Blasius Solution: Exact Fluid Dynamics Solution for Flat Plate, Study notes of Fluid Dynamics

An in-depth analysis of the blasius solution for a flat plate boundary layer, the first exact solution to the laminar boundary layer equations. The mathematical problem, the blasius equations, and the boundary conditions. It also discusses the concept of similarity solutions and the successful similarity variable for this problem. The document concludes with the governing equation, the choice of constants, and the boundary conditions in terms of the similarity variable. This solution is crucial for understanding the behavior of laminar boundary layers and is used throughout fluid dynamics studies.

Typology: Study notes

2021/2022

Uploaded on 09/12/2022

plastic-tree
plastic-tree 🇬🇧

4.4

(8)

213 documents

1 / 4

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
An Internet Book on Fluid Dynamics
Blasius Solution for a Flat Plate Boundary Layer
The first exact solution to the laminar boundary layer equations, discovered by Blasius (1908), was for a
simple constant value of U(s) and pertains to the case of a uniform stream of velocity, U,encounteringan
infinitely thin flat plate set parallel with that stream as shown in Figure 1:
Figure 1: Blasius flow of a boundary layer on a flat plate
Thus the laminar boundary layer equations (Bjb20) and (Bjb21) become
∂u
∂s +∂v
∂n =0 (Bjd1)
u∂u
∂s +v∂u
∂n =ν2u
∂n2(Bjd2)
and, using the stream function, ψ,givenby
u=∂ψ
∂n and v=∂ψ
∂s (Bjd3)
the single governing equation that Blasius set out to solve is
∂ψ
∂n
2ψ
∂s∂n ∂ψ
∂s
2ψ
∂n2=ν3ψ
∂n3(Bjd4)
with the following boundary conditions:
1. ψ= 0 on the solid surface, n=0,sincev=0
2. ∂ψ/∂n = 0 on the solid surface, n=0sinceu=0
3. ∂ψ/∂n Uas n→∞since uU
Notice that we have avoided using δin the formulation of the mathematical problem. The definition and
evaluation of the boundary layer thickness will follow from the results.
The Blasius solution is best presented as an example of a similarity solution to the non-linear, partial
differential equation (Bjd4). In a similarity solution we seek a similarity variable (here symbolized by η)
which is a function of sand nsuch that the unknown ψmay be written as a function of the single variable
η. If this is possible then the problem is reduced to finding the solution to an ordinary differential equation
pf3
pf4

Partial preview of the text

Download Blasius Solution: Exact Fluid Dynamics Solution for Flat Plate and more Study notes Fluid Dynamics in PDF only on Docsity!

An Internet Book on Fluid Dynamics

Blasius Solution for a Flat Plate Boundary Layer

The first exact solution to the laminar boundary layer equations, discovered by Blasius (1908), was for a simple constant value of U(s) and pertains to the case of a uniform stream of velocity, U, encountering an infinitely thin flat plate set parallel with that stream as shown in Figure 1:

Figure 1: Blasius flow of a boundary layer on a flat plate

Thus the laminar boundary layer equations (Bjb20) and (Bjb21) become

∂u ∂s

∂v ∂n

= 0 (Bjd1)

u

∂u ∂s

  • v

∂u ∂n

= ν

∂^2 u ∂n^2

(Bjd2)

and, using the stream function, ψ, given by

u =

∂ψ ∂n

and v = −

∂ψ ∂s

(Bjd3)

the single governing equation that Blasius set out to solve is

∂ψ ∂n

∂^2 ψ ∂s∂n

∂ψ ∂s

∂^2 ψ ∂n^2

= ν

∂^3 ψ ∂n^3

(Bjd4)

with the following boundary conditions:

  1. ψ = 0 on the solid surface, n = 0, since v=
  2. ∂ψ/∂n = 0 on the solid surface, n = 0 since u=
  3. ∂ψ/∂n → U as n → ∞ since u → U

Notice that we have avoided using δ in the formulation of the mathematical problem. The definition and evaluation of the boundary layer thickness will follow from the results.

The Blasius solution is best presented as an example of a similarity solution to the non-linear, partial differential equation (Bjd4). In a similarity solution we seek a similarity variable (here symbolized by η) which is a function of s and n such that the unknown ψ may be written as a function of the single variable η. If this is possible then the problem is reduced to finding the solution to an ordinary differential equation

for ψ(η). Of course, it may not be possible to find such a similarity variable in which case no such solution exists. Consequently the key step is in finding a possible η.

Let us speculate that a successful similarity variable for the present problem takes the form

η = An/sk^ (Bjd5)

where A and k are unknown constants. Then

∂η ∂n

A

sk^

and

∂η ∂s

Akn sk+^

kη s

(Bjd6)

We further speculate that the solution, ψ(η), to equation (Bjd4) takes the form

ψ = BskF (η) (Bjd7)

where B is another unknown constant and F (η) is an unknown function. It follows that

u =

∂ψ ∂n

= Bsk^

dF dη

∂η ∂n

= AB

dF dη

(Bjd8)

and it is therefore convenient and without loss of generality to choose AB = U so that

u = U

dF dη

(Bjd9)

Note that this choice is merely for convenience and could have been omitted from the solution. Next we observe that the following derivatives that appear in the governing equation (Bjd4) take these forms:

v =

∂ψ ∂s

kUsk−^1 A

η

dF dη

− F

(Bjd10)

∂^2 ψ ∂n^2

∂u ∂n

UA

sk

d^2 F dη^2

∂^2 ψ ∂n∂s

∂u ∂s

kUAn sk+

d^2 F dη^2

∂^3 ψ ∂n^3

∂^2 u ∂n^2

UA^2

s^2 k

d^3 F dη^3

(Bjd11)

and substituting these expressions into equation (Bjd4) and rearranging yields

kU^2 F

d^2 F dη^2

νUA^2 s^2 k−^1

d^3 F dη^3

(Bjd12)

This is the critical point at which we recognize that the similarity solution will work in this case if k = 1/ 2 and the governing equation (Bjd12) can be written with only F and η and without any stray s or n as ( U 2 νA^2

F

d^2 F dη^2

d^3 F dη^3

= 0 (Bjd13)

We still have the flexibility to choose a value for A and we choose A = (U/ 4 ν)

1 (^2) so that the governing equation (Bjd13) becomes

2 F

d^2 F dη^2

d^3 F dη^3

= 0 (Bjd13)

From this it is evident that F will be a function of η alone provided the boundary conditions can be written in terms of F and η alone. Indeed the boundary conditions listed above become

  1. v = 0 on the solid surface, η = 0, and therefore (dF/dη)η=0 = 0
  2. u = 0 on the solid surface, η = 0, and therefore Fη=0 = 0

involving water at normal temperatures with a kinematic viscosity of ν = 10−^6 m^2 /sec, we note that the length of the region near the leading edge would be 1μm for a flow with U = 1 m/s. Therefore it is often adequate to neglect this very small region in the flow and to assume it has negligible effect on the flow further downstream.

The above solution for a steady, planar, laminar boundary layer will be used in the sections that follow to illustrate various properties of laminar boundary layers.