0% found this document useful (0 votes)
108 views101 pages

ME530 ch3

The document summarizes ideal fluid flow and potential flow theory. It describes ideal fluids as inviscid, incompressible fluids governed by the continuity and Euler equations. Potential flow is introduced as a special case of ideal flow where velocity derives from a potential function. Two-dimensional potential flows are then analyzed using streamfunctions, which satisfy the continuity equation by construction. Complex analysis tools like complex potentials and velocities are introduced to simplify analysis of things like source/sink/vortex flows.

Uploaded by

Victor Sanchez
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
108 views101 pages

ME530 ch3

The document summarizes ideal fluid flow and potential flow theory. It describes ideal fluids as inviscid, incompressible fluids governed by the continuity and Euler equations. Potential flow is introduced as a special case of ideal flow where velocity derives from a potential function. Two-dimensional potential flows are then analyzed using streamfunctions, which satisfy the continuity equation by construction. Complex analysis tools like complex potentials and velocities are introduced to simplify analysis of things like source/sink/vortex flows.

Uploaded by

Victor Sanchez
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 101

II.

Ideal – fluid flow


● Ideal fluids are
● Inviscid
● Incompressible
● The only ones decently understood mathematically
● Governing equations
∇⋅u=0
Continuity

∂u 1
+( u⋅∇ ) u=− ρ ∇ p+ f
∂t Euler
Boundary conditions Normal to surface
Free-slip
u⋅n=U⋅n (velocity is parallel
to surface)
Velocity of surface

Potential flow (special case)


u =  (u = /x, v = /y, w = /z)
Potential flow is irrotational
Continuity equation for potential flow
  = 0
2

Continuity equation (with boundary conditions)


can be solved alone for velocity
Then plug  into momentum equation (Bernoulli
form) to solve for pressure
4. 2D potential flows
4.1. Stream function
● 2D ideal continuity equation
∂u ∂ v
+ =0
∂x ∂ y

Velocity potential 
∂φ ∂φ
u= , v=
∂x ∂y

Introduce streamfunction  (counterpart of
potential) so that
∂ψ ∂ψ
u= , v=−
∂y ∂x
Streamfunction satisfies continuity equation by
construction
2 2
∂ ψ ∂ ψ
− =0
∂x∂ y ∂ y∂ x
Streamfunction exists for any ideal 2D flow
Before going further, consider vorticity in 2D flow

[ ]
i j k
∂ ∂ ∂
ω=∇ ×u=det
∂x ∂y ∂z
u v w
Streamfunction satisfies continuity equation by
construction
2 2
∂ ψ ∂ ψ
− =0
∂x∂ y ∂ y∂ x
Streamfunction exists for any ideal 2D flow
Before going further, consider vorticity in 2D flow

[ ]
i j k
∂ ∂ 0
ω=∇ ×u=det
∂x ∂y
u v 0
Vorticity in 2D flow For 2D,
effectively
a scalar

ω=k
∂v ∂u

∂x ∂y(=k ω
)
Now consider an irrotational 2D flow
∂v ∂u
ω= − =0
∂x ∂y
Express velocity in terms of streamfunction

ω=

∂x

( ) ( )
∂ψ
∂x

∂ ∂ψ
∂y ∂y
=0

2
∇ ψ=0
Properties of streamfunction


Streamlines are lines of  = const

Difference in the value of  between two
streamlines equals the volume of fluid flowing
between them

Streamlines  = const and potential lines
 = const are orthogonal at every point in the
flow
Why  = const is a streamline

ds
dy dx

d ψ=

ds
ds=
(
∂ψ ∂ x ∂ψ ∂ y
+
∂ x ∂s ∂ y ∂s )
ds=−v dx+u dy

dy dx
d ψ=0 means v dx=u dy ; =
v u
Streamline
equation!
Flow rate between two streamlines
 =1 Direction along AB:
B ds = (dx,dy)
ds
v u
dn u
Direction normal to AB:
 =2 dn = (dy,-dx)
A
Volume flow rate
B B B B
Q=∫A u⋅n ds=∫A u⋅d n=∫ A u dy−∫A v dx
B
Q=∫A d ψ=ψ1−ψ2
Orthogonality between streamlines and potential
lines
Along a streamline d ψ=−v dx+u dy=0

Along an isopotential line ( = const)...


∂φ ∂φ
d φ= dx+ dy=u dx+v dy=0
∂x ∂y

Normal to streamline: (-v, u)


Normal to isopotential line: (u, v)
They are orthogonal: (-v, u)(u, v)  0
4.2. Complex potential and velocity

● Complex variable z = x+iy


● Function of a complex variable
F(z) =  (x,y) + i  (x,y)
● Cauchy-Riemann condition for function of a
complex variable to be holomorphic*
∂φ ∂ ψ ∂φ ∂ψ
= ; =− ;
∂x ∂y ∂y ∂x
Holomorphic function – complex-valued function of a complex
variable which is differentiable in a neighborhood of every point
within its domain
Complex potential constructed from velocity
potential and streamfunction
F(z) =  (x,y) + i  (x,y)
Cauchy-Riemann condition satisfied by
construction
Advantages of using complex potential

If  and  are the real and imaginary parts of
any holomorphic function, 2 = 0 and 2 = 0
automatically
● Complex velocity w = dF/dz = u - iv – directly
related to flow velocity
Magnitude of complex velocity
w*w = (u + iv)(u – iv) = u2+v2 = uu = 

Polar coordinates in complex plane


y
x + iy = r (cos  + i sin ) = rei
u = ur cos  - u sin 
eq
er

v = ur sin  + u cos 


r
x
q
w = (ur - iu) e-i
4.3. Uniform flow
−i α
F (z )=C e z
dF −i α
w (z )= =C e =C cos α−iC sin α
dz
u=C cos α , v=C sin α
y

x
4.4. Source, sink, and vortex

F (z )=C log z=C log (r e )=C (log r+i θ)
First, let C be real and positive
φ=C log r , ψ=C θ
dF C C −i θ
y w (z )= = = e
dz z r
C
u r= , uθ =0
x r

Source at z = 0
Source strength (discharge rate)
2π 2π
m=∫0 u r r d θ=∫0 C d θ=2 πC
Complex potential of a source of strength m at
z = z0
m
F ( z )= log(z −z 0)

Complex potential of a sink of strength m at z = z0
m
y F ( z )=− log ( z−z 0 )

x
Now consider a purely imaginary constant in the
logarithmic potential:

F (z )=−iC log z =−iC log (r e )=−iC log r+C θ
φ=C θ , ψ=−C log r
dF C C −i θ
w (z )= =−i =−i e
y
dz z r
C
u r=0, u θ=
r
x Point vortex
Vortex strength (circulation)

Γ=∮ u⋅d l =∫0 u θ r d θ=2 π C
L
Complex potential of a vortex with circulation G at
z = z0
F ( z )=−i Γ log (z−z 0 )

Note 1. z = z0 is a singularity (uq  )
Note 2. This flow field is called a free vortex:

Γ L' =∮ u⋅d l ≡0
L' Any contour not
including z0
4.5. Flow in a sector
n
F (z )=U z , n⩾1
Abraham de Moivre's formula

Abraham de Moivre
1667-1754
Author of The Doctrine of Chances
4.5. Flow in a sector
n
F (z )=U z , n⩾1
Abraham de Moivre's formula
inθ n
e =( cos θ+i sin θ ) =cos (n θ)+i sin(n θ)

Use polar coordinates



z=r e
n n
F (z )=U r cos(nθ)+i U r sin (n θ)
Potential and stream function
n n
φ=U r cos (n θ) , ψ=U r sin( n θ)
Complex velocity
n−1 n−1 i(n−1)θ
w (z )=nUz =n U r e =
=( n U r cos n θ+i nU r sin n θ ) e
n−1 n−1 −i θ

Velocity components
n−1
u r=n U r cos n θ
n−1
u θ=−n U r sin n θ
n = 1: uniform flow
n = 2: flow in a
right-angle corner
q=p/n
n = 3: shown
4.6. Flow around a sharp edge
1 /2 1 /2 i θ /2
F (z )=C z =C r e
Potential and streamfunction
φ=C r 1 /2 θ 1/ 2
cos , ψ=C r sin θ
2 2
Complex velocity
dF 1 −1 / 2 C −iθ / 2
w (z )= = C z = 1 /2 e =
dz 2 2r
C −iθ i θ/ 2 C θ θ
2r 2r (
= 1 /2 e e = 1 /2 cos +i sin e
2 2)−i θ

C θ C θ
u r= 1/ 2 cos , u θ=− 1 /2 sin
2r 2 2r 2
y

Singularity

 = 0, q = 0 x
 = 0, q = 2p

C θ C θ
u r= 1/ 2 cos , u θ=− 1 /2 sin
2r 2 2r 2
4.7. Doublet
y

Source at x = -e Sink at x = +e

Now let e  0


Complex potential of source and sink
m m
F (z )= log(z +ε)− log ( z−ε)
2π 2π
m z+ε m 1+ε/ z
F (z )= log = log
2π z−ε 2 π 1−ε/ z
For small /z, expand denominator into series:
−1
(1−ε/ z) =1+ε/ z+…
Plug that into F(z)
m
F (z )= log ( ( 1+ε /z )( 1+ε/ z+…) )

m ε
F (z )=
2π (
log 1+2 +…
z )
Use series expansion for logarithm near 1
m ε m ε
F (z )=
2π (
log 1+2 +… =
z )
2 +…
2π z ( )
If we take the limit of this as   0, the result will
be trivial: F(z) = 0
For a non-trivial result, let lim mε=πμ
ε →0
Then
μ μ x−iy x−iy
lim F (z)= = =μ =μ 2 2
ε →0 z x+iy (x+iy)(x−iy) x +y
x y
φ=μ 2 2 , ψ=−μ 2 2
x +y x +y
Consider a streamline  = const
y
ψ=−μ 2 2
x +y
2 2
ψ( x + y )=−μ y
2 2 μ
x + y + ψ y=0
2 2
μ μ μ
2 2
x + y + ψ y+
( ) ( )

=

2 2
μ μ
2
( 2ψ) ( 2ψ)
x + y+ =

Circle of radius m/(2y) and center at x = 0, y = -m/(2y)


y

x
μ μ −2i θ μ −iθ
w (z )=− 2 =− 2 e =− 2 e (cos θ−i sin θ)
z r r
μ
y u r=− 2 cos θ
r
μ
u θ=− 2 sin θ
r
x
Doublet of strength m at z = z0
μ
F (z )=
z−z 0
4.8. Circular cylinder flow
Let uniform flow go past a doublet
μ
F ( z )=Uz+
z
Potential and stream function
μ μ μ

( )
F (z )=Ure + i θ = Ur+ cos θ+i Ur− sin θ
re r r ( )
Potential Stream function

Consider streamline y = 0


Ur = m/r means that this streamline is a circle of
1/2
radius a = (m/U)
Can rewrite complex potential as

( )
2
a
F ( z )=U z+
z
z → ∞ , F (z)→ U z
Uniform flow dominates the far field
2
a
z → 0, F ( z )→U
z
Doublet dominates the flow near the origin
Singularity at origin
Velocity=0
Velocity=0
(rear
(forward
stagnation
stagnation
point)
point)

Flow symmetry: F(-z) = -F(z)


4.9. Cylinder with circulation
Take cylinder flow, add rotation around the origin

( )
2
a iΓ
F (z )=U z+ + log z+C
z 2π Constant to
Vortex at origin keep y = 0 at r = a

Pretty easy to find C, tuck it into the logarithm

( )
2
a iΓ z
F (z )=U z+ + log
z 2π a
Complex velocity

( )
2
dF a iΓ 1
w= =U 1− 2 +
dz z 2π z
( ) ( )
2 2
a iΓ 1 a −2 i θ i Γ 1 −iθ
w=U 1− 2 + =U 1− 2 e + e
z 2π z r 2π r

[( ) ]
2
a −i θ i Γ 1 −i θ

w= U e − 2 e + e
r 2π r

[( ) (( ) )]
2 2
a a Γ −i θ
w= U 1− 2 cos θ+i U 1+ 2 sin θ+ e
r r 2πr

Remember that w = (ur-iuq)e-iq

( ) ( )
2 2
a a Γ
u r=U 1− 2 cos θ , uθ =−U 1+ 2 sin θ−
r r 2π r
On the surface (r = a),
Γ
u r=0, u θ=−2 U sin θ−
Boundary! 2πa
Find stagnation points (velocity = 0, r = a)

sin θs =− Γ
4 πU a
Possibilities:
2 stagnation points on the cylinder
1 stagnation point on the cylinder
0 stagnation points on the cylinder (but maybe
somewhere else in the flow?)
Two stagnation points

0< Γ <1
4 πUa

One stagnation point

Γ =1
4 π Ua
No stagnation points on the cylinder
Γ >1
4 π Ua
Look for stagnation point (rs, qs) elsewhere (for
rs > a) Cannot
be 0 Must be 0!

( )
2
a
u r=U 1− 2 cos θs =0,
rs

( )
2
a Γ
u θ=−U 1+ 2 sin θ s− =0
rs 2πrs
cos qs = 0 means qs = p/2 or qs = 3p/2

( )
2
a Γ
U 1+ 2 sin θ s=−
rs 2 π rs
positive negative

Must be
-1, so qs = 3p/2

( )
2
a Γ
U 1+ 2 =
rs 2π rs
Solve this for rs
√(
2
r s= Γ ± Γ
4πU 4πU )
−a2

Two stagnation points


- inside the cylinder (so who cares?)
+ outside the cylinder (good stuff)
4.10. Blasius integral laws

● Find potential
● Find velocity components
● Plug velocity into Bernoulli equation to find
pressure on body surface
● Integrate to find
● Hydrodynamic force on the body
● Hydrodynamic moment on the body
● MUCH simpler with complex potential!
Any contour fully enclosing the body

C0 Body of an arbitrary shape


Surface: streamline y = 0

Complex force:
Y FFororcece X - iY
y c.g.
M X

Blasius first law Blasius second law

ρ ρ
( ∮zw )
2
X −iY =i ∮ w dz
2
M= ℜ dz
2C 0
2 C0
Evaluating complex integrals
Taylor series (real variable)
(n)

f ( x 0)
f ( x−x 0)=∑ a n ( x−x 0 ) , a n=
n

n=0 n!
This expansion is valid in an interval |x - x0| < dx
Evaluating complex integrals
Laurent series (complex variable)

f ( z−z 0 )= ∑ a n ( z−z 0 ) ,
n

n=−∞
1
a n= ∮
2π i C
f (ζ )( ζ−z 0)−n−1

This expansion is valid in an annulus where f is


holomorphic: R1 < |z - z0| < R2
C
If R1 = 0, z0 – isolated singularity
 z0
Coefficient a-1 of Laurent series:
residue of f at z0
Cauchy theorem
If complex function f(z) is holomorphic
everywhere inside contour C,

∮ f (z)dz=0
C

Cauchy residue theorem


If complex function f(z) is holomorphic
everywhere inside contour C, except isolated
singularities,
∮ f (z)dz=2iπ ∑ a−1, k
C k
Example
ez – holomorphic everywhere in a disk of radius r
with center at z = 0
2 3
z z z
e =1+z+ + +…
2 6
ez/z – holomorphic everywhere in a disk of radius r
with center at z = 0, except at it center
z 2
e 1 z z
= +1+ + +…
z z 2 6
a-1=1

Note. a-m  0, a-m-k  0, k = 1,2, ... at z = z0 –


z0 is a pole of order m
4.11. Force and moment on a circular
cylinder
Complex potential

( )
2
a iΓ z
F (z )=U z+ + log
z 2π a
Complex velocity

( )
2
dF a iΓ
w= =U 1− 2 +
dz z 2πz
Blasius first law
ρ
X −iY =i ∮ w dz
2
2C 0
2 2 2 4 2
2 2U a U a i U Γ i U Γ a
2 Γ 2
w =U − + 4 + − − 2 2
z 2
z πz πz 3
4π z
0 -2 -4 -1 -3 -2
Term order in z
iU Γ
a−1= π

z = 0 – sole isolated singularity of w2, thus


iU Γ
X −iY =2iπ ∑ a−1,k =2iπ π =−iρ U Γ
k
X = 0 (D'Alembert's paradox)
Y = rUG (Zhukovsky-Kutta law)

Similar analysis for zw2 produces M = 0


4.12. Conformal transformations
Helps deal with boundaries
z = x+iy y z = x+ih h

x x
z = f(z)

It's only good if the Laplace equation is also


transformed into something nice...
● Consider f – holomorphic function mapping (x,y)
into (x,h)

In (x,y) plane, let 2(x,y) = 0

Then in (x,h) plane, 2(x,h) = 0
(proof: p. 93)
● Laplace equation is preserved by conformal
mapping
● What happens with complex velocity?
dF dF ( ζ) d ζ d ζ
w (z )= = = w( ζ)
dz d ζ dz dz
Velocity scales during conformal mapping
Let's prove that conformal mapping preserves
sources, sinks, etc.
C Γ=∮ u⋅dl =∮ ( u dx+v dy )
C C

Circulation of all
point vortices inside
C
dy dl
dx
m=∮ u⋅dn=∮ ( u dy−v dx )
C C
Strength of all
sources/sinks inside
C
∮ w( z ) dz=
C

=∮ ( u−iv )( dx+idy )=∮ ( u dx+v dy )+i ∮ ( udy−vdx )=


C C C
=Γ+i m
Could have proven the same with residue
theorem...
Now consider a conformal mapping (x,y)  (x,h)
( Γ+i m )∣z =∮ w(z )dz=

C∣z
=∮ w(ζ) dz =
C∣
dz
=∮ w(ζ) d ζ=( Γ+i m )∣ζ
z

C ∣ζ
Conformal mapping preserves strength of
sources, sinks, and vortices
4.13. Zhukovsky transformation
2
c
z=ζ+
ζ
∣ζ∣→ ∞ , z → ζ
2
dz c
=1− 2
dζ ζ
z = 0: singularity (let's contain it Nikolai Egorovich
Zhukovsky
inside the body) (1847-1921)
“Man will fly using the
dz power of his intellect
ζ=±c , =0 rather than the strength
dζ of his arms.”

z = c: critical points (angle not preserved)


Critical points of Zhukovsky transform
z = x+iy z = x+ih h
y z0
z0

q2 q1 x n2 n1 x
-2c +2c -c +c

2
c
ζ=±c z=±c+ =±2c
±c
Can prove: q1 - q2 = 2 (n1 - n2)
A smooth curve passing through z = c will
correspond to a curve with a cusp in z-plane
Example: z = cein
z = x+iy z = x+ih h
y

x x
-2c +2c -c +c

2
c
z=ce + i ν =c ( e +e ) =2c cos ν
iν iν −i ν

ce
Zhukovsky transform recipe. Start with flow
around a cylinder in z-plane, map to something
4.14. Flow around ellipses
Circle in z-plane, radius a > c, center at origin

ζ=a e

( ) ( )
2 2 2
c −i ν
iν c c
z=a e + e = a+ cos ν+i a− sin ν
a a a
Major semiaxis Minor semiaxis
Parametric equation of an ellipse
z = x+iy z = x+ih h
y

x x
-2c +2c -c +c
Flow past a cylinder

( )
2
a
F ( z )=U z+
z
Now consider freestream flow at an angle
Can get this by conformal
mapping too (in plane z':
x'
z = eiaz' - rotation)
y'

Correspondingly,
z' = e-iaz
In plane z'

( )
2
a
F (z ')=U z ' +
z'

( ) ( )
2 2
−i α a −i α a iα
F =U z e + −i α =U z e + e
ze z
Let's have this flow in z-plane:

( )
2
−i α a iα
F (ζ)=U ζ e + e
ζ
Now recall that
2
c
z=ζ+
ζ
Express z in terms of z:
2 2
ζ +c −ζ z =0

(√ )
2
z z 2
ζ= ± −c
2 2
Recall that for z  , z  z. Thus select

(√ )
2
z z 2
ζ= + −c
2 2
Plug this into F(z) to get F(z)... (skip derivation)
[ ( )( √( ) )]
2 2
−i α a i α −i α z z 2
F ( z )=U ze + 2 e −e − −c
c 2 2

Uniform flow at angle a approaching an ellipse


with major semiaxis a + c2/a and minor semiaxis
a - c2/a
a

Stagnation points: z =aeia


Stagnation points in z-plane...
2
c −i α

z=±ae ± e
a

a
( ) ( )
2 2
c c
z=± a+ cos α±i a− sin α
a a

( )
2
c
x=± a+ cos α
a

( )
2
c
y=± a− sin α
a
- forward stagnation point
+ downstream stagnation point
a = 0: horizontal flow approaching horizontal
ellipse
a = p/2: vertical flow, horizontal ellipse (or
horizontal flow, vertical ellipse)
4.15. Kutta condition and the flat-plate airfoil

Martin Wilhelm Kutta


(1867-1944)
4.15. Zhukovsky-Chaplygin postulate and
the flat-plate airfoil

Sergey Chaplygin
(1869-1942),
Hero of Socialist Labour
(1 February 1941)
4.15. Zhukovsky-Chaplygin postulate and
the flat-plate airfoil
Flow around a sharp edge (section 4.6)...
1 /2
F (z )=C z
dF C
w (z )= = 1/2
dz 2 z
z = 0: singularity

● At a sharp edge, velocity goes to infinity


4.15. Zhukovsky-Chaplygin postulate and
the flat-plate airfoil
Flow around a sharp edge (section 4.6)...
1 /2
F (z )=C z
dF C
w (z )= = 1/2
dz 2 z
z = 0: singularity

● At a sharp edge, velocity goes to infinity


● This is not the case in experiment, luckily
● Need a fix for theory near sharp edges
● That's not the only problem though...
2
c
z=ζ+
ζ
z = x+iy z = x+ih

r = c

a a

Herein lies
the problem!
Stagnation point is ALWAYS
at the trailing sharp edge!

Smoke visualization of wind tunnel flow past a lifting surface


Alexander Lippisch, 1953
Zhukovsky-Chaplygin postulate:
For bodies with sharp trailing edges at modest
angles of attack to the freestream, the rear
stagnation point will stay at the trailing edge

Dealing with trailing-edge singularity


In modeling real lifting surfaces, trailing edge has
sharp but finite curvature
How to “fix” the flat-plate flow?
z = x+ih

Angle of attack
Add circulation...
z = x+iy z = x+ih

...to move the


stagnation point to the
trailing edge!
We want to move the rear stagnation point to
z = 2c
That would correspond to z = c in the z-plane
Need to move it there from z = ceia
For cylinder flow with circulation...
sin θs =− Γ
4 πU a
If sin qs = - sin a,
Γ=4 π U a sin α
Recipe for constructing a complex potential for
corrected flat-plate flow (Eq. 4.22b)
● Cylinder flow

Add circulation G = 4p a U sin a

Rotate the plane a degrees counterclockwise
● Zhukovsky transform
● ???
● Profit!
Lift on a flat-plate airfoil extending from -2a to 2a

Blasius law for cylinder flow:


Y =ρ U Γ
In our case
2
Y =4 π ρU a sin α
Introduce dimensionless lift coefficient
Y
C L=
1 2
wing ρU l
2
chord Characteristic length scale
(for wings – chord length)

For our flat plate, l = 4a and


C L =2π sin α

At small angles of attack, lift coefficient on a flat


plate increases with angle of attack!
4.16. Symmetrical Zhukovsky airfoil
Goal: airfoil with sharp trailing edge and blunt
leading edge
z = x+iy z = x+ih h
y

r = a = c(1 + e)

x x
l
t
-2c +2c -c +c

Center:
-m = -ec
-(c + 2m)
small
Leading edge in z-plane: -(c + 2m)
In z-plane, the leading edge is...
c 2
z=−c (1+2ε)− =−2c+O(ε )≈−2c
1+2 ε
Chord length l = 4c
Similarly (more series expansions, linearization)
thickness t 3 √3
t =3 √ 3c ε , = ε
l 4
Thickness ratio

Maximum thickness occurs at x = -c


Extra Flugzeugbau EA300, 1987, Walter
Extra design, Zhukovsky wing profile
Can find e in z-plane from desired l and t in z-
plane:
4 t t
ε= ≈0.77
3 √3 l l
Equation for symmetric Zhukovsky profile in z-
plane

)√ ( )
2
y
l

2
3√3 (
1−2
x
l
1− 2
x
l

At zero angle of attack, stagnation point is at


trailing edge, lift = 0
Add angle of attack a...
To satisfy the Zhukovsky/Kutta/whatever
condition...
z = x+ih

Need to move this


r = a stagnation point...
x

...here!
a

For a cylinder of radius a, the needed amount of


circulation is (same as for flat plate...)
G = 4p a U sin a
Express radius a in terms of l and t...
l
a=c+m=c (1+ε)= 1+
4
4 t
3 √3 l ( )
For an angle of attack a, circulation we need to
add is...

Γ=4 π U a sin α=π U l 1+


(
4 t
3√3 l
sin α
)
Lift coefficient for symmetrical Zhukovsky airfoil

( t
C L ≈2 π 1+0.77 sin α
l )
t  0, this reduces to lift coefficient of flat plate
Zhukovsky symmetrical profile has better lift!
4.17. Arc airfoil
Airfoil of zero thickness but finite curvature
Use cosine theorem to z = x+ih h

get r a
2 2 2 π
a =r +m −2 r m cos −ν
( )

r
x

m
n
2 -c +c
In z-plane,
2
c −i ν

z=r e + e =
r
x y

( ) ( )
2 2
c c
= r+ cos ν+i r− sin ν
r r
( ) ( )
4 4
2 c 2 2 22 2 2 c 2
x = r +2c + 2 cos ν , y = r −2c + 2 sin ν
r r
2
 sin n 2
 cos n

( )
4
2 2 2 c 2 22 2 2
r cos ν sin ν=x sin ν− 2 c + 2 cos ν sin ν
r
=

( )
4
2 2 2 c 2 2 2 2 2
r cos ν sin ν= y cos ν+ 2c − 2 cos ν sin ν
r
2 2 2 2 2 2 2
x sin ν− y cos ν=4c cos ν sin ν
Use cosine theorem:

( )
2 2 2
r −c c 1 y
sin ν= = r− =
2 rm r 2 m 2 msin ν
y cannot be 2 y 2 y
sin ν= , cos ν=1−
negative!!! 2m 2m
2 2 2 2 2 2 2
x sin ν− y cos ν=4c cos ν sin ν

x
2 y
2m
2
− y 1−
2
(
y
2m
2
)
=4 c
2
2 y
2m( 1−
y
2m )
x y 2c 2 y
− y+ = −c 2
2m 2m m m
2 2 2 2 y
x −2 m y+ y =4 c −2 c
m

[ ( )] [ ( ) ]
2 2
2 c m 2 c m
x + y +c − =c 4+ −
m c m c
[ ( )] [ ( ) ]
2 2
2 c m 2c m
x + y +c − =c 4+ −
m c m c
y⩾0
Equation of an arc in the z-plane
z = x+iy z = x+ih h
y

a
h – will find

r
x

m
x n
-2c +2c -c +c
Otto Lilienthal and his glider, 1895
[ ( )] [ ( ) ]
2 2
2 c m 2 c m
x + y +c − =c 4+ −
m c m c

Recall that m/c = e, linearize (not essential here


but nice)
2 2

( ) ( )
2
2 c 2 c
x + y+ =c 4+ 2
m m
Find arc height h
Since y = 2m sin2n, ymax = h =2m
Next have to add circulation to put stagnation
point at the trailing edge (trickier, because
cylinder is moved upward in the z-plane)
Stagnation point needs to rotate by a + tan-1(m/c)
Angle of attack Vertical shift

Linearize:
tan-1(m/c)  m/c = e, a  c
Amount of circulation to be added:
m
( )
Γ=4 π U a sin α+ ≈4 πU c sin α+
c
m
c ( )
Lift coefficient:

( )
m
C L =2π U c sin α+ =2 π U c sin α+2
c
h
l ( )
Again, more lift than flat plate!
4. 18. Zhukovsky airfoil

● Know how to create lifting surfaces with:


● Straight chord, finite thickness
● Zero thickness, small finite curvature (camber)
● Both improve lift, compared with flat plate
● Create a lifting surface with both thickness and
camber (Zhukovsky profile)
z = x+ih
h

r
h/2
x
-c +c
0.77 tc/l

l – chord
t – max. thickness
h – max. camber
z = x+iy
y

x
t
h
-2c +2c
l
Maxim Gorky (ANT-20, PS-124) plane, 1935
Circulation

( t
Γ=π U l 1+0.77 sin α+
l
thick
) (
2h
l
cam
)
ness ber
Lift coefficient

( t
C L =2π 1+0.77 sin α+
l ) (
2h
l )
Kalinin K-7 (Russia, 1930)
Dornier X flying boat (Germany, 1929)

You might also like