3.

Course Number: CE 365K

Course Title: Hydraulic Engineering Design
Course Instructor: R.J. Charbeneau

Subject: Open Channel Hydraulics

Topics Covered:
8.

Open Channel Flow and Manning Equation

9.

Energy, Specific Energy, and Gradually Varied Flow

10. Momentum (Hydraulic Jump)

11. Computation: Direct Step Method and Channel Transitions
12. Application of HEC-RAS
13. Design of Stable Channels

3.2

Topic 8: Open Channel Flow

Geomorphology of Natural Channels:
Geomorphology of natural channels concerns their shape and
structure. Natural channels are of irregular shape, varying from
approximately parabolic to approximately trapezoidal (Chow, 1959).

Trapezoidal fit

3.3

Channel Geometry Characteristics

Depth, y
Area, A
Wetted perimeter, P
Top width, T

Hydraulic Radius, Rh = Area / Wetted perimeter

Hydraulic Depth, Dh = Area / Top width

3.4

Trapezoidal Channel

3.5

Parabolic Channel
z

If 0 < (4 a y)

1/2

< 1 Then

3.6

Example #12: Parabolic Channel

A grassy swale with parabolic cross-section shape has
top width T = 6 m when depth y = 0.6 m while
carrying stormwater runoff. What is the hydraulic
radius?

3.7

Flow in Open Channels: Manning

Equation
Mannings equation is used to relate the average
channel (conduit) velocity to energy loss, Sf = hf/L.
Manning equation (metric units: m, s)

UNITS ?!!? Does n have units? Tabulated values?

3.8

Manning Equation (Cont.)

To change to US Customary
units multiply by
General case

= 1 (metric) or 1.486 (English)

3.9

Channel Conveyance, K
For Mannings equation, K combines roughness
and geometric characteristics of the channel

K = (/n) A Rh2/3
Mannings equation: Q = K Sf1/2

3.10

Roughness and Mannings n

Equivalence between roughness size (k) and Mannings n:
n = 0.034 k 1/6 (k in ft)
Strickler (1923)

Examples

k (cm)

Concrete (finished)

0.012

0.06

Asphalt

0.016

0.3

Earth channel (gravel)

0.025

Natural channel (clean)

0.030

15

Floodplain (light brush)

0.050

300

* Compare with Mannings n for sheet flow

3.11

Grassed Channels

Vegetal Retardance Class (Table 8.4)

SCS (after Chow, 1959)

English Units (ft, s)

, V Rh
Figure 8.11, page 311

3.12

Example #13: Estimate the channel

discharge capacity for So = 0.008

Drainage
channel
from Far
West Pond

3.13

Example #13: Qmax (cfs) =

1. For grass channels, use Slide 3.11. Guess an
initial value n = 0.05
2. Geometry:

Bottom width, b = 5 ft (~)

Side slope (z:1), z = 2

Q = 240 ft3/s

Maximum depth, y = 4 ft
--------------------------------------Hydraulic radius, Rh = 2.3 ft
Velocity, V = 4.6 ft/s

V Rh = 10.6 ft2/s

3.14

Example #13 (Cont.)

V Rh > 10 ft2/s n
= 0.03 (slide 3.11)

Qmax =
400 cfs

3.15

Normal Depth
Normal depth is the depth of uniform flow in an
prismatic open channel. Since the flow is
uniform, the depth and discharge are related
through Mannings equation with Sf = So.

Given Q, n, A(y), Rh(y) and So: solve for yn

3.16

Waves (Small Disturbances) in a

Moving Stream

Wave (disturbance) can move upstream if

Froude Number

3.17

Critical Depth Froude number

Critical flow occurs when the velocity of water is the
same as the speed at which disturbances of the free
surface will move through shallow water. The speed or
celerity of disturbances in shallow water is given by c =
(g Dh)1/2, where Dh is the hydraulic depth. Critical flow
occurs when v = c, or more generally

Importantly, critical depth is independent of the

channel slope.

3.18

Topic 9: Energy, Specific Energy,

and Gradually Varied Flow
1D Energy Equation:
Closed Conduit:

Open Channel Flow:

zB z hereafter

3.19

Energy in Open Channels

Friction Slope:

=H

Sf = - dH/dx
Channel Slope:

Kinetic
Energy
Correction
Factor

So = - dz/dx

Specific
Energy

3.20

What do we do with ?
For simple channels assume = 1
For complex channels (main channel plus left
and right-bank floodplains), velocity variation
at a given station can be significant, and
should be calculated and used in a 1D energy
equation (HEC-RAS does this automatically!)

10

3.21

Kinetic Energy Coefficient,

1
Left Overbank

Main
Channel

A1, P1, n1

Right Overbank

A3, P3, n3

A2, P2, n2

3.22

Specific Energy, E
Hydraulic energy head measured with respect to
the local channel bottom, as a function of depth y
y
For fixed Q
Alternate depths

11

3.23

Specific Energy at Critical Flow

Rectangular channel: Dh = y
E = V2/2g + y = [(1/2) (V2/gy) + 1] y
= (1/2 + 1) y

Froude Number

For critical flow (in a rectangular channel):

y = (2/3) E
V2/2g = (1/3) E

3.24

Energy Equation
E2 + z2 = E1 + z1 + hL(2 1)
E = Specific Energy = y + V2/2g

Head Loss:
Major Losses friction losses along channel
Minor Losses channel expansion and contraction

12

3.25

Friction Losses in Open Channel

Flow:
Slope of the EGL: Sf = hf / L
Mannings equation: Q = K Sf1/2
Bed-friction head loss: hf = (Q/K)2 L

3.26

Minor (Expansion and Contraction)

Losses
Energy losses at channel expansions and contractions

Default values:
Channel Contraction Channel Expansion -

C = 0.1
C = 0.3

Abrupt Expansion: (C = 1)

13

3.27

Gradually Varied Flow Profiles

Physical laws governing the head variation in open
channel flow

1)

Gravity (So) is the driving force for flow

2)

If So = Sf then dE/dx = 0 and flow is uniform (normal depth)

3)

Gravity (So) is balanced by friction resistance (Sf ) and

longitudinal adjustment in specific energy (dE/dx)

4)

Adjustments in specific energy are constrained through

specific energy diagram

3.28

Gradually Varied Flow: Mild Slope

(yn > yc)
1. Point 1 (M1 Curve): y > yn
Sf < So dE/dx > 0
depth increases downstream
2. Point 2 (M2 Curve): y < yn
Sf > So dE/dx < 0
depth decreases downstream

3. Point 3 (M3 Curve): y < yc <

yn Sf > So dE/dx < 0
depth increases
downstream

yn

1
2

yc

For fixed Q

14

3.29

Mild Slope

Note that for the M1 and M2 curves, the depth

approaches normal depth in the direction of flow
computation for subcritical flow.

3.30

Gradually Varied Flow: Steep

Slope (yn < yc)
1. Point 1 (S1 Curve): y > yc >
yn Sf < So dE/dx > 0
depth increases
downstream
2. Point 2 (S2 Curve): y > yn
Sf < So dE/dx > 0
depth decreases downstream
3. Point 3 (S3 Curve): y < yn
Sf > So dE/dx < 0
depth increases downstream

y
For fixed Q
1

yc
yn

15

3.31

Steep Slope

Note that for curves S2 and S3 the depth

approaches normal depth in the direction for flow
computation for supercritical flow

3.32

Horizontal Slope (So = 0)

yc

16

3.33

Topic 10: Momentum and Hydraulic

Jump
yp = depth to pressure center
(X-section centroid)

FB

v1

v2

Momentum Equation: F = Q (V2 V1)

yp1 A1 yp2 A2 FB = Q2 (1/A2 1/A1)
Write as:

M1 = M2 + FB/

Specific Force: M = yp A + Q2/(g A)

3.34

Hydraulic Jump
y

2
Sequent depths
(Conjugate)

1
Specific force, M

17

3.35

Energy Loss and Length

y2

yj
y1

Lj

Energy Loss:

Rectangular
Channel

Exercise: convince yourself that

this is the same as E = E1 E2.

Lj

Jump Length:
Lj = 6.9 yj = 6.9 (y2 y1)

yj

3.36

Example #14:
Normal depth
downstream in a
trapezoidal channel
is 1.795 m when
the discharge is 15
m3/s. What is the
upstream sequent
depth?
Answer:
1.065 m
How do you find
the energy loss?
E = 0.14 m
Power = 20.6 kW

18

3.37

Topic 11: Direct Step Method and

Channel Transitions
Returning to the Energy Equation
Step calculation

Solve for x:

3.38

Direct Step Method: Approach

In this case you select a sequence of depths and

solve for the distance between them: y1, y2, y3,
With known y: E1, E2, Q2/2gA2, etc. are calculated
The average friction slope is calculated from

19

3.39

Example #15: Channel Transition

What is the headwater upstream of a control section in a
downstream box culvert: Q = 10 m3/s; channel n = 0.025
Headwater = ?

z1 = 1

Control
section
Fr = 1

5m

b2 = 2.5 m

b1 = 2 m

upstream

downstream

Control Section:
V = (g Dh)0.5 = (g y)0.5 Q = (g y)0.5 (b y)
y = [Q / b g0.5]2/3 = 1.37 m

3.40

Example #15 (Cont.): Specific Energy

A 1.37 m depth in
the upstream
channel (Section
1, blue)
corresponds to
specific energy E
= 1.55 m,
compared with E
= 2.05 m in the
control section.
The upstream E2
must exceed 2.05
m, which requires
an increase in
depth.

20

3.41

Example #15 (Cont.): Direct Step

Method
Assume a 5-m long
transition section. Adjust
depth until x = 5 m
y2 = 2.051 m
Resulting upstream
specific energy is E2 =
2.110 m.
Contraction losses at
entrance ~ 6 cm
Friction losses in
transition section ~ 1 cm

3.42

Example #16: Specific Energy

and Channel Transitions
Trapezoidal channel with b = 8 ft, z = 2, n =
0.030. Normal depth occurs upstream and
downstream.
Rectangular culvert (b = 5 ft, n = 0.012) added
with concrete apron extending 10 feet downstream
from culvert outlet.
Develop flow profile, especially downstream of
the culvert, for Q = 250 cfs.

21

3.43

Ex. #16 (Cont.): Normal and Critical

Depth in Main Channel
Upstream and
Downstream
channel have
yn = 3.13 ft
yc = 2.51 ft
yn > yc
Mild Slope
Downstream
Control

3.44

Ex. # 16 (Cont.): Normal and Critical

Depth in Culvert

yn = 4.22 ft
yc = 4.26 ft

yn < yc Steep Slope Upstream control

22

3.45

Ex. #16 (Cont.): Specific Energy

Diagram
Normal depth:
E = 3.13 ft
Minimum specific
energy in culvert:
E = 6.40 ft
To enter the
culvert from
upstream, specific
energy increased
through M1 curve

Normal depth
upstream and
downstream

Culvert
1

Channel

(Culvert as choke)

3.46

Ex. #16 (Cont.) Upstream Transition

and Entry to Culvert
Trapezoidal Channel
M1 Curve

Headwater = 6.51 ft

Culvert

M1 curve
Control in culvert
1

Channel

23

3.47

Ex. #16 (Cont.): Control in Culvert

yn < yc Steep Slope Control at Culvert Entrance

S2 curve in culvert;
normal depth reached
before culvert end

3.48

Ex. #16 (Cont.): Possible Depths at

Concrete Apron

Set y2 = yn in culvert and adjust y1 = ya until x = 10 ft for depth at end

of concrete apron

Possible depths for ya = 1.389 ft or 5.74 ft

24

3.49

Ex. #16 (Cont.): Which Depth for ya?

How do you determine which depth at the end of the culvert apron is
correct?
The flow profile downstream of the apron must follow an M-curve
(since yc < yn in the downstream channel) and the depth must end in
normal depth.

Normal depth is yn = 3.13 ft

while ya = 1.39 ft or 5.74 ft.
Cannot reach normal depth
from y = 5.74 ft following an
M1 curve.
Correct depth is ya = 1.39 ft,
which will follow an M3 curve
leading to a hydraulic jump,
which must end at normal
depth (no M-curves approach
normal depth in the
downstream direction).

3.50

Ex. #16 (Cont.): Transition to Apron

(Pt. 5)
End of culvert apron at
Pt. 5. M3 Curve leading
to Hydraulic Jump, which
in turn exits at normal
depth for the
downstream channel

Culvert
2

M1 curve
3

Control in culvert
4

1
5

S2 curve
Channel

25

3.51

Ex. #16 (Cont.): M3 Profile to

Hydraulic Jump
How far does the flow profile follow the M3 curve
downstream of the concrete apron?
From this supercritical flow profile, the flow must
reach subcritical flow at normal depth through a
hydraulic jump.
The flow profile must follow the M3 curve until the
depth is appropriate for the jump to reach yn (this is
the sequent depth to yn).

3.52

Ex. #16 (Cont.): Estimation of

Sequent Depth
Plot y versus specific
force values. For y =
3.13 ft, the specific
force (M) value is
103.1 ft3. Look for
the sequent depth on
the supercritical limb
of the specific-force
curve, to find ys =
1.97 feet.
Conclusion: the M3
curve is followed
from a depth of ya =
1.39 ft to a depth of
ys = 1.97 ft, which
marks the entry to a
hydraulic jump.

Sequent
Depths

26

3.53

Ex. #16 (Cont.): Distance along

M3 Curve and Jump Length

The Direct Step method may

be used to estimate the
distance along the M3-curve
from the culvert apron to the
entrance of the hydraulic
jump.
Taking a single step, this
distance is estimated to be 44
feet.
The jump length is calculated
as
Lj = 6.9 (3.13 1.97) = 8 ft

3.54

Ex. #16 (Cont.): Transition

Example
1 Upstream normal
2 Upstream culvert
3 Culvert entrance
(control)

Culvert
2

4 Culvert exit
5 Culvert apron
6 Jump upstream
7 Jump downstream
(normal depth)

M1 curve
3
1

Jump

Control in culvert
4

M3 curve

S2 curve
Channel

27

3.55

Ex. #16 (Cont.): Hydraulic Profile of

Channel Transition

3.56

Ex. #16 (Cont.): Energy Loss through

Jump
E2 + z2 = E1 + z1 + hL(2 1)
hL = Eu Ed + zu zd = Eu Ed + So x
hL = 3.724 3.62 + 0.005 x 8 = 0.144 ft
Very weak hydraulic jump

28

3.57

Ex. #16 (Cont.): Discussion

Tired yet??
Advantages of hand/spreadsheet calculation
include control of each step in calculations
Disadvantages include 1) tedious, and 2)
requires some level of expert knowledge
Alternatives? Computer application using
HEC-RAS (River Analysis System)

3.58

Topic 12: Solve Energy (and

Momentum) Equations Using HEC-RAS
E2 + z2 = E1 + z1 + hL(2 1)
E = Specific Energy = y + v2/2g

Head Loss:
Major Losses friction losses along channel
Minor Losses channel expansion and contraction

29

3.59

Computation Problem
For subcritical flow, compute from downstream to upstream.
Discharge may vary from station to station, but are assumed known.
The depth, area, etc. at station 1 (downstream) are known.

Energy Equation:

WS = Water
Surface = y + z

Unknowns at Station 2: WS, A, K, (Rh)

This is
what
HEC-RAS
does.

3.60

Compound Channel Section

For one-dimensional
flow modeling, the
slope of the EGL is
uniform across the
channel section

1
Left Overbank

2
Main
Channel

A1, P1, n1

3
Right Overbank

A3, P3, n3

A2, P2, n2

Thus

30

3.61

Example # 17: Same problem

(Ex. #16) using HEC-RAS

3.62

Create a Project
Enter Reach on Geometric Data
Enter X-Section information for stations
Enter Discharge and Boundary Conditions
Compute

Ex. #17 (Cont.): Simple Run on Straight

Channel
Entry Data:
Q = 250 cfs
BC: Normal depth
downstream with So
= 0.005

Conclusion: HEC-RAS
correctly computes
normal depth

yn

31

3.63

Ex. #17 (Cont.): Add X-Section Data

for Culvert

3.64

Ex. #17 (Cont.): Compute for

channel with culvert section
Headwater =
6.51 ft
Tailwater =
3.13 ft

Depth = 6.51 ft

Conclusion:
Headwater
correctly
calculated but
tailwater
uniform at
normal depth
need more
X-sections

32

3.65

Ex. #17 (Cont.): Add X-sections and

Compute
Need to change computation method to mixed flow

3.66

Ex. #17 (Cont.): Results - 1

6.51
3.13

4.26
1.77

1.34

4.28

Culvert

33

3.67

Ex. #17 (Cont.): Results - 2

Bed Shear Stress is Important in assessing
channel stability to erosion

End of
Jump

High Erosion Potential

Culvert

3.68

Ex. #17 (Cont.): Discussion

Hand/Spreadsheet calculation and HECRAS calculation give equivalent results
Both are a pain (1st for computation and
2nd for set-up)
Life as a Hydraulic Engineer is a pain
HEC-RAS CAN SIMULATE FLOWS IN
COMPLEX CHANNELS THAT CONNOT
BE ADDRESSED THROUGH SIMPLE
ALTERNATIVES

34

3.69

Topic 13: Design of Stable

Channels
A stable channel is an unlined channel that will carry
water with banks and bed that are not scoured
objectionably, and within which objectionable deposition
of sediment will not occur (Lane, 1955).
Objective: Design stable channels with earth, grass and
riprap channel lining under design flow conditions

3.70

(Two) Methods of Approach

Maximum Permissible Velocity maximum
mean velocity of a channel that will not cause
erosion of the channel boundary.
Critical Shear Stress critical value of the bed
and side channel shear stress at which sediment
will initiate movement. This is the condition of
incipient motion. Following the work of Lane
(1955), this latter method is recommended for
design of erodible channels, though both
methods are still used.

35

3.71

Bed Shear Stress

Force balance in downstream direction:

p1 w y p2 w y + w g w y L sin() w L = 0

= w g y sin() = w y So

3.72

Bed Shear Stress Balance Between

Gravity and Bed Shear Forces
o P L = w A L sin()

Bed o
shear
stress

o = w Rh S

Wetted
perimeter

w A L sin()
Gravity force

36

3.73

Calculation of Local Bed Shear Stress

Uniform Flow: o = Rh S
Local Bed Shear Stress: y Sf
Manning Equation: Sf = V2 n2/(2 y4/3)
Result:

o = V2 n2/(2 y1/3)

3.74

Distribution of Tractive Force

Varies with side slope, channel width, and
location around the channel perimeter
Following criteria for trapezoidal channels
(Lane, 1955)
Channel bottom -

(o)max = w y So

Channel sides -

(o)max = 0.75 w y So
(See Figure 8.6, page 303)

37

3.75

Maximum Permissible Velocity

Unlined (Earth-lined) Channel
WB Table 9.1 (Fortier and Scobey, 1926)

For well-seasoned channels of small slopes and for depths of flow

less than 3 ft. Tractive force calculated from o = 30 n2 V2.

4.76

Example #18: Unlined Channel

An earthen channel is to be excavated in a soil that consists of colloidal
graded silts to cobbles. The channel is trapezoidal with side slope 2:1 and
bottom slope 0.0016. If the design discharge is 400 cfs, determine the size
for an unlined channel using the maximum permissible velocity.
From the table we have V = 4 ft/s and n = 0.030.
Required channel cross-section area: A = Q/V = 100 ft2
Manning equation to find hydraulic radius:

Wetted perimeter: P = A/Rh = 100/2.87 = 34.8 ft

38

3.77

Example #18 (Cont.)

Trapezoidal Channel:

Combine into the quadratic equation (eliminating b)

Solution:

3.78

Example #18 (Cont.)

For this problem

Channel base width (use y = 4 ft since y = 10 ft gives negative b)

Bed shear stress

Note: small velocity values are

required for stable unlined channels

39

3.79

Grassed Channels

Vegetal Retardance (Table 9.2)

SCS (after Chow, 1959)

, v Rh
WB, Figure 9.3, page 332

3.80

Grassed Channels
Permissible Velocities for Channels Lined with Grass (SCS, 1941)

WB, Table 9.3 (from Chow, 1959)

40

3.81

Movement of Sediment Incipient Motion

Coefficient of Friction

= Angle of Repose

Sliding Downhill F = 0

Force F necessary to move

object

3.82

Application to Incipient Motion

Shields Number, Sh
The Shields
number is
analogous to
the angle of
internal friction

41

3.83

Shields Curve Sh(Re*)

o, d, ,

= Sh

Sh depends on

Shear Reynolds

Motion

number:
No Motion

3.84

Shields Number, Sh
The Shields parameter Sh

As long as Sh < Shcritical (found from Shields Curve),

there will be no motion of bed material
Mobility of bed material depends on particle diameter
and weight (specific gravity), and on channel flow
depth (hydraulic radius) and slope

42

3.85

Design of Riprap Lining

For Re* > 10, the critical Shields number satisfies

For problems in Stormwater Management, Re* is

very large (1000s) and Shcritical = 0.06. For design
purposes, a value Shc = 0.05 or 0.04 is often
selected.
Increased bed material size (riprap) results in
decreased Sh (improved bed stability) but increases
bed roughness (Mannings n)

3.86

Design of Riprap Lining

An approach to design of stable channels using riprap
lining:
1. For specified discharge, channel slope and geometry,
select a test median particle diameter d50 (the
designation d50 means that 50 percent of the bed
material has a smaller diameter)
2. Use Stricklers equation to relate material size to
Mannings n: n = 0.034 d501/6
3. Use Mannings equation to find the depth and Rh
4. Check bed stability using Shields curve

43

3.87

Example #19: Riprap Lining

What size riprap is required for a channel carrying a
discharge Q = 2,500 ft3/s on a slope So = 0.008. The
channel has trapezoidal cross section with bottom width
b = 25 ft and side slope z:1 = 3:1?
Solution:
1. Select d50 = 3 inches (0.25 ft).
2. n = 0.034 (0.25)1/6 = 0.0270
3. For the discharge and slope, Mannings equation gives y = 5.24 ft;
Rh = 3.67 ft.
4. These values give Sh = 3.67 x 0.008/(1.65 x 0.25) = 0.071; Re* =
(g Rh So)1/2 d50/ = (32.2 x 3.67 x 0.008)1/2 x 0.25/10-5 = 24,000.
From the Shields curve, Sh > Shc, and the bed will erode.

3.88

Ex. #19 (Cont.):

Riprap Lining
2.1 Try d50 = 6 inches (0.5 ft)
2.2 n = 0.034 (0.5)1/6 = 0.0303
2.3 From Mannings equation,
y = 5.56 ft; Rh = 3.85 ft
2.4 Sh = 0.037, Re* = 50,000
OK
3.1 Try d50 = 4 inches
Channel [Link]
Check values here

Enter d50

44

3.89

Design of Channel Lining

Permissible Shear Stress for Lining
Materials (US DOT, 1967)
For Riprap Lining:
Recommended
Grading allows
voids between
larger rocks to be
filled with smaller
rocks (Simons and
Senturk, 1992)
d20 = 0.5 d50
d100 = 2 d50

45