•CHAPTER 5

•5. Internal Flows (Pipe Flow)

Objective:
This chapter is completely devoted to an
important practical fluids engineering
problem:
flow in ducts with
 various velocities,
 various fluids,
and various duct shapes.
•Correlations, those are adequate to solve most
piping problems, are discussed in this chapter.

•After completion of this chapter students will be

able to to solve :
basic piping problems given the pipe
geometry, and its added components (such
as fittings, valves, bends, and diffusers), the
desired flow rate and fluid properties, what
pressure drop is needed to drive the flow?
Or given the pressure drop available from a
pump, what flow rate will result?
•Flow Regimes
Reynolds classical experiment
•He measured the pressure drop as a function of
flow rate for water in a tube
•At low flow rates the pressure drop was directly
proportional to the flow rate(Laminar).
•As the flow rate was increased a point was
reached where the relation was no longer linear
and the ‘‘noise’’ or scatter in the data increased
considerably(Transition).

•At still higher flow rates the data became more

reproducible, but the relationship between
pressure drop and flow rate became almost
quadratic instead of linear(Turbulent).
•Flow Regimes
Reynolds classical experiment
•He measured the pressure drop as a function of
flow rate for water in a tube
•At low flow rates the pressure drop was directly
proportional to the flow rate(Laminar).
•As the flow rate was increased a point was
reached where the relation was no longer
linear and the ‘‘noise’’ or scatter in the data
increased considerably(Transition).

•At still higher flow rates the data became more

reproducible, but the relationship between
pressure drop and flow rate became almost
quadratic instead of linear(Turbulent).
•To investigate this phenomenon further, Reynolds introduced a
trace of dye into the flow

• At the low flow rates where the linear relationship was observed,
the dye was seen to remain a coherent, rather smooth thread
throughout most of the tube.

• However, where the data scatter occurred, the dye trace was
seen to be rather unstable, and it broke up after a short distance.

•At still higher flow rates, where the quadratic relationship was
observed, the dye dispersed almost immediately into a uniform
‘‘cloud’’ throughout the tube.
 Laminar flow
 Also known as
streamline flow
 Occurs when the fluid
flows in parallel
layers(lamella), with
no disruption
between the layers
 The opposite of
turbulent flow
(rough)
Laminar flow
Turbulent
 Laminar flow over a flat Flow
and horizontal surface can
be pictured as consisting of
parallel and thin layers

 Layers slide over each Laminar Flow

other, thus the name
‘streamline’ or smooth.
 The paths are regular and
there are no fluctuations
Turbulent flow
 Usually occurs when the
liquid is moving fast
 The flow is ‘chaotic’ and
there are irregular
fluctuations
 Includes
 rapid variation of
pressure and velocity of
the fluid
 Good way to transfer
thermal energy
Reynold’s number
 A dimensionless number in fluid mechanics
 It quantifies the relative importance of Inertial force and
Viscous force for a given flow conditions.
 Arises when performing analysis of fluid dynamics
 Can be used to determine dynamic similitude in such cases.
Concept used in the testing of models, e.g. testing miniature
airplanes/submarines
Reynold’s number
for flow in a pipe
 ρ is the density of the fluid
 V is the mean fluid velocity
 D is the diameter Inertial force
 Q is the volumetric flow rate

 μ is the dynamic viscosity of the fluid Viscous force

 v is the kinematic velocity of the fluid
 A is the pipe cross-sectional area.
Reynold’s number
0 < Re < 1: highly viscous laminar “creeping” motion
 1 < Re < 100: laminar, strong Reynolds-number
dependence
100 < Re < 103 : laminar, boundary-layer theory useful
 103 < Re < 104 : transition to turbulence
104 < Re < 106 : turbulent, moderate Reynolds-number
dependence
106 < Re : turbulent, slight Reynolds-number dependence
The critical Reynolds number, Recr , below which the flow is
laminar and above which the flow is turbulent for pipe-flow
transition is now taken to be

•Laminar when Re < 2300

•Turbulent when Re > 2300
Flow development in a conduit
•For internal flow in the figure below(incompressible fluid)
•There is an entrance region where a nearly inviscid upstream.
• Flow converges and enters the tube.
•Viscous boundary layers grow downstream, retarding the axial
flow u(r, x) at the wall and thereby accelerating the center-core
flow to maintain the incompressible continuity requirement
•At a finite distance from the
entrance, the boundary
layers merge and the inviscid
core disappear.

•The tube flow is then entirely

viscous, and the axial velocity
adjusts slightly further until at
x = Le it no longer changes
with x and is said to be fully
developed u≈u(r) only.

•Downstream of x = Le the
velocity profile is constant,
the wall shear is constant,
and the pressure drops
Fig. Developing velocity profiles
linearly with x, for either
and pressure changes in the
laminar or turbulent flow.
entrance of a duct flow
•Dimensional analysis shows that the Reynolds number is the only
parameter affecting entrance length. If
Le = f(d, V, 𝜇 , 𝜌) V= Q/A

then

•For laminar flow, the accepted correlation is

•The maximum laminar entrance length, at Recr = 2300, is Le =

138d which is the longest development length possible.

•In turbulent flow the boundary layers grow faster, and Le is

relatively shorter, according to the approximation for smooth
walls
Conservation of Energy
Consider flow in a pipe, driven by pressure or gravity or
both
Conservation of Energy
 Kinetic, potential, and thermal energy

p1 V12 p2 V22
 1  z1  h p    2  z2  ht  hL
 2g  2g

hp = head supplied by a pump

ht = head given to a turbine

hL or hf = mechanical energy converted to thermal(frictional loss)

downstream from cross section 1!

Cross section 2 is ____________

Point to point or control volume? irreversible

Why ? _____________________________________
V is average velocity, kinetic energy
V 2
 EGL (or TEL) and HGL
p V2
EGL   z
 2g
pressure elevation
head (w.r.t. velocity head (w.r.t.
reference pressure) head datum)

downward (in
•The energy grade line must always slope ___________
direction of flow) unless energy is added (pump)
•The decrease in total energy represents the head loss or
energy dissipation per unit weight
•EGL and HGL are coincident and lie at the free surface for
water at rest (reservoir)
Energy equation
V2 Energy Grade Line
velocity head  Hydraulic G L
2g

static head pressure

p
Why is static head 
head important?

z elevation
pump
z=0 datum
p1 V12 p2 V22
 1  z1  hp   2  z2  ht  h f
 2g  2g
 Flow in a Circular Pipe
Consider flow in a full pipe, driven by pressure or gravity or
both

Making a control-volume
analysis of the flow between
sections 1 and 2
Q1=Q2= const, or

since the pipe is of constant

area. The steady-flow energy
equation reduces to (no shaft
work)
Fig(B). Control volume of steady,
fully developed flow between two
sections in an inclined pipe =
•Now assume that the flow is fully developed and correct later for
entrance effects.
• Then the kinetic energy correction factor α1 = α2 and since V1=V2
from continuity equation, the energy equation reduces to

•The pipe-head loss equals the change in the sum of pressure

and gravity head, i.e., the change in height of the hydraulic
grade line (HGL).

•Since the velocity head is constant through the pipe, hf also

equals the height change of the energy grade line (EGL)

•Finally apply the momentum relation to the control volume,

accounting for applied forces due to pressure, gravity, and shear
 •This equation relates hf to the wall shear stress

………….A

•we can assume that

Where ᵋ is the wall roughness height

•Then the dimensional analysis tell us that

…………….B

•The dimensionless parameter f is called the Darcy friction factor.

Combining Eqs. (A) and (B), we obtain the desired expression for
finding pipe head loss
…………….(C)
•Our only remaining problem/ is to find the form of the function F in Eq.
(B) and plot it in the Moody chart of Fig. below.

Fig. the Moody chart for pipe friction with smooth and
rough walls
 Equations of Motion
•For either laminar or turbulent flow, the
continuity equation in cylindrical coordinates is
given by (App. D, F. White)

For no swirl or circumferential variation

and fully developed flow: u= u(r) there fore

or

•But at the wall, r = R, Vr = 0 (no slip); therefore Vr = 0

everywhere.
•Thus in a fully developed flow there is only one velocity
component, u = u(r).
•The momentum differential equation(equation of motion) in
cylindrical coordinates reduces to

•Left hand side vanishes because u = u(r) only

•gx = g sin ϕ
∴ ……….(M)

•Since the left-hand side varies only with r and the right-hand
side varies only with x, it follows that both sides must be equal
to the same constant.

………(E)
•Thus the shear varies linearly from the
centerline to the wall, for either laminar or
turbulent flow.
•This is also shown in Fig. (B). At r = R, we
have the wall shear

• We can now complete our study of pipe flow by applying

either laminar or turbulent assumptions.

•Laminar Flow Solution

Note in Eq. (E) that the HGL slope d(p + 𝜌gz)/dx is negative
because both pressure and height drop with x.
 •For laminar flow, 𝜏 = 𝜇du/dr, which we substitute in Eq. (E)

………..(G)

Integrate once
………..(H)

•The constant C1 is evaluated from the no-slip condition at the

wall: u = 0 at r = R
………..(I)

•Introduce into Eq. (I) to obtain the exact solution for laminar
fully developed pipe flow

………..(J)
 •The laminar flow profile is thus a parabolic which is falling to
zero at the wall and reaching a maximum at the axis.

……..(K)

•It resembles a sketch of u(r) in Fig. (B).

•Other pipe-flow results follow immediately from Eq. (J). The

volume flow is
……..(L)

……..(N)
•Thus the average velocity in laminar flow is one–half the
maximum velocity (from K and N)

……..(O)
 •For horizontal tube (∆z = 0), Eq. (N) is of the predicted by
Hagen’s experiment,

……..(P)

•The wall shear is computed from the wall velocity gradient

……..(P)

•This gives an exact theory for laminar Darcy friction factor

……..(Q)

……..(R)
•The laminar head loss follows from Eq. (C)

……..(S)

•It can be clearly seen that head loss is proportional to V.

Turbulent Flow Solution
•A similar analysis is not readily available for turbulent flow.
•Colebrook equation, shown below, provides an excellent
representation for the variation of the Darcy friction factor in
the turbulent flow regime.
•Note that the equation depends on both the pipe Reynolds
number and the roughness ratio and cannot be expressed
explicitly for f.

•Eq. (5.35) was plotted in 1944 by Moody into what is now

called the Moody chart for pipe friction .
Fig (E). the Moody chart for pipe friction with smooth and rough walls
Notice that the roughness curves are nearly horizontal in the fully rough
regime to the right of the dashed line.
It is accurate to ±5 percent for design calculations over the full
range shown in Fig.
It can be used for circular and noncircular pipe flows and for
open-channel flows.
A good approximate equation for the turbulent region of the
Moody chart is given by Haaland’s equation:
•Table 5.1 Recommended Roughness Values for Commercial Ducts
 Revision
• Flow Regimes
Laminar
Transition Re, cr = 2300
Turbulent
• Flow development
– Fully developed flow, u = u(r) only
– For laminar flow,
– For turbulent flow,
• Flow in a circular pipe
the Darcy-Weisbach equation,
For laminar flow,
For turbulent flow,
01/24/13 25
 Effect of Rough walls

•The effect is negligible for laminar pipe flow, and all the
laminar formulas derived in this section are valid for rough
walls also.

•But turbulent flow is strongly affected by roughness.

•For a fully rough pipe the friction factor is given by:

 Three Types of Pipe-Flow Problems
1. Given d, L, and V or Q, 𝜇, 𝜌, and g, compute the head loss hf
(head-loss problem).
2. Given d, L, hf, and g, compute the velocity V or flow rate Q (flow-
rate problem).
3. Given Q, L, hf, 𝜇, 𝜌, and g, compute the diameter d of the pipe
(sizing problem).
•Only problem 1 is well suited to the Moody chart. We have to
iterate to compute velocity or diameter because both d and V are
contained in the ordinate and the abscissa of the Moody chart.
Problem #2

Even though velocity (or flow rate) appears in both the ordinate and the
abscissa on the Moody chart, iteration for turbulent flow is nevertheless
quite fast, because f varies so slowly with Red. Alternately we could
change the scaling variables to (𝜌, 𝜇, and d) and thus arrive at
dimensionless head loss versus dimensionless velocity. The result is

A formula valid for all turbulent pipe flows is found by simply

rewriting the Colebrook interpolation, Eq. (5.35), in the form of Eq.
(5.38):
•Example
Problem #3

A similar difficulty arises for problems involving unknown pipe

difficulty, except a closed form, analytical solution is not available.
Again, a trial and error solution is appropriate for use to obtain the
solution and the problem can again include losses due to KE, PE,
and piping components with no additional difficulty.
 Flow in non Circular Conduits
•For flow in non-circular ducts or ducts for which the flow does
not fill the entire cross-section, we can define the hydraulic
diameter Dh as

Where A = cross-sectional area of actual flow,

P = wetted perimeter, i.e. the perimeter on which viscous
shear acts

•With this definition, all previous equations for the Reynolds

number Re friction factor f and head loss hf are valid as
previously defined and can be used on circular and non-
circular flow cross sections.
 Minor Losses
•In addition to frictional losses for a
length L of pipe, we must also consider
losses due to various fittings (valves,
unions, elbows, tees, etc.). These
losses are expressed as

where
•hm = the equivalent head loss across
the fitting or flow component
•V = average flow velocity for the pipe
size of the fitting
•Ki = the minor loss coefficient for given
flow component; valve, union, etc.
•Table 5.2 shows minor loss K values for several common
types of valves, fully open, and for elbows and tees.

Table 5.2 Minor loss coefficient for common valves and piping components
 Fig. Butterfly
pump

Figure 5.5 Average loss coefficients for partially open valves.

•Note that: the K valves shown here are for the indicated
fractional opening. Also, fully open values may not be consistent
with values indicated in Table 5.2 for fully open valves or for the
valve of a particular manufacturer. In general, use specific
manufacturer’s data when available.
•Exit losses are K = 1.0 for all shapes of exit (reentrant, sharp,
beveled, or rounded).

•For exit losses K ≈ 1 for all submerged exits, e.g., fluid

discharged into a tank at a level below the fluid surface.

•Also, for an open pipe discharge to the atmosphere, there is

no loss coefficient when the energy equation is written only to
the end of the pipe.
Figure 5.6 Entrance and exit loss coefficients (a) reentrant inlets; (b) rounded and
beveled inlets
Figure 5.7 Sudden contraction and expansion losses
•These equations would be appropriate for a single pipe size (with
average velocity, V).
•For multiple pipe/duct sizes, this term must be repeated for each
pipe size; and the energy equation must still be used to determine the
total head loss and pressure drop from all possible contributions.
Example

Fig. E6.16
To solve basic pipe flow pressure drop problem, use the following
procedure:

1. Use known flow rate to determine Reynolds number.

2. Identify whether flow is laminar or turbulent.
3. Use appropriate expression to find friction factor (with ε/D if
necessary).
4. Use definition of hf to determine friction head loss.
5. Tabulate and sum minor loss coefficients for piping components.
6a. Use general energy equation to determine total pressure drop, or
6b. Determine pump head requirements as appropriate
7. Determine pump power and motor size if required.
 THE END!

THANK YOU!!!