For a thin boundary layer that is contiguous to the solid surface on which it has formed, the

full equations of motion for a constant-density constant-viscosity fluid, (4.10) and (9.1), may be
simplified. Let dðxÞ be the average thickness of the boundary layer at downstream location x
on the surface of a body having a local radius of curvature R. The steady-flow momentum
equation for the surface-parallel velocity component, u, is:

2 
vu vu 1 vp v u v2 u
u þv ¼  þn þ ; (10.1)
vx vy r vx vx2 vy2
which is valid when d=R  1. The more general curvilinear form for arbitrary R(x) is given in
Goldstein (1938) and Schlichting (1979), but the essential features of viscous boundary layers
can be illustrated via (10.1) without additional complications.
Let the characteristic magnitude of u be UN, the velocity at a large distance upstream of the
body, and let L be the stream-wise distance over which u changes appreciably. The longitudinal
length of the body can serve as L, because u within the boundary layer may change in the
stream-wise direction by a large fraction of UN over a distance L (Figure 10.2). A measure of
vu/vx is therefore UN/L, so that the approximate size of the first advective term in (9.1) is:

uðvu=vxÞ w UN2
L; (10.2)
where w is to be interpreted as “of order.” We shall see shortly that the other advective term
in (10.1) is of the same order. The approximate size of the viscous stress term in (10.1) is:
    2
ð1=rÞðvs=vyÞ ¼ n v2 u vy2 w nUN d : (10.3)

The magnitude of d can now be estimated by noting that the advective and viscous terms
should be of the same order within the boundary layer. Equating the magnitudes of advective
and viscous terms in (10.2) and (10.3) leads to:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffi
d w nL=UN or d=L w 1= Re: (10.4)
This estimate of d can also be obtained by noting that viscous effects diffuse to a distance of
order [nt]1/2 in time t and that the time-of-flight for a fluid element along a body of length L is
of order L/UN. Substituting L/UN for t in [nt]1/2 suggests the viscous layer’s diffusive thick-
ness at x ¼ L is of order [nL/UN]1/2, which is the duplicates (10.4).
EXAMPLE 10.1
For time-averaged turbulent boundary-layer flow, the advective acceleration scaling (10.2) is still
appropriate. However, the laminar shear stress relationship (10.3) should be replaced with
vs=vy w sw =d. What is the scaling for the skin friction coefficient in this case?

Solution
As was done to reach (10.4), equate the advective and shear-stress accelerations:


UN2
1 sw 1 1 2
w w rUN Cf ;
L r d rd 2
where the second scaling follows from (10.15), the definition of the skin friction. Canceling com-
mon terms between the two ends of this relationship then produces:

Cf w 2d=L:

Although the Reynolds number dependence of Cf is not revealed by this simple relationship, it does
suggest Cf will be much less than unity for attached turbulent boundary-layer flows. Measurements
in flat-plate turbulent boundary-layer flows on smooth walls typically produce Cf w 0.001 to 0.004
with the lower values occurring at higher Reynolds number; see Section 10.7.

EXAMPLE 10.2
In boundary-layer flows the shape factor, d*/q, is often of interest because an increasing shape
factor indicates that a boundary layer is headed toward separation. Compute the shape factor for
the following approximate-laminar (ul) and approximate-turbulent (ut) boundary-layer profiles:
( ) ( )
y y 2 y 1=7
ul ðyÞ 2

for y < d ut ðyÞ
d for y < d
¼ d d ; and ¼ ;
Ue Ue
1 for y > d 1 for y > d

where d is a profile constant in each case. Which boundary layer is closer to separation?

Solution
For the laminar profile, use (10.16) and (10.17) to find:

Zd
 Zd


y y 2 d y y 2 y y 2 2d
dl ¼ 1  2
þ
dy ¼ and ql ¼ 2

1  2
þ
dy ¼ ;
d d 3 d d d d 15
0 0
d 5
so ¼ :
q 2
Repeat for the turbulent profile to find:

Zd
y  Zd 
y 
1=7 d y 1=7 1=7 7d d 9
dt ¼ 1
dy ¼ and qt ¼
d 1
dy ¼ ; so ¼ :
d 8 d 72 q 7
0 0

For the given profiles, the laminar boundary layer has a larger shape factor and is closer to sepa-
ration. In general, turbulent boundary layers resist separation better than laminar ones.
TABLE 10.1 Blasius Boundary-Layer Profile Functions
h f df/dh d2f/dh2

0.0 0.0 0.0 0.3321

0.2 0.0068 0.0664 0.3320

0.4 0.0268 0.1328 0.3314
0.6 0.0611 0.1989 0.3299
0.8 0.1074 0.2646 0.3272
1.0 0.1667 0.3297 0.3228
1.2 0.2390 0.3937 0.3164

1.4 0.3252 0.4559 0.3074

1.6 0.4225 0.5163 0.2962
1.8 0.5310 0.5743 0.2826
2.0 0.6502 0.6297 0.2667
2.2 0.7823 0.6809 0.2481
2.4 0.9240 0.7282 0.2278

2.6 1.0741 0.7716 0.2063

2.8 1.2319 0.8109 0.1840
3.0 1.3969 0.8459 0.1614
3.2 1.5697 0.8756 0.1392
3.4 1.7478 0.9010 0.1181
3.6 1.9302 0.9226 0.0984

3.8 2.1164 0.9407 0.0804

4.0 2.3058 0.9555 0.0643
4.2 2.4983 0.9666 0.0508
4.4 2.6924 0.9758 0.0391
4.6 2.8883 0.9826 0.0296

4.8 3.0853 0.9878 0.0219

5.0 3.2833 0.9915 0.0160
5.2 3.4819 0.9942 0.0114
5.4 3.6809 0.9961 0.0080
5.6 3.8803 0.9975 0.0055
5.8 4.0799 0.9984 0.0037
6.0 4.2796 0.9990 0.0024
6.2 4.4795 0.9994 0.0016
6.4 4.6794 0.9996 0.0010
6.6 4.8793 0.9998 0.0006
6.8 5.0793 0.9999 0.0004
7.0 5.2792 0.9999 0.0002
 EXAMPLE 10.3
Using the information in Table 10.1 plot streamlines and d99 in the Blasius boundary layer for a
1.0 m/s airflow over a 3.0-m-long surface.

Solution
Start with (10.19) and insert (10.26) to reach:
pffiffiffiffiffiffiffiffiffi  pffiffiffiffiffiffiffiffiffiffiffi
j ¼ nUxf y U=nx :

The goal is to use this formula and Table 10.1 to plot an x-y curve that represents j ¼ constant. To
get started denote the first two entries on the ith row of Table 10.1 by hi and fi, and look for
an algebraic parameterization of the streamline’s coordinates at discrete locations: xi ¼ x(hi, fi) and
yi ¼ y(hi, fi). The first parameterization can be found directly from the above equation:
pffiffiffiffiffiffiffiffiffiffi  
j ¼ nUxi fi or xi ¼ j2 fi2 nU :

The second comes from (10.26):

pffiffiffiffiffiffiffiffiffiffiffiffi  
hi ¼ yi U=nxi or yi ¼ hi j fi U :

Thus, once a value of j is selected, xi-yi coordinate pairs on this streamline can be obtained by
evaluating the equations for xi and yi using the hi and fi entries in Table 10.1. For the conditions
given, such a streamline plot is provided in Figure 10.7. Here, a few additional h and f values from
high in the boundary (h > 7.0) were needed to plot streamlines starting from x ¼ 0. And, the darker
line is d99 from (10.30a). For x < 0, the streamlines are straight and horizontal.
This figure shows several important phenomena. First, even at this modest size and flow speed
the boundary layer’s thickness (centimeters) is much less than the corresponding development
length (meters). Second, there is a kink in the streamlines at x ¼ 0. This occurs because the
boundary-layer equations are parabolic so the plate has no upstream influence. This kink would be
absent if the full equations of fluid motion were used near the plate’s leading edge. And third,
streamlines that originate in the outer irrotational flow continually enter the boundary layer with
increasing downstream distance.

0.04

0.03 ψ = 0.02

δ99
y (m)

0.02 ψ = 0.01
ψ = 0.004
0.01 ψ = 0.001

0
0.00 0.50 1.00 1.50 2.00 2.50 3.00
x (m)
FIGURE 10.7 Blasius boundary-layer streamlines and 99% thickness for a 1.0 m/s air flow over a 3.0-m-long
surface. At x ¼ 3 m, the Reynolds number based on downstream distance, Rex, is 200,000. The lighter curves are
streamlines from the solution of (10.27), and the stream function values are in m2/s. The heavier curve is the overall
boundary-layer thickness, d99 from (10.30a); it reaches 3.3 cm at x ¼ 3.0 m.
 EXAMPLE 10.4
Determine a formula for the surface-normal velocity v in a Falkner-Skan boundary-layer flow. Is
v positive or negative when the exterior flow is accelerating and n is positive?

Solution
Start with (10.34) and differentiate:
vj v hpffiffiffiffiffiffiffiffiffiffiffiffiffi  pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi i n þ 1 pffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffi n  1 df
v ¼  ¼  naxnþ1 f y axn1 =n ¼  naxn1 f þ naxn1 h :
vx vx 2 2 dh
Divide this result by U(x) ¼ axn and simplify to find:
v 1 nþ1 n  1 df
¼  1=2 fþ h :
Ue ðxÞ Rex 2 2 dh

The plotted results in Figure 10.8 show that h, f, and df/dh are all positive for all n in the range of
interest. Thus, this equation suggests that v is negative when n is positive, and this is certainly so
when n  1. Thus, an accelerating outer flow is mildly drawn towards the wall, and this prevents
boundary-layer separation. However, the opposite is also true; a decelerating flow (n < 0) will
produce a mild positive wall normal velocity (recall that f < h(df/dh) in the Blasius boundary layer
where n ¼ 0). Thus, a decelerating outer flow is pushed away from the wall, and this may lead to
boundary-layer separation.

10.5 VON KARMAN MOMENTUM INTEGRAL EQUATION

Exact solutions of the boundary-layer equations are possible only in simple cases. In more
complicated problems, approximate methods satisfy only an integral of the boundary-layer
equations across the layer thickness. When this integration is performed, the resulting ordi-
nary differential equation involves the boundary layer’s displacement and momentum thick-
nesses, and its wall shear stress. This simple differential equation was derived by von
Karman in 1921 and applied to several situations by Pohlhausen (1921).
The common emphasis of an integral formulation is to obtain critical information with
minimum effort. The important results of boundary-layer calculations are the wall shear
stress, displacement thickness, momentum thickness, and separation point (when one exists).
The von Karman boundary-layer momentum integral equation explicitly links the first three
of these, and can be used to estimate, or at least determine the existence of, the fourth. The
starting points are (7.2) and (10.9), with the pressure gradient specified in terms of Ue(x)
from (10.11) and the shear stress s ¼ m(vu/vy):
vu vu dUe 1 vs
u þv ¼ Ue þ : (10.37)
vx vy dx r vy
 10.5 VON KARMAN MOMENTUM INTEGRAL EQUATION 487
Multiply (7.2) by u and add it to the left side of this equation:


vu vv vu vu vðu2 Þ vðvuÞ dUe 1 vs
u þ þu þv ¼ þ ¼ Ue þ : (10.38)
vx vy vx vy vx vy dx r vy
Move the term involving Ue to the other side of the last equality, and integrate (7.2) and
(10.38) from y ¼ 0 where u ¼ v ¼ 0 to y ¼ N where u ¼ Ue and v ¼ vN:
ZN
 ZN ZN
vu vv vu vv N
þ dy ¼ 0 / dy ¼  dy ¼ ½v0 ¼ vN; (10.39)
vx vy vx vy
0 0 0

ZN
 ZN ZN
2 
vðu2 Þ vðvuÞ dUe 1 vs vðu Þ dUe 1
þ  Ue dy ¼ þ dy /  Ue dy þ Ue vN ¼  sw ;
vx vy dx r vy vx dx r
0 0 0

(10.40)
where sw is the shear stress at y ¼ 0 and s ¼ 0 at y ¼ N. Use the final form of (10.39) to elim-
inate vN from (10.40), and exchange the order of integration and differentiation in the first
term of (10.40):
ZN ZN ZN
d dUe vu 1
u dy  Ue
2
dy  Ue dy ¼  sw : (10.41)
dx dx vx r
0 0 0

Now, note that

0 1
ZN ZN ZN ZN
vu d d @ dUe
Ue dy ¼ Ue u dy ¼ Ue u dyA  u dy:
vx dx dx dx
0 0 0 0

and use this to rewrite the third term on the left side of (10.41) to find:
ZN ZN
d  2  dUe 1
u  Ue u dy þ ðu  Ue Þdy ¼  sw : (10.42)
dx dx r
0 0

A few final algebraic rearrangements produce:

2 3
ZN
 ZN

1 d 4 2 u u 5 dUe u
sw ¼ Ue 1 dy þ Ue 1 dy;
r dx Ue Ue dx Ue
0 0 (10.43)
1 d dUe
or sw ¼ U 2 q þ Ue d :
r dx e dx
Throughout these manipulations, Ue and dUe/dx may be moved inside or taken outside the
vertical-direction integrations because they only depend on x.
Equation (10.43) is known as the von Karman boundary-layer momentum integral equation,
and it is valid for steady laminar boundary layers and for time-averaged flow in turbulent
boundary layers.
EXAMPLE 10.5
Use the von Karman boundary-layer momentum integral equation to determine how the wall
shear stress depends on downstream distance in an accelerating flow where Ue(x) ¼ (Uo/L)x.

Solution
The given exterior flow velocity follows a power law with n ¼ 1, so the generic Falkner-Skan
boundary-layer thickness is:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffi
dðxÞ ¼ nx=Ue ðxÞ ¼ nL=Uo ¼ const:
From this, we can deduce that q and d* are constant as well since are both defined as integrals of the
velocity profile and are therefore proportional to the generic thickness, d. For example:
ZN
 ZN
u u  
qh 1 dy ¼ d f 0 ðhÞ 1  f 0 ðhÞ dh ¼ d$const:
Ue Ue
y¼0 y¼0

where h and f(h) are defined by (10.34). Now use (10.43) and Ue(x) ¼ (Uo/L)x to find:

1 d Uo2 x2 Uo x  d Uo x 2Uo2 x U2 x
sw ¼ 2
q þ d ¼ 2
q þ o2 d ;
r dx L L dx L L L
and this can be mildly simplified to:


sw 4q þ 2d x
1 2
¼ :
2
rU o L L

Thus, the skin friction increases linearly with downstream distance in this case. However, the three-
unknowns-and-one-equation problem persists since values for q and d* are needed to determine sw.
Thwaites method provides an approximate remedy for this problem.