CHEN 801:

ADVANCED TRANSPORT PHENOMENA

Lecture Note

By
Dr. A. Abubakar

Department of Chemical Engineering

Ahmadu Bello University, Zaria
1
▪ COURSE OUTLINE
➢ Fundamentals of Fluid Flow
➢ Differential Equations of Fluid Flow
➢ Approximate Solutions of the Fluid Flow Equations
➢ Turbulence Models
➢ Introduction to Computational Fluid Dynamics (CFD)

▪ REFERENCE TEXT BOOKS

➢ ‘Fluid Mechanics: Fundamentals and Applications’ by
Yunus S. Cengel and John M. Cimbala.
➢ ‘Fundamental of momentum, Heat and Mas Transfer’ by
James R, Welty, Charles E. Wicks, Robert E. Wilson,
Gregory L. Rorrer.
➢ ‘Fluid Mechanics’ by Frank M. White
➢ ‘Coulson and Richardson’s Chemical Engineering’, Vol. 1.
2
 TOPIC 3

APPROXIMATE SOLUTIONS
OF THE FLUID FLOW
EQUATIONS

3
3.1 Introduction
▪ Vast majority of practical fluid flow problems (especially those
involving Navier-Stokes equation) cannot be solved analytically and
require either (1) further approximations for laminar flow problems or
(2) computer assistance (usually CFD) for some laminar flow
problems and all turbulent flow problems. We will consider the first
option for now.
▪ The approximations of the fluid flow problems are appropriate in a
whole flow field, but in most cases, they are appropriate only in
certain regions of the flow field.
▪ The term exact is used when the solution starts with the full Navier–
Stokes equation. An approximate solution, on the other hand, is one in
which the Navier–Stokes equation is simplified in some region of the
flow before we even start the solution. In other words, term(s) are
eliminated a priori depending on the class of problem.

“Exact” solutions Approximate solutions

Full Navier-Stokes equation Simplified Navier-Stokes equation
Analysis Analysis
Solution Solution
4
▪ It is important to stress that in most practical fluid flow problems, a
particular approximation may be appropriate in a certain region of the
flow field, but not in other regions, where a different approximation
may perhaps be more appropriate. Fig. 3.1 below illustrates this point
qualitatively for flow of a liquid from one tank to another.

Fig. 3.1: Flow of a liquid from one tank to another with different types
of approximations for different regions of the flow field.
▪ Now, how do we determine if an approximation is appropriate? We do
this by comparing the orders of magnitude of the various terms in the
equations of motion to see if any terms are negligibly small compared
to other terms. These orders of magnitude of the terms are determined
by nondimensionalizing the conservations equations. 5
3.2 Nondimensionalized Differential Equations
▪ Recalling the incompressible continuity Navier-Stokes equations:
𝛁∙𝐯=0 (3.1)
𝐷𝐯 𝜕𝐯
𝜌 =𝜌 + (𝐯 ∙ 𝛁)𝐯 = 𝜌𝐠 − 𝛁𝑃 + 𝜇𝛁 𝟐 𝐯 (3.2)
𝐷𝑡 𝜕𝑡
Pressure term
Unsteady Viscous term
Gravity term
acceleration term
Advective or inertial
acceleration term

▪ We introduce in Table 3.1 some characteristic (reference) scaling

parameters that are used to nondimensionalize Eqs. (3.1) and (3.2).
Table 3.1: Scaling parameters used to nondimensionalize the continuity
and momentum equations, along with their primary dimensions
Scaling Parameter Description Primary Dimensions
𝐿 Characteristic length {L}
𝑉 Characteristic speed {Lt-1}
𝑓 Characteristic frequency {t-1}
𝑃𝑜 − 𝑃∞ Reference pressure difference {mL-1t-2}
𝑔𝑜 Gravitational acceleration {Lt-2} 6
▪ We then define several non-dimensional variables and one non-
dimensional operator based on the scaling parameters in Table 3.1 as
follows:
∗ ∗
𝒙 ∗
𝐯 ∗
𝑃 − 𝑃∞ ∗
𝒈
𝑡 = 𝑓𝑡 ; 𝑥 = ; 𝐯 = ; 𝑃 = ; 𝑔 =
𝐿 𝑉 𝑃𝑜 − 𝑃∞ 𝑔𝑜
𝜕 𝜕 𝜕 𝜕 𝜕 𝜕
𝛁= , , = 𝑥 , 𝑦 , 𝑧
𝜕𝑥 𝜕𝑦 𝜕𝑧 𝐿𝜕 𝐿 𝐿𝜕 𝐿 𝐿𝜕 𝐿 (3.3)
1 𝜕 𝜕 𝜕 1 ∗
= , , = 𝛁
𝐿 𝜕𝑥 ∗ 𝜕𝑦 ∗ 𝜕𝑧 ∗ 𝐿

𝛁 ∗ = 𝐿𝛁

▪ Each of the starred quantities in Eq. (3.3) is non-dimensional. For

example, although each component of the gradient operator 𝛁
has dimensions of {L-1}, each component of 𝛁 ∗ has dimensions of {1}.
▪ Therefore, we first substitute Eq. (3.3) into Eq. (3.1), treating each
term carefully as follows:
𝛁∗ 𝑉 ∗ ∗
∙ 𝑉𝐯 ∗ = 0 𝛁 ∙𝐯 =0
𝐿 𝐿 7
▪ After dividing both sides by 𝑉/𝐿 to make the above equation
dimensionless, we get;
Nondimensionalized continuity equation:
𝛁∗ ∙ 𝐯∗ = 0 (3.4)

▪ Similarly, Eq. (3.2), after appropriate substitutions, becomes:

𝜕𝑉𝐯 ∗
𝜌 ∗
+ (𝑉𝐯 ∗ ∙ 𝛁 ∗ Τ𝐿)𝑉𝐯 ∗ = 𝜌𝑔∗ 𝑔𝑜 − 𝛁 ∗ Τ𝑳 𝑃𝑜 − 𝑃∞ 𝑃∗ + 𝜇 𝛁 ∗ Τ𝑳 𝟐 𝑉𝐯 ∗
𝜕 𝑡 Τ𝑓
▪ Rearranging the above equation gives;
𝜕𝐯 ∗ 𝜌𝑉 2 ∗ ∗ ∗ ∗
𝑃𝑜 − 𝑃∞ ∗ ∗ 𝜇𝑉 ∗ 2 ∗
𝜌𝑉𝑓 ∗ + 𝐯 ∙ 𝛁 𝐯 = 𝜌𝑔𝑜 𝑔 − 𝛁 𝑃 + 2𝛁 𝐯
𝜕𝑡 𝐿 𝐿 𝐿
▪ Now, multiplying the above equation by the collection of constants
𝐿/(𝜌𝑉 2 ) to make all the terms dimensionless gives;
𝑓𝐿 𝜕𝐯 ∗ 𝑔𝑜 𝐿 ∗ 𝑃𝑜 − 𝑃∞ 𝜇
∗ ∗ ∗
+ 𝐯 ∙𝛁 𝐯 = 𝑔 − ∗ ∗
𝛁 𝑃 + 𝛁∗2𝐯∗ (3.5)
𝑉 𝜕𝑡 ∗ 𝑉2 𝜌𝑉 2 𝜌𝑉𝐿

▪ Each of the terms in square brackets in Eq. (3.5) is a non-

dimensional grouping of parameters. They are listed below.
8
▪ Strouhal number, St = 𝑓𝐿Τ𝑉; Froude number, Fr = 𝑉 Τ 𝑔𝑜 𝐿; Euler
number, Eu = 𝑃𝑜 − 𝑃∞ Τ𝜌𝑉 2 ; and Reynolds number, Re = 𝜌𝑉𝐿Τ𝜇.
▪ Thus, Eq. (3.5) becomes;
Nondimensionalized Navier-Stokes equation:
𝜕𝐯 ∗ ∗ ∗ ∗
1 ∗ ∗ ∗
1 ∗2 ∗
St + 𝐯 ∙𝛁 𝐯 = 𝑔 − Eu 𝛁 𝑃 + 𝛁 𝐯 (3.6)
𝜕𝑡 ∗ Fr 2 Re

▪ The following comments can be made about Eqs. (3.4) and (3.6):
➢ Eq. (3.4) contains no additional dimensionless parameters. Hence, it
cannot be simplified further because all the terms are of the same
order of magnitude.
➢ The order of magnitude of the non-dimensional variables is unity if
they are nondimensionalized using variables that are characteristic of
the flow field. Thus, the relative importance of the terms in Eq. (3.6)
depends only on the relative magnitudes of the dimensionless
parameters St, Eu, Fr, and Re.
➢ If the flow is steady, then f = 0 and the Strouhal number drops out of
the list of dimensionless parameters (St = 0). The first term on the left
side of Eq. (3.6) then disappears, as does its corresponding unsteady
term 𝜕𝐯Τ𝜕𝑡 in Eq. (3.1). 9
▪ For flows without free-surface effects, gravity does not affect the
dynamics of the flow—its only effect is to superpose a hydrostatic
pressure on the dynamic pressure field.
▪ Thus, we can define a modified pressure 𝑃′ that absorbs the effect of
hydrostatic pressure. For the case in which z is defined vertically
upward (opposite to the direction of the gravity vector), and in which
we define some arbitrary reference datum plane at z = 0, we can write;
Modified pressure: 𝑃′ = 𝑃 + 𝜌𝑔𝑧 (3.7)

▪ The idea is to replace the two terms −𝛁𝑃 + 𝜌𝐠 with one term −𝛁𝑃′
using the modified pressure of Eq. (3.7) so that the equation becomes;

𝐷𝐯 𝜕𝐯
𝜌 =𝜌 + (𝐯 ∙ 𝛁)𝐯 = −𝛁𝑃′ + 𝜇𝛁 𝟐 𝐯 (3.8)
𝐷𝑡 𝜕𝑡
▪ With 𝑃 replaced by 𝑃′ , and with the gravity term disappeared from Eq.
(3.8), the Froude number drops out of the list of dimensionless
parameters. The advantage is that we can solve a form of the Navier–
Stokes equation that has no gravity term. After solving the Navier–
Stokes equation in terms of modified pressure 𝑃′ , it is a simple matter
to add back the hydrostatic pressure. 10
3.3 Types of Approximate Solutions
3.3.1 Creeping flow approximation
▪ Creeping flow, also known as stokes flow or low Reynolds number
flow is a class of fluid in which the Reynolds number is very small (Re
≪ 1). E.g. the flow of honey or syrup is a creeping flow.
▪ For simplicity, we assume that gravitational effects are negligible, or
that they contribute only to a hydrostatic pressure component, as
discussed previously. We also assume either steady flow so the
unsteady acceleration term can be dropped out. In addition, the
advective acceleration term can also be dropped out since its order of
magnitude which is one is much smaller than the order of magnitude
of the viscous term.
▪ Therefore, approximate Navier-Stokes equation for creeping flow is;
1
Eu 𝛁∗ 𝑃∗ ≅ 𝛁∗2𝐯∗ (3.9)
Re
Or in dimensional form;
𝛁𝑃 ≅ 𝜇𝛁 𝟐 𝐯 (3.10)
11
3.3.2 Inviscid and irrotational regions of flow approximations
▪ Inviscid flow does not mean flow without viscosity because all fluids
of engineering relevance have viscosity. Inviscid flow actually mean
flow of a viscous fluid in a region of the flow field in which net
viscous forces are negligible compared to pressure and/or inertial
forces.
▪ An irrotational flow (also known as potential flow) is characterized
by negligible vorticity. An irrotational flow region of the flow also have
negligible net viscous forces—not because there is no friction, but
because the frictional (viscous) stresses cancel each other out.
▪ Therefore, in both inviscid and irrotational regions of flow field, the
viscous term of the Navier-Stokes equation drops out and then
reduces to Euler equation as follows:
𝜕𝐯 ∗ ∗ ∙ 𝛁∗ 𝐯∗ =
1
St ∗
+ 𝐯 2
𝑔∗ − Eu 𝛁 ∗ 𝑃∗ (3.11)
𝜕𝑡 Fr

Or in dimensional form;
𝜕𝐯
𝜌 + (𝐯 ∙ 𝛁)𝐯 = 𝜌𝐠 − 𝛁𝑃 (3.12)
𝜕𝑡 12
▪ For steady flow, the unsteady acceleration term drops out and Eq.
(3.12) can be written in expanded forms as follows:
𝜕𝑣𝑥 𝜕𝑣𝑥 𝜕𝑣𝑥 𝜕𝑃
𝑥 direction: 𝜌 𝑣𝑥
𝜕𝑥
+ 𝑣𝑦
𝜕𝑦
+ 𝑣𝑧
𝜕𝑧
= 𝜌𝑔𝑥 −
𝜕𝑥
𝜕𝑣𝑦 𝜕𝑣𝑦 𝜕𝑣𝑦 𝜕𝑃
𝑦 direction: 𝜌 𝑣𝑥 + 𝑣𝑦 + 𝑣𝑧 = 𝜌𝑔𝑦 − (3.13)
𝜕𝑥 𝜕𝑦 𝜕𝑧 𝜕𝑦

𝜕𝑣𝑧 𝜕𝑣𝑧 𝜕𝑣𝑧 𝜕𝑃

𝑧 direction: 𝜌 𝑣𝑥 + 𝑣𝑦 + 𝑣𝑧 = 𝜌𝑔𝑧 −
𝜕𝑥 𝜕𝑦 𝜕𝑧 𝜕𝑧

Bernoulli equation in inviscid and irrotational flow regions

▪ Since both inviscid and irrotational approximations involve negligible
net frictional (or viscous) forces, Bernoulli equations derived for both
cases in regions of steady, incompressible flow are as follows:
𝑃 v2
Inviscid approximation: + + 𝑔𝑧 = 𝐶 = constant along streamlines (3.14a)
𝜌 2
𝑃 v2
Irrotational approximation: + + 𝑔𝑧 = 𝐶 = constant everywhere (3.14b)
𝜌 2

where v is the magnitude of vector v 13

▪ Eqs. (3.14) above means that in an inviscid region of flow, Bernoulli
constant only holds along a streamline and may change from
streamline to streamline while in an irrotational region of flow, the
Bernoulli constant is the same everywhere even across streamlines.
▪ Thus, the irrotational approximation is more restrictive than the
inviscid approximation.

3.3.3. Boundary layer approximation

▪ The boundary layer approximation bridges the gap between the
Euler equation and the Navier–Stokes equation, and between the
slip condition and the no-slip condition at solid walls (see Fig. 3.2).

(a) (b)
Fig. 3.2: (a) A huge gap exists between the Euler equation (which allows
slip at walls) and the Navier–Stokes equation (which supports the no-
slip condition); (b) the boundary layer approximation bridges that gap
14
▪ Boundary layer is the flow region adjacent to the wall in which the
viscous effects (and thus the velocity gradients) are significant.
▪ Although dropping the viscous terms to form Euler equation greatly
simplifies the calculation, there are some serious deficiencies
associated with application of the equation to practical engineering
flow problems. High on the list of deficiencies is the inability to
specify the no-slip condition at solid walls.
▪ Thus, Ludwig Prandtl (1875–1953) in 1904 introduced the boundary
layer approximation to correct some of the major deficiencies.
Prandtl’s idea was to divide the flow into two regions: an outer flow
region that is inviscid and/or irrotational, and an inner flow region
called a boundary layer—a very thin region of flow near a solid wall
where viscous forces and rotationality cannot be ignored (Fig. 3.3).
▪ In the outer flow region, we use the continuity and Euler equations to
obtain the outer flow velocity field, and the Bernoulli equation to
obtain the pressure field. We solve for the outer flow region first, and
then fit in a thin boundary layer in regions where rotationality and
viscous forces cannot be neglected. Within the boundary layer, we
solve the boundary layer equations, which are presented below.
15
 Fig. 3.3: Prandtl’s boundary layer concept splits the flow into an outer
flow region and a thin boundary layer region (not to scale)

▪ The key to successful application of the boundary layer approximation

is the assumption that the boundary layer is very thin.

Fig. 3.4: Flow of a uniform stream parallel to a flat plate (drawings not to
scale): (a) 𝑹𝒆𝒙 ~𝟏𝟎𝟐 , (b) 𝑹𝒆𝒙 ~𝟏𝟎𝟒 . The larger the Reynolds number, the
thinner the boundary layer along the plate at a given x-location. 16
▪ At a given x-location, it has found that the higher the Reynolds
number, the thinner the boundary layer. This is illustrated in Fig. 3.4
▪ Boundary layer thickness is denoted as 𝛿. It is defined as the distance
away from the wall at which the velocity component parallel to the
wall is 99 percent of the fluid speed outside the boundary layer.
▪ In other words, the higher the Reynolds number, the thinner the
boundary layer, all else being equal, and the more reliable the
boundary layer approximation. We are confident that the boundary
layer is thin when δ ≪ 𝑥 (or, expressed non-dimensionally, δ/𝑥 ≪ 1).

Derivation of boundary layer equations

▪ For simplicity we consider only steady, two dimensional flow in the 𝑥𝑦-
plane in Cartesian coordinates. In addition,
❖ We consider only laminar boundary layers
❖ We neglect gravity since we are not dealing with free surfaces
or with buoyancy-driven flows (free convection flows), where
gravitational effects dominate.
▪ We will begin with the nondimensionalized Navier–Stokes equation.
With the unsteady term and the gravity term neglected, Eq. (3.6)
becomes; 17
 1 ∗2 ∗
𝐯 ∗ ∙ 𝛁 ∗ 𝐯 ∗ = − Eu 𝛁 ∗ 𝑃∗ + 𝛁 𝐯 (3.15)
Re

▪ Now, how do we decide which terms to keep and which to neglect? To

answer this question, we redo the nondimensionalization of the
equations of motion (i.e. continuity and Navier-Stokes equations)
based on appropriate length and velocity scales within the boundary
layer using the magnified view of a portion of the boundary layer of
Fig. 3.5 sketched in Fig. 3.6.

Fig. 3.5: The boundary layer coordinate Fig. 3.6: Magnified view of
system for flow over a body; x follows the boundary layer along
the surface and is typically set to zero at the surface of a body,
the front stagnation point of the body, showing length scales 𝒙
and y is everywhere normal to the and 𝜹, and velocity scale 𝑼
surface locally
18
▪ NOTE:
❖ Since the order of magnitude of 𝑥 is 𝐿, we use 𝐿 as an appropriate
length scale for distances in the streamwise direction and for
derivatives of velocity and pressure with respect to 𝑥. However, this
length scale is much too large for derivatives with respect to 𝑦. It
makes more sense to use 𝛿 as the length scale for distances in the
direction normal to the streamwise direction and for derivatives with
respect to 𝑦.
❖ Similarly, while the characteristic velocity scale is 𝑉 for the whole
flow field, it is more appropriate to use 𝑈 as the characteristic
velocity scale for boundary layers, where U is the magnitude of the
velocity component parallel to the wall at a location just above the
boundary layer (Fig. 3.5b). 𝑈 is in general a function of 𝑥.
▪ Thus, within the boundary layer at some value of x, the orders of
magnitude are;
𝜕 1 𝜕 1
𝑣𝑥 ~ 𝑈 ; 𝑃 − 𝑃∞ ~ 𝜌𝑈 2 ; ~ ; ~ (3.16)
𝜕𝑥 𝐿 𝜕𝑦 𝛿
▪ The order of magnitude of velocity component 𝑣𝑦 is not specified in
Eq. (3.16), Instead, it is obtained from the continuity equation by
substituting Eq. (3.16) as follows: 19
 𝜕𝑣𝑥 𝜕𝑣𝑦
+ =0
𝜕𝑥 𝜕𝑦
𝑈 𝑣𝑦 𝑈 𝑣𝑦
+ =0 ~
𝐿 𝛿 𝐿 𝛿

▪ Since the two terms have to balance each other, they must be of the
same order of magnitude. Thus we obtain the order of magnitude of
velocity component 𝑣𝑦 as;.
𝑈𝛿
𝑣𝑦 ~ (3.17)
𝐿

▪ Since δ/𝐿 ≪1 in a boundary layer (the boundary layer is very thin), we

conclude that 𝑣𝑦 ≪ 𝑣𝑥 in a boundary layer.
▪ From Eqs. 3.16 and 3.17, we define the following non-dimensional
variables within the boundary layer:
𝑥 𝑦 𝑣𝑥 𝑣𝑦 𝐿 𝑃 − 𝑃∞
𝑥∗ = ; 𝑦∗ = ; 𝑣𝑥 ∗ = ; 𝑣𝑦 ∗ = ; 𝑃∗ = (3.18)
𝐿 𝛿 𝑈 𝑈𝛿 𝜌𝑈 2
▪ Since we used appropriate scales, all these non-dimensional variables
are of order unity—i.e., they are normalized variables (Topic 1).
20
▪ We now substitute the non-dimensional variables into the 𝑥 − and
𝑦 −components of the Navier–Stokes equation.
1. 𝒚-component of the Navier–Stokes equation
𝜕𝑣𝑦 𝜕𝑣𝑦 1 𝜕 𝑃 − 𝑃∞ 𝜕 2 𝑣𝑦 𝜕 2 𝑣𝑦
𝑣𝑥 + 𝑣𝑦 =− +𝜗 +𝜗
𝜕𝑥 𝜕𝑦 𝜌 𝜕𝑦 𝜕𝑥 2 𝜕𝑦 2
where 𝜗 = 𝜇Τ𝜌 = Kinematic viscosity
∗ ∗ ∗
∗
𝜕 𝑣𝑦 𝑈𝛿 Τ 𝐿 𝑣𝑦 𝑈𝛿 𝜕 𝑣𝑦 𝑈𝛿 Τ𝐿 1 𝜕 𝑃∗ 𝜌𝑈 2
𝑣𝑥 𝑈 + =−
𝜕 𝑥 ∗𝐿 𝐿 𝜕 𝑦 ∗𝛿 𝜌 𝜕 𝑦 ∗𝛿

𝜕 2 𝑣𝑦 ∗ 𝑈𝛿 Τ𝐿 𝜕 2 𝑣𝑦 ∗ 𝑈𝛿 Τ𝐿
+𝜗 +𝜗
𝜕 𝑥 ∗𝐿 2 𝜕 𝑦 ∗𝛿 2

▪ After some algebra and after multiplying each term by 𝐿2 /(𝑈 2 𝛿), we
get;
2 2
∗
𝜕𝑣𝑦 ∗ ∗
𝜕𝑣𝑦 ∗ 𝐿 𝜕𝑃∗ 𝜗 𝜕 2 𝑣𝑦 ∗ 𝐿 𝜗 𝜕 2 𝑣𝑦 ∗
𝑣𝑥 + 𝑣𝑦 =− + + (3.19)
𝜕𝑥 ∗ 𝜕𝑦 ∗ 𝛿 𝜕𝑦 ∗ 𝑈𝐿 𝜕𝑥 ∗ 2 𝛿 𝑈𝐿 𝜕𝑦 ∗ 2

▪ Comparing terms in Eq. 3.19, the middle term on the right side is
clearly orders of magnitude smaller than any other term since 𝑅𝑒𝐿 =
𝑈𝐿Τ𝜗 ≫ 1. 21
▪ For the same reason, the last term on the right is much smaller than
the first term on the right. Neglecting these two terms leaves the two
terms on the left and the first term on the right.
▪ However, since 𝐿 ≫ 𝛿 , the pressure gradient term is orders of
magnitude greater than the advective terms on the left side of the
equation. Thus, the only term left in Eq. 3.19 is the pressure term.
Since no other term in the equation can balance that term, we have
no choice but to set it equal to zero.
▪ Thus, the non-dimensional 𝑦-momentum equation reduces to;
𝜕𝑃∗
≅0
𝜕𝑦 ∗
or, in terms of the physical variables,
𝜕𝑃
≅0 (3.20)
𝜕𝑦

▪ In words, although pressure may vary along the wall (in the 𝑥 -
direction), there is negligible change in pressure in the direction
normal to the wall. This is illustrated in Fig. 3.7.
▪ In summary, the pressure across a boundary layer (𝑦-direction) is
nearly constant. 22
 Fig. 3.8: The pressure in the
Fig. 3.7: Pressure may change
irrotational region of flow outside
along a boundary layer (x-
of a boundary layer can be
direction), but the change in
measured by static pressure taps
pressure across a boundary
in the surface of the wall. Two
layer (y-direction) is negligible
such pressure taps are sketched.

▪ One immediate consequence of Eq. (3.20) and the statement just

presented is that at any x-location along the wall, the pressure at the
outer edge of the boundary layer (𝑦 ≅ 𝛿) is the same as that at the
wall ( 𝑦 = 0 ). This leads to a tremendous practical application;
namely, the pressure at the outer edge of a boundary layer can be
measured experimentally by a static pressure tap at the wall directly
beneath the boundary layer (Fig. 3.8).
23
2. 𝒙-component of the Navier–Stokes equation
𝜕𝑣𝑥 𝜕𝑣𝑥 1 𝜕 𝑃 − 𝑃∞ 𝜕 2 𝑣𝑥 𝜕 2 𝑣𝑥
𝑣𝑥 + 𝑣𝑦 =− +𝜗 +𝜗
𝜕𝑥 𝜕𝑦 𝜌 𝜕𝑥 𝜕𝑥 2 𝜕𝑦 2

∗
𝜕 𝑣𝑥 ∗ 𝑈 𝑣𝑦 ∗ 𝑈𝛿 𝜕 𝑣𝑥 ∗ 𝑈 1 𝜕 𝑃∗ 𝜌𝑈 2
𝑣𝑥 𝑈 + =−
𝜕 𝑥 ∗𝐿 𝐿 𝜕 𝑦 ∗𝛿 𝜌 𝜕 𝑥 ∗𝐿

𝜕 2 𝑣𝑥 ∗ 𝑈 𝜕 2 𝑣𝑥 ∗ 𝑈
+𝜗 +𝜗
𝜕 𝑥 ∗𝐿 2 𝜕 𝑦 ∗𝛿 2

▪ After some algebra, and after multiplying each term by 𝐿/𝑈 2 , , we get

2
∗
𝜕𝑣𝑥 ∗ ∗
𝜕𝑣𝑥 ∗ 𝜕𝑃∗ 𝜗 𝜕 2 𝑣𝑥 ∗ 𝐿 𝜗 𝜕 2 𝑣𝑥 ∗
𝑣𝑥 + 𝑣𝑦 =− ∗+ + (3.21)
𝜕𝑥 ∗ 𝜕𝑦 ∗ 𝜕𝑥 𝑈𝐿 𝜕𝑥 ∗ 2 𝛿 𝑈𝐿 𝜕𝑦 ∗ 2

▪ Comparing terms in Eq. 3.21, the middle term on the right side is
clearly orders of magnitude smaller than the terms on the left side,
since 𝑅𝑒𝐿 = 𝑈𝐿Τ𝜗 ≫ 1.
▪ What about the last term on the right? If we neglect this
term, we throw out all the viscous terms and are back to the Euler
equation. Clearly this term must remain.
24
▪ Furthermore, since all the remaining terms in Eq. 3.21 are of order
unity, the combination of parameters in parentheses in the last term
on the right side of the equation must also be of order unity. i.e.
2
𝐿 𝜗
~1
𝛿 𝑈𝐿

▪ Again recognizing that 𝑅𝑒𝐿 = 𝑈𝐿Τ𝜗, we see immediately that

𝐿 1
~ (3.22)
𝛿 𝑅𝑒𝐿

▪ Eq. (3.22) confirms our previous statement that at a given streamwise

location along the wall, the larger the Reynolds number, the thinner
the boundary layer.
▪ In terms of the original (physical) variables, Eq. (3.21), after removing
the middle term on the right side, is written as;
𝜕𝑣𝑥 𝜕𝑣𝑥 1 𝜕𝑃 𝜕 2 𝑣𝑥
𝑣𝑥 + 𝑣𝑦 =− +𝜗 (3.23)
𝜕𝑥 𝜕𝑦 𝜌 𝜕𝑥 𝜕𝑦 2
▪ Since 𝑃 not a function of 𝑦 as earlier found from y-momentum
equation analysis , we can replace 𝜕𝑃Τ𝜕𝑥 in Eq. (3.23) with 𝑑𝑃Τ𝑑𝑥 to
give; 25
 𝜕𝑣𝑥 𝜕𝑣𝑥 1 𝑑𝑃 𝜕 2 𝑣𝑥
𝑣𝑥 + 𝑣𝑦 =− +𝜗 (3.24)
𝜕𝑥 𝜕𝑦 𝜌 𝑑𝑥 𝜕𝑦 2

▪ Note that the last term in Eq. (3.24) is not negligible in the boundary
layer, since the 𝑦-derivative of velocity gradient 𝜕𝑣𝑥 /𝜕𝑦 is sufficiently
large to offset the (typically small) value of kinematic viscosity 𝜗.
▪ Finally, since we also know from our 𝑦-momentum equation analysis
that the pressure across the boundary layer is the same as that
outside the boundary layer, we can apply the Bernoulli equation to
the outer flow region to find expression for the the pressure.
Therefore, neglecting the elevation term in the Bernoulli equation
[Eq. (3.14)] and differentiating it with respect to 𝑥, we get
𝑃 𝑈2 1 𝑑𝑃 𝑑𝑈 (3.25)
+ = constant − =𝑈
𝜌 2 𝜌 𝑑𝑥 𝑑𝑥
where we note that both 𝑃 and 𝑈 are functions of 𝑥 only, as
illustrated in Fig. 3.9 below.
▪ Substitution of Eq. (3.25) into Eq. (3.24) yields;
𝜕𝑣𝑥 𝜕𝑣𝑥 𝑑𝑈 𝜕 2 𝑣𝑥
𝑣𝑥 + 𝑣𝑦 =𝑈 +𝜗 (3.26)
𝜕𝑥 𝜕𝑦 𝑑𝑥 𝜕𝑦 2
26
 Fig. 3.9: Outer flow speed parallel to the wall is 𝑼(𝒙) and is obtained from
the outer flow pressure, 𝑷(𝒙). This speed appears in the 𝒙-component of
the boundary layer momentum equation, Eq. (3.26).

▪ In summary, for a steady, incompressible, laminar boundary layer in

the xy-plane without significant gravitational effects, the boundary
layer equations are;
𝜕𝑣𝑥 𝜕𝑣𝑦
Continuity equation: + =0
𝜕𝑥 𝜕𝑦

𝜕𝑣𝑥 𝜕𝑣𝑥 𝑑𝑈 𝜇 𝜕 2 𝑣𝑥
𝑥-momentum equation: 𝑣𝑥 + 𝑣𝑦 =𝑈 + (3.27)
𝜕𝑥 𝜕𝑦 𝑑𝑥 𝜌 𝜕𝑦 2

𝜕𝑃
𝑦-momentum equation: ≅0
𝜕𝑦
27
▪ Mathematically, the full Navier–Stokes equation is elliptic in space,
which means that boundary conditions are required over the entire
boundary of the flow domain. In contrast, the x-momentum boundary
layer equation (the second equation of Eq. 3.27) is parabolic. This means
that we need to specify boundary conditions on only three sides of the
(two-dimensional) flow domain. Physically, flow information is not passed
in the direction opposite to the flow (from downstream). This fact greatly
reduces the level of difficulty in solving the boundary layer equations.
Specifically, we don’t need to specify boundary conditions downstream,
only upstream and on the top and bottom of the flow domain (Fig. 3.10).

Fig. 3.10: The boundary layer equation set is parabolic, so boundary

conditions need to be specified on only three sides of the flow domain.
28