Prandtl/Blasius boundary layer solution (for laminar boundary layers):

( )
2 2
∂𝑢 ∂𝑢 1 ∂𝑝 ∂𝑢 ∂𝑢
​ 𝑢 ∂𝑥
+𝑣 ∂𝑦
=− ρ ∂𝑥
+ ν 2 + 2
∂𝑥 ∂𝑦

+ ν( )
2 2
∂𝑣 ∂𝑣 1 ∂𝑝 ∂𝑣 ∂𝑣
​ 𝑢 ∂𝑥
+𝑣 ∂𝑦
=− ρ ∂𝑦 2 + 2
∂𝑥 ∂𝑦
∂𝑢 ∂𝑣
​ ∂𝑥
+ ∂𝑦
=0

The fluid velocity (both u and v) far from the body is the upstream velocity and zero at the solid
body surface.

Approximations:

Since the boundary layer is thin, it is expected that the component of velocity normal to the plate
is much smaller than that parallel to the plate and that the rate of change of any parameter θ
across the boundary layer should be much greater than that along the flow direction.
∂ ∂
​ 𝑣 ≪ 𝑢, ∂𝑥
θ ≪ ∂𝑦
θ
∂𝑢 ∂𝑣
​ ∂𝑥
+ ∂𝑦
= 0
2
∂𝑢 ∂𝑢 1 ∂𝑝 ∂𝑢
​ 𝑢 ∂𝑥
+𝑣 ∂𝑦
=− ρ ∂𝑥
+ ν 2
∂𝑦

Similarly, the pressure gradient in the x-direction is much smaller than other pressure gradients
in the boundary layer region.
∂𝑢 ∂𝑣
​ ∂𝑥
+ ∂𝑦
=0
2
∂𝑢 ∂𝑢 ∂𝑢
​ 𝑢 ∂𝑥
+𝑣 ∂𝑦
= ν 2
∂𝑦

The nature of the above formulation and the original equations is different. The former is
parabolic, and the latter is elliptical. Therefore, the solutions are different in nature, and what
happens downstream of a given location in a boundary layer cannot affect what happens
upstream of that point. Consequently, the presence of the plate has no effect on the flow ahead
of the plate.

Substitution:
𝑥 𝑦 𝑢 𝑣 𝑝
​ 𝑥' = 𝑐
, 𝑦' = δ
, 𝑢' = 𝑈
, 𝑣' = 𝑉
, 𝑝' = 𝑝0
2 2
𝑈 ∂𝑢' 𝑈𝑉 ∂𝑢' 1 ∂ 𝑢'
​ 𝑐
𝑢' ∂𝑥'
+ δ
𝑣' ∂𝑦'
= ν𝑈 2 2
δ ∂𝑦'
2 2
​ 𝑢'
∂𝑢'
∂𝑥'
+
𝑉𝑐
𝑈δ
𝑣'
∂𝑢'
∂𝑦'
=
𝑣
𝑈𝑐 ( ) 𝑐
δ
∂ 𝑢'
∂𝑦'
2

​ c: the length of the flat plate; c ∝ x

One can argue that, in dimensionless form, the boundary layer velocity profiles on a flat plate
should be similar regardless of the location along the plate, thus being independent of x.
​
𝑢
𝑈
=𝑔 ( ) 𝑦
δ
​ 𝑔: 𝑎𝑛 𝑢𝑛𝑘𝑛𝑜𝑤𝑛 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 𝑡𝑜 𝑏𝑒 𝑑𝑒𝑡𝑒𝑟𝑚𝑖𝑛𝑒𝑑

Inside the boundary layer, viscous forces balance inertia and pressure gradient forces. Thus,
inertia and viscous forces are of the same order.
𝑐 2 ν𝑥 1/2
​
ν
𝑈𝑐 ( )
δ
∝1 ⇒ δ∝ ( ) 𝑈
𝑈
η=𝑦 ν𝑥
∂η 𝑦 𝑈 1
∂𝑥
=− 2𝑥 ν𝑥
=− 2𝑥
η
1/2
ψ = (ν𝑥𝑈) 𝑓(η)
∂ψ
𝑢 = ∂𝑦 = 𝑈𝑓'(η)
𝑣
∂ψ ν𝑈
=− ∂𝑥 = 4𝑥 {η𝑓'(η) − ( ) 𝑓(η)}
f: an unknown function

Substitution:
∂𝑢
∂𝑥
+
∂𝑣
∂𝑦
=−
𝑈η
2𝑥
𝑓'' + ( ) ν𝑈
4𝑥
𝑈
ν𝑥
(𝑓' + η𝑓'' − 𝑓') = 0

(−1+
ν
2
𝑈
𝑣𝑥 )
𝑓'' = ( νη
2𝑦 )
− 1 𝑓'' = 0 ⇒
νη
2𝑦
=1
2
𝑢
∂𝑢
∂𝑥
+𝑣
∂𝑢
∂𝑦
= ν
∂𝑢
∂𝑦
2 ⇒ 𝑈𝑓' − ( 𝑈η
2𝑥 ) ( )(η𝑓' − 𝑓)𝑈
𝑓'' +
ν𝑈
4𝑥
𝑈
ν𝑥 ( )𝑓'''
𝑓'' = ν𝑈
𝑈
ν𝑥
η
2 ( νη
2𝑦 )
− 1 𝑓'𝑓'' −
νη
4𝑦
𝑓𝑓'' = 𝑓'''
νη 1
​ 𝑓''' + 4𝑦
𝑓𝑓'' = 𝑓''' + 2
𝑓𝑓'' = 0
​ 2𝑓''' + 𝑓𝑓'' = 0

Boundary conditions:
​ η = 0: 𝑓 = 0, 𝑓' = 0 ∵ 𝑢 = 𝑣 = 0
​ η → ∞: 𝑓' → 1 ∵ 𝑢 = 𝑈

Although there is no known analytical solution to the above ODE, it is relatively easy to
numerically integrate the equation.
From the solution, it is found that u/U ≈ 0.99 when η = 5.0.
𝑈
​ η𝑦=δ = δ ν𝑥
=5
ν𝑥
​ δ=5 𝑈
δ ν 5
​ 𝑥
=5 𝑈𝑥
=
𝑅𝑒𝑥

Similarly,
*
δ 1.721
​ 𝑥
=
𝑅𝑒𝑥
Θ 0.664
​ 𝑥
=
𝑅𝑒𝑥

As postulated, the boundary layer δ is thin provided that Rex is large.

●​ (i.e., δ/x → 0 as Rex → ∞)

Notice: This type of reduction is not generally possible.

With the velocity profile known, it is easy to determine the wall shear stress.
∂𝑢
​ τ𝑤𝑎𝑙𝑙 = µ |
∂𝑦 𝑦=0

∂𝑢 𝑈 3/2 1
​ |
∂𝑦 𝑦=0
=𝑈 ν𝑥
𝑓''(η)|𝑦=0 = 0. 332𝑈 ν𝑥
3/2 ρµ
​ τ𝑤𝑎𝑙𝑙 = 0. 332𝑈 𝑥

Drag calculation:

The momentum integral method provides a simple alternate approximation to calculate the drag
force.

Assume that the pressure variation is negligible throughout the flow. Also, the fluid is considered
incompressible.
→ →
​ Σ𝐹𝑥 = ρ∫ ∫ 𝑢𝑉 · 𝑛 𝑑𝐴 + ρ∫ ∫ 𝑢𝑉 · 𝑛 𝑑𝐴
(1) (2)

​ Σ𝐹𝑥 =− 𝐷 =− ∫ ∫ τ𝑤𝑎𝑙𝑙 𝑑𝐴 =− 𝑏 ∫ τ𝑤𝑎𝑙𝑙 𝑑𝑥

𝑝𝑙𝑎𝑡𝑒 𝑝𝑙𝑎𝑡𝑒
 → → 2
​ − 𝐷 = ρ∫ ∫ 𝑢𝑉 · 𝑛 𝑑𝐴 + ρ∫ ∫ 𝑢𝑉 · 𝑛 𝑑𝐴 = ρ∫ ∫ 𝑈(− 𝑈) 𝑑𝐴 + ρ∫ ∫ 𝑢 𝑑𝐴
(1) (2) (1) (2)
δ
2 2
​ 𝐷 = ρ𝑈 𝑏ℎ − ρ𝑏 ∫ 𝑢 𝑑𝑦
0

Conservation of mass:
δ
​ 𝑈ℎ = ∫ 𝑢 𝑑𝑦
0
δ
2
​ ρ𝑈 𝑏ℎ = ρ𝑏 ∫ 𝑈𝑢 𝑑𝑦
0
δ
2
​ 𝐷 = ρ𝑏 ∫ 𝑢(𝑈 − 𝑢) 𝑑𝑦 ≈ ρ𝑏𝑈 Θ
0
𝑑𝐷 2 𝑑Θ
​ 𝑑𝑥
= ρ𝑏𝑈 𝑑𝑥
𝑑𝐷
​ 𝑑𝑥
= 𝑏τ𝑤𝑎𝑙𝑙
2 𝑑Θ
​ τ𝑤𝑎𝑙𝑙 = ρ𝑈 𝑑𝑥

General velocity profile:

𝑦
​ 𝑌= δ
∈ [0, 1]
𝑢
​ 𝑈
= 𝑔(𝑌)
𝑢
​ 𝑈
= 1 𝑓𝑜𝑟 𝑌 > 1

δ 1
2
​ 𝐷 = ρ𝑏 ∫ 𝑢(𝑈 − 𝑢) 𝑑𝑦 = ρ𝑏𝑈 δ ∫ 𝑔(𝑌){1 − 𝑔(𝑌)} 𝑑𝑌
0 0

( )
1
2
​ 𝐷 = ρ𝑏𝑈 δ𝐶1 𝐶1 = ∫ 𝑔(𝑌){1 − 𝑔(𝑌)} 𝑑𝑌
0
𝑑𝐷 2 𝑑δ
​ 𝑑𝑥
= ρ𝑏𝑈 𝐶1 𝑑𝑥
= 𝑏τ𝑤𝑎𝑙𝑙
​ (The value of C1 is independent of x.)

​ τ𝑤𝑎𝑙𝑙 = µ
∂𝑢
|
∂𝑦 𝑦=0
= µ
𝑈 𝑑𝑔
|
δ 𝑑𝑌 𝑌=0
= µ
𝑈
δ (
𝐶2 𝐶2 =
𝑑𝑔
| )
𝑑𝑌 𝑌=0
2 𝑑δ 𝑈
​ ρ𝑈 𝐶1 𝑑𝑥 = µ δ
𝐶2
ν𝐶2
​ δ 𝑑δ = 𝑈𝐶1
𝑑𝑥
δ 𝑥 ν𝐶 ν𝐶2𝑥
2 1 2
∫ δ 𝑑δ = ∫ 𝑈𝐶1
𝑑𝑥 ⇒ 2
δ = 𝑈𝐶1
0 0

2ν𝐶2𝑥
δ= 𝑈𝐶1
 δ 2ν𝐶2 2 𝐶2
𝑥
= 𝑈𝐶1𝑥
= 𝑅𝑒𝑥 𝐶1

𝑈 𝐶1𝐶2 3/2 ρµ 3/2 ρµ

τ𝑤𝑎𝑙𝑙 = µ δ
𝐶2 = 2
𝑈 𝑥
∼𝑈 𝑥

To use the above equation, one must first determine the values of C1 and C2. To do so, one uses
one of the approximations given above, where the Blasius solution is considered to be exact.

Local friction coefficient:

τ𝑤𝑎𝑙𝑙
​ 𝑐𝑓 = 1 2
2
ρ𝑈

µ 2𝐶1𝐶2
​ 𝑐𝑓 = 2𝐶1𝐶2 ρ𝑈𝑥
= 𝑅𝑒𝑥

0.664
Blasius solution: 𝑐𝑓 =
𝑅𝑒𝑥

The friction drag coefficient for a flat plate of length l and width b:
𝑙

𝐷𝑓 𝑏∫τ𝑤𝑎𝑙𝑙 𝑑𝑥 𝑙
1
​ 𝐶𝐷𝑓 = 1 2 = 0
1 2 = 𝑙
∫ 𝑐𝑓 𝑑𝑥
2
ρ𝑈 𝑏𝑙 2
ρ𝑈 𝑏𝑙 0
𝑙
1 µ −1/2
​ 𝐶𝐷𝑓 = 𝑙
∫ 2𝐶1𝐶2 ρ𝑈
𝑥 𝑑𝑥
0
𝑙
​ =
1
𝑙
2𝐶1𝐶2
µ
ρ𝑈 [2𝑥1/2]0 = 2 2𝐶1𝐶2
µ −1/2
ρ𝑈
𝑙 = 8𝐶1𝐶2
µ
ρ𝑈𝑙

8𝐶1𝐶2 1.328
​ 𝐶𝐷𝑓 = 𝑅𝑒𝑙
=
𝑅𝑒𝑙

Usually, a parabolic approximation of the Blasius solution is used.