the first throat, with
cross-sectional area At1 = A*; the diffuser throat is called the second throat, with
area At2. Due to the entropy increase in the diffuser, At2 > At1 .
•To prove this, assume that sonic flow exists at both the first and second throats.

a* and hence T*
 A a  A a
*
1
*
t1 1
*
2
*
t2 2
are constant
throughout a given
adiabatic flow.
∗
𝑃1 Thus, a1*/a2* = 1
∗ ∗ ∗ ∗ ∗
𝐴𝑡 2 𝜌 1 𝑎1 𝜌1 𝑅 𝑇1 𝑃1
= ∗ ( ∗ )= ∗ = ∗
= ∗
𝐴𝑡 1 𝜌 2 𝑎2 𝜌2 𝑃2 𝑃2
𝑅 𝑇∗
2
Since total press
At 2 P01 decreases across a shock
  P02  P01 always  At 2  At1
At1 P02
If we know the total pressure ratio, then we can
determine the size of 2nd throat
 At 2 P01

At1 P02
• Since the total pressure always decreases across shock waves and within
boundary layers, po2 will always be less than po1 .
• Thus, from Eq., the second throat must always be larger than the first
throat.
• Indeed, if we know the values of total pressure at the two throats, then
Eq. tells us precisely how large to make the second throat.
• If At2 is made smaller than demanded by Eq. the mass flow through the
tunnel cannot be handled by the diffuser; the diffuser "chokes," and
supersonic flow in the nozzle and test section is not possible.
• For typical supersonic diffusers, the
efficiency ηD is very sensitive to At2.
• Note that as At2 is decreased from a
large value ηD first increases,
reaches a peak value, then rapidly
decreases.
• The peak efficiency is obtained by
a value of At2 slightly larger than
given by Eq. (5.36).
• Keep in mind that the value of At2
obtained from Eq. (5.36) is the
minimum allowed value that will
pass the incoming mass flow from
the nozzle.
• Below this value, the flow will be
choked, and the diffuser efficiency
falls.
•• At
Themuch
valuehigher
of At2 values of (5.36)
from Eq. At2 there
is are no problems with passing the incoming
mass flow; however,
represented the diffuser
by the dashed efficiency is compromised because the
vertical
supersonic flow from the inlet is not sufficiently compressed and hence
line in Fig. 5.25.
remains supersonic in the second throat.
• In the downstream divergent portion, this supersonic flow first accelerates, and
then passes through a normal shock near the diffuser exit.
• Since the Mach number is fairly high in front of the shock, the total pressure
loss across the normal shock is large.
• This defeats the purpose of an oblique shock diffuser (namely. to have a weak
normal shock occur at the second throat in a near sonic flow).
 9.7. CRITICAL MACH NUMBER
■ Consider airfoil at different incoming M. At M=0.61, Mach number at A is 1.0. This
free stream M is called Critical Mach Number

Assuming
Isentropic flow,
pressure ratios
can be found

Pressure coefficient
at A

This equation tells for a given M∞ , value of Cp

at A and Mach number at A are uniquely related
at given point A
 ; Setting Ma=1, M∞=MCr and
Cp=CpCr

• Cp is a unique function of MCr and plot is shown

• Prandtl Glauret Rule

• Obtain incompressible pressure coefficient at

minimum pressure point
• Plot curve B using Compressibility correctios
• Plot CpCr vs MCr (Curve C)
• Intersection of B and C gives MCr
Thick vs Thin airfoil
• When freestream exceeds MCr, supersonic flow
on top surface of airfoil
• At high enough subsonic M, this supersonic
region will be terminated by weak shock
• Total pressure loss associated with shock
causes adverse pressure gradient- separate
boundary layer and large pressure drag
• Free stream M at which drag rise begins is
DRAG DIVERGENCE MACH NUMBER and it is
slightly large than MCr
• Higher drag divergence M
then aircraft can fly close
to M=1 before drag rise is
encountered.
• One way is to make wing
thinner (thickness to
chord ratio) --- see figure.
• By making wing thinner,
magnitude of drag
increase is reduced and
Mdrag_divergence comes closer
to 1.
• 2nd way is to give sweep
back the wing
• Consider Straight rectangular wing
with thickness to chord ratio of 0.15.
Air flows straight over this wing with
t1/c1=0.15.
• Now consider swept back wing with 45
degree sweep. Streamline CD flows
over the wing and sees an effective
airfoil shape with same thickness
(t1=t2).

• However effective chord length c2 is

longer by factor 1.141 (c2=1.41c1) –

considering same span (c2=c1/sin45).

• This makes effective thickness to
chord ratio equal to t2/c2=0.106 –
thinner by almost 1/3 compared to
straight wing.
• Therefore by sweeping wing, flow
behaves as if wing is thinner and M drag
 END OF COURSE

Thankful for your patient

ears, glowing eyes and
attentive minds

Good Luck