VOLUME 1

ISSUE 2
JULY 2013

A QUARTERLY PUBL ICAT ION FOR WESTERN D AM ENGINEERS

In this issue of the Western Dam Engineering

Technical Note we present articles on the importance The Western Dam Engineering Technical
of protecting dams against the erosive action of Note is sponsored by the following
waves and also discuss how to protect conduits from agencies:
cavitation damage through air venting. This quarterly
newsletter is meant as an educational resource for S Colorado Division of Water Resources
civil engineers who practice primarily in rural areas of S Montana Department of Natural
Resources
the western United States. This publication focuses on
S Wyoming State Engineer’s Office
technical articles specific to the design, inspection,
safety, and construction of small dams. It provides This news update was compiled, written,
general information. The reader is encouraged to use and edited by URS Corporation in Denver,
the references cited and engage other technical Colorado
experts as appropriate.
Funding for the News Update has been
provided by the FEMA National Dam Safety
Act Assistance to States grant program.
Valuable Low-Cost Reference:
The Embankment Dam Reference Toolbox provides a Article Contributors:
comprehensive collection of design standards and URS Corporation: Max Shih, PhD, PE;
references for dam engineering available from ASDSO. Casey Robertson; Ed Villano, PE; Greg Glunz,
PE Consultant: Hank Falvey, Dr. Ing;
Upcoming ASDSO Webinar Dam Safety Training:
 Hydraulic Design of Labyrinth Weirs, by Blake P. Tullis, Ph.D., Editorial Review Board:
and Brian Crookston. August 20, 2013 Michele Lemieux, PE, Montana Dam Safety
 Stability Evaluations of Concrete Dams, by Guy S. Lund, P.E., Program Supervisor; Bill McCormick, PE, PG,
and Robert Kline, Jr., P.E. October 8, 2013 Colorado Chief Dam Safety Branch;
ASDSO Training Website Link Mike Hand, PE, Wyoming Dam Safety
Program; Mark Ogden, PE, Association
of State Dam Safety Officials;
An error was found in Issue 01 of this publication. Please note the Matthew Lindon, PE, Loughlin Water
following correction to the article titled “Simple Steps to Siphoning”: Associates; and Steve Becker, PE,
Hmax must be greater than the value of (DCE-RWS), Hmax>(DCE-RWS)
Natural Resources Conservation Service
The corrected Issue 01 can be found via the following link:
Western Dam Tech Note Issue 01
The material in this publication has been prepared in accordance with generally recognized engineering principles and practices, and is for general information only. The information
presented should not be used without first securing competent advice from qualified professionals with respect to its suitability for any general or specific application. No reference made in
this publication constitutes an endorsement or warranty thereof by URS Corporation or sponsors of this newsletter. Anyone using the information presented in this newsletter assumes all
liability arising from such use.

Introduction……………………………………….….. 1 Design of Riprap for Slope Protection against Wave Action ……………7
Predicting Wave Runup on Dam Slopes ……...…2 Design Considerations for Outlet Works Air Vents…………………. .....12

1
Predicting Wave Runup on Dam proportionally higher waves will be produced. TR-69
recommends two approaches to determine the design
Slopes fetch and wind direction: (1) U.S. Weather Service
Introduction climatological data or (2) site orientation. Because
most dams/reservoirs are ungauged, wind data does
When wind blows over an open water surface, such as
not typically exist and the site orientation method is
within a reservoir, wind-generated waves can strike
preferred to define the effective fetch and design wind
the upstream slope of the dam embankment. This can
direction.
cause erosion of the embankment material and if
severe enough, waves can overtop the embankment, The design wind direction is obtained by determining
both of which are dam safety issues. Therefore, the the longest stretch of open water from a point on the
dam embankment design must consider the potential shoreline opposite to the dam embankment. It is
effects of wave action and protect against erosion of assumed that wind and waves are developed along the
the embankment materials and overtopping due to longest fetch of open water from the dam. According
wave runup. This is done by extending the to Saville’s 1954 study, the width of the fetch on inland
embankment up from the still water flood pool level to reservoirs normally places a definite restriction on the
an elevation equal to the still water pool plus the length of effective fetch, which is the effective distance
maximum calculated wave runup and wind setup of the water over which the wind blows without
height. appreciable change in direction. Figure 1
diagrammatically shows the central (longest) and radial
This article describes a procedure for calculating the
fetch lines for a hypothetical reservoir. Simplistically,
wind-generated wave characteristics for inland
this method involves drawing the central radial line
reservoirs and lakes and the resulting wave runup on a
and then drawing seven radial lines at 6-degree
sloping dam embankment for small dams.
intervals on each side of the central radial line.
Dominant Factors and Procedure
The major variables used to calculate wind-generated
wave height on open water surfaces, such as
reservoirs, and influence embankment design are:
 Effective Fetch and Wind Direction
 Wind Speed over Water
 Wind Setup, Wave Height and Runup
The procedure presented in this article is based on
information presented in TR-69 (USDA, 1983) and
Bureau of Reclamation ACER TM-No. 2. Additional
information related to US Army Corps of Engineers
(USACE) procedures is presented in the reference
documents included at the end of this article.
Effective Fetch and Design Wind Direction
Figure 1: Central and radial fetch lines
The procedure in this article is limited to The effective fetch, Fe, can then be computed using
reservoirs where 1.) Effective fetch is less than Equation 1.
10 miles and 2.) Wave height is less than 5 feet. ∑ ( )
∑ ( )
Eq.1
The fetch is an overwater length blown on at a xi = Length of Radial Line i
constant wind speed and direction. The longer the αi = Angle Degree between the Central Radial Line
fetch and the faster the wind speed, the more wind and the Radial Line i
energy is imparted to the water surface and

2
Wind Speed over Water
There are two common procedures for determining
the design wind speed. They are:
1. A constant overwater wind speed of 100 mph
(Reclamation, 1987)
2. Site-specific wind speed and duration curves
The 100 mph wind speed recommended by
Reclamation is a simple but conservative approach.
The more detailed site-specific approach is presented
in the following paragraphs.
According to the guidelines titled Reclamation ACER
Figure 2: Plot of wind speed vs. duration
TM-No. 2 and TR-69, the design wind speed and
duration can be selected by using the observed Table 1: Maximum wind speed relationship
maximum wind speed and the effective fetch. Ratio of Land Wind Speed to the Fastest
Fastest Mile Wind Mile Wind for the Durations
Commonly, the observed fastest mile wind speed is Speed, mph
considered as the maximum overland wind speed, UL, 1 min* 30 min 60 min 100 min
and can be obtained from the National Oceanic and 100 100% 52% 46% 41%
Atmospheric Administration (NOAA) National Climatic 80 100% 57% 51% 47%
Data Center websites shown at the end of this article. 60 100% 65% 59% 55%
The NOAA wind data, including wind speed, duration, * Duration of fastest mile wind speed is one minute.
and direction, indicates the overland wind
characteristics at 25 feet above ground. The blue curve in Figure 2 needs to be generated using
Figure 2 in TR-69 or the empirical relationship
The duration of a given wind speed that needs to be (Equation 2) of overland wind speed and duration for
maintained to fully develop the maximum waves is a the site specific effective fetch.
function of the effective fetch. The longer the effective
fetch, the longer the duration for the sustained wind ( ) Eq.2
speed. Figure 2 graphically shows the selection of
design wind speed based on the relationship between g = Gravitational Acceleration, 32.2 ft/sec2
the maximum wind speed and the effective fetch T = Wave Duration in seconds. Wave duration is
response to wind speed. The intersection of the red equal to the minimum wind duration required
curve and blue curve identifies the “design wind for generation of wave heights for a specific
speed.” effective fetch and wind speed.
The red line on Figure 2 can be developed using the UL = Overland Wind Speed in ft/sec
observed fastest mile wind speed and the information Fe = Site Specific Effective Fetch in ft
contained in Figure 5 of TR-69. Alternatively, Table 1 is Because of smoother and more uniform surface
provided as a simplification of the information shown conditions, overwater wind speeds, Uw, are higher
in Figure 5 of TR-69. than overland wind speeds, UL. To consider this speed
enhancement, the overwater wind speed can be
computed using the following equation.
Eq.3
Careful!! The units for effective fetch and
wind speed vary for the various equations in this β = Wind Speed Adjustment Factor or Ratio, Uw/UL,
article. Make note of units required for each eqn. Shown on Figure 3.

3
 Equations 5 and 6 are empirical equations developed
from deep-water waves, which are defined as waves
having lengths equal to or less than 2D. They also give
Wind Ratio (Uw/UL)

conservative wave height estimations for shallow-

water waves.

The significant wave height defined above

would be exceeded by approximately 33
percent of the expected waves generated by the
associated wind speed. If a lower potential of
Figure 3: Wind speed relationship – water to land exceedance is desired, a wave height of 1.27Hs and
1.67Hs have a corresponding potential for
Wind Setup, Wave Height, and Runup exceedance of 10 percent and 1 percent,
A sketch of waves striking an embankment slope is respectively.
illustrated in Figure 4. When wind is blowing over a
water surface, horizontal shear stress acts on the When waves reach a sloping embankment, the waves
water surface, and the water surface is tilted in the will eventually break on the slope and run up to a
direction of the wind. This wind effect is termed “wind height governed by the angle of the slope, and the
setup” and can be estimated using the empirical surface roughness and permeability. Wave runup
equation from TR-69 shown below. height, R, is the difference between the maximum
elevation reached by wave runup on a slope and the
Eq. 4 storm water level. The steeper the embankment slope
the greater the wave runup height. Many studies have
S = Wind Setup in feet been published that provide guidance for determining
Uw = Wind Speed in miles per hour the wave runup height on slopes. The runup from a
F = Wind Fetch in miles (Approximately equal to Fe) significant wave on an embankment slope with riprap
D = An approximation of the average water depth protection can be predicted using:
along the fetch length in feet
Slope protection is generally designed for what is Eq. 7
( )
known as the “significant wave height.” The significant
wave height is the average height of the highest one- R = Wave Runup Height in feet
third of the wind-generated waves. This means that 33 Hs = Significant Wave Height in feet
percent of the waves that hit the slope will be higher L = Wave Length in feet
than this value. Based on the selected design θ = Angle of the Dam Face from Horizontal
overwater wind speed and the effective fetch, the Equation 7 should be used only for embankment
significant wave height, Hs, and wave length, L, can be slopes steeper than 5H:1V.
estimated using the following dimensionless equations
from TR-69.

( ) Eq. 5

√
( ) Eq. 6

g = Gravitational Acceleration, 32.2 ft/sec2

Hs = Significant Wave Height in feet
L = Wave Length in feet Figure 4: Sketch illustrating wave terms
UL = Overland Wind Speed in ft/sec
Fe = Effective Fetch in feet

4
Conclusions Table 2: Procedure to determine the effective fetch
The wind-generated wave characteristics and the Radial Radial Length α 2
No. (mi), Xi (Degree) cos α Xi·cos α
related wind setup and wave runup on a sloping 1 1.7 42 0.74 0.96
embankment within a reservoir must be considered for 2 1.8 36 0.81 1.20
3 1.9 30 0.87 1.45
the purposes of designing embankments and 4 2.0 24 0.91 1.70
embankment slope protection. Slope protection for 5 2.2 18 0.95 2.02
6 2.3 12 0.98 2.23
the embankment must also be considered and a 7 2.4 6 0.99 2.41
procedure for the design of riprap slope protection is 8 2.6 0 1.00 2.63
9 2.5 6 0.99 2.51
described in the following article titled, “Design of 10 2.4 12 0.98 2.33
Riprap for Slope Protection against Wave Action.” 11 2.3 18 0.95 2.11
12 2.1 24 0.91 1.78
13 2.0 30 0.87 1.53
NOAA Climatological Data Links 14 1.8 36 0.81 1.20
15 1.7 42 0.74 0.96
Local Climatological Data:
Sum= 13.51 27.02
http://www.ncdc.noaa.gov/IPS/lcd/lcd.html
Climate Maps of the United States: ∑ ( )
∑ ( )
miles
http://cdo.ncdc.noaa.gov/cgi-bin/climaps/climaps.pl
This effective fetch of 2.0 miles or 10,560 feet
NOAA Climate Data Online: from the given reservoir with a longest fetch of
http://www.ncdc.noaa.gov/cdo-web/ 2.6 miles is estimated.
References (with Links where available) 2. Refer to Figure 5 of TR-69 or Table 1 in this article,
 USDA (1983), Technical Release No. 69: “Riprap for Slope Protection the generalized maximum wind speed-duration
against Wave Action.”
 Reclamation (1992), ACER Technical Memorandum No. 2: “Freeboard
relationship is plotted as the red line on Figure 5.
Criteria and Guidelines for Computing Freeboard Allowances for This is computed by using the observed fastest
Storage Dams.” mile wind speed, 75 mph, interpolating the ratio
 Reclamation (1987), Design of Small Dams, Third Edition.
 USACE (1976), Engineering Technical Letter No. 1110-2-221: “Wave
of land wind speed to the fastest mile wind for
Runup and Wind Setup on Reservoir Embankments.” each of the durations shown and then multiplying
 Saville, Thorndike J. (1954), “The Effect of Fetch Width on Wave this ratio by the observed fastest wind speed. The
Generation,” Technical Memorandum No. 70, Beach Erosion Board,
USACE.
results of these computations are shown in Table
2.
Example #1: Table 2: Maximum Wind Speed-Duration Relationship
Find the wind setup, the wave height and the wave for a Fastest Mile Wind of 75 mph
runup of a reservoir as shown on Figure 1. The
observed fastest mile wind speed is 75 mph for this 1 min 30 min 60 min 100 min
site. The average depth of the reservoir is 10 feet, and Interpolated Ratio
100% 59% 53% 49%
from Table 1
the riprap protected embankment has a 3H:1V or 18°
Corresponding Max.
slope. 75 44 40 37
Wind Speed (mph)
Calculations: 3. By using Equation 2 and the effective fetch, 2.0
1. To measure the lengths of the central (longest) miles, the relationship of overland wind speed-
and radial lines as shown in Figure 1, compute the duration for the selected fetch is determined for a
effective fetch using Equation 1. The computation range of selected speeds (in this case, UL= 90
is shown in Table 2. mph, 60 mph, and 35 mph). Remember to first
convert UL to ft/sec and fetch length to feet. T is
calculated in seconds with Equation 2 and then
converted to minutes for the plot. The results are
shown as the blue curve in Figure 5.

5
 ( )

Figure 5: Plot of wind speed vs. duration

4. The intersection of the red curve and blue curve
identifies the “Design Overland Wind Speed ( )”
of 52 mph or 76 ft/sec. Find the overwater wind
speed using Figure 3 and Equation 3. This gives a
wind ratio of 1.21 from the figure and an adjusted
overwater wind speed ( ) of 63 mph.

5. Find the wind setup using Equation 4.

6. Find the wave height using Equation 5.

( )
7. Find the wave length using Equation 6.

( ( ) )
8. Find the wave runup height using Equation 7.

( ) ( )

Results:
The estimated maximum significant wave height is 3.2
feet with an overwater wind speed of 63 mph. The
corresponding maximum height the water will reach
from the still water flood level is 3.5 feet, which is the
sum of wind setup (0.6 foot) and runup (2.9 feet).

6
Design of Riprap for Slope
Protection against Wave Action
Introduction
This article is intended to provide practical guidance to
engineers for the design and construction of riprap for
embankment dams, particularly small embankment
dams. This article is not intended to be an all-inclusive
guide. A list of commonly used references on the topic
is provided at the end of this article.
As discussed in the previous article of this issue,
earthen embankment dams can be subject to erosion
by wave action within the reservoir. In 1983 the USDA Figure 1: Erosion of a small embankment dam in
developed a technical release (TR-69) that describes Montana caused by wave action.
procedures for the design of rock riprap protection for
earthen embankments to protect against wave action.
TR-69 was used as the basis for this article. Detail not
found herein can be found in TR-69 and the associated
references. As mentioned in the previous article the
design procedures described in TR-69 are generally
limited to reservoirs having an effective fetch length of
less than 10 miles and significant wave height of less
than 5 feet.
Additional relevant publications for guidance on the
design of riprap slope protection include Chapter 7 of
“Embankment Dams” (Reclamation 1992), “Design of
Small Dams” (Reclamation 1987), “Design of Coastal
Revetments, Sea Walls, and Bulkheads” (USACE 1995)
and “Design of Riprap Revetment” (FHWA 1989).
Why Riprap?
Slope armoring acts as primary protection against
embankment erosion caused by wind and wave action
within the reservoir. Excessive erosion of a dam
embankment can lead to embankment failure.
Inadequately designed or installed riprap can pose a
dam safety risk. For successful performance, a riprap
layer must be designed to:
 Protect the individual rock particles from
displacement by the wave force, and
 Keep the protected earth, filter, and bedding
underlying the riprap from being washed out Figure 2: Erosion of a small embankment dam in
through the voids in the riprap. Montana caused by wave action.
Figures 1 and 2 are examples of the embankment Riprap is one material commonly used as armoring for
erosion that can occur without adequate protection upstream slope protection. There are other
against wave action. commercially available armoring materials, each with
their own design considerations and methodologies.

7
Some of these alternate materials include articulated increases. Conversely as the embankment slope and or
concrete blocks, cellular concrete mats, and in some significant wave height decreases, the calculated
low wave-energy sites, vegetation or geosynthetic rock weight reduces.
reinforced vegetation. This article focuses on the Determining Type, Size, Thickness and
design of riprap armoring, as it is the most commonly Gradation
preferred and installed material.
There are two types of rock placement described in TR-
Procedure 69:
In general terms and in TR-69, the procedure for the  Type 1 – Dumped (Equipment-Placed) Rock
design of riprap can be summarized as a flow chart as  Type 2 – Hand-Placed Rock
shown on Figure 3. This procedure is described in the
Dumped rock is regarded as superior to hand-placed
following sections of this article and an example
rock because of historically low maintenance costs.
(Example #2) is provided at the end of this article.
Experience has also shown that in most cases dumped
Example #2 is a continuation of Example #1 from the
rock provides the best upstream slope protection at
previous article in this newsletter.
the lowest ultimate cost. For these reasons, only
Determine Rock Weight dumped rock is discussed further in this article.
The procedure for determining the physical riprap
characteristics described in TR-69 for Type 1 (dumped)
Determine Type and Size of Riprap rock is as follows:
 Size: using the W50 weight of rock, find rock
size (D50) using Figure 9 (TR-69) or the
Dumped Hand-Placed equations provided with the figure. Usually the
equation for spherically shaped rock is used to
estimate rock size for riprap as follows:
Determine Gradation and Thickness of Riprap
and/or Bedding and/or Filter √ Eq.2
Where, GS = Specific Gravity
 Gradation: using the rock size, find the
Determine Limit of Riprap Protection gradation limits using Figure 10 (TR-69).
 Thickness: two times the D50 rock size.
Figure 3: Summary flow chart procedure for design of
riprap.
According to the hazard category of the dam a safety
Determining Rock Weight factor can also be applied to the calculated D50 rock
In accordance with TR-69, the equation to estimate the size and this is described in “Slope Protection for Dams
required riprap rock weight (W50) can be given as: and Lakeshores” (USDA 1989). Alternative methods for
determining riprap size, thickness, and gradation are
Eq.1 described in Chapter 7 of “Embankment Dams” (USBR
( )
1992).
GS = Specific Gravity Generally riprap should be hard, dense angular stone,
HS= Significant Wave Height (See previous article for graded as designed, comprising sound fragments
calculation method) resistant to abrasion and weathering and be free of
cot  = Horizontal Component of Embankment Slope cracks, seams, clay, organic material and other defects.
Rounded boulders or cobbles are not generally
acceptable as riprap.
Rock weight can also be estimated using Figure 8 in TR-
69. As the embankment slope and or significant wave
height increases, the calculated rock weight also

8
Bedding and Filters wind setup (S). This can be calculated as described in
Once the gradation of the riprap is determined, the the previous article. The lower limit of riprap is
gradation and thickness of the bedding layer should be determined by the lower of either the (a) vertical
determined. In principal, the bedding layer provides a distance of 1.5 times Hs below the still water flood
foundation for the riprap placement and also provides pool, or (b) lowest controlled outlet elevation.
a filter-compatible transition layer to finer, underlying The upper and lower limits of riprap are shown on
embankment materials. The finer embankment Figure 4.
material underlying the riprap could be washed out
through the rock particles during reservoir fluctuations
and wave action. Retention of the underlying
embankment materials is attained by placing a finer-
grained layer of bedding under the riprap. Where very
large riprap is used, a progressively finer two-stage
bedding/filter layer may be required. The bedding
layer needs to be filter-compatible with both the
underlying embankment material and overlying riprap Figure 4: Typical upper and lower limits for riprap
to limit the potential of erosion and washout of both placement.
embankment and bedding material between the voids
of the riprap. For owners of existing small dams the extent of a
Generally bedding should be a well-graded mixture of riprap revetment may be limited by the budget
gravel and sand that is filter-compatible with both the available to complete the project. When this is the
riprap and the embankment materials. There is some case the owner and designer should carefully consider
general guidance on developing the filter-compatible where the riprap can offer best value from a dam
gradation and the recommended thickness provided in safety and operational perspective. Priorities could
Chapter 7 of Embankment Dams (Reclamation 1992) include, but may not be limited to, providing riprap on
and in Design of Riprap Revetment (FHWA 1989). The sections of the embankment where erosion has
general guidance for bedding thickness is summarized previously occurred, is deemed likely to occur (i.e.,
in Table 1. adjacent to concrete structures and other
Table 1: Bedding layer thickness according to riprap infrastructure), and or in horizontal bands at the
layer thickness. reservoir normal water level or normal operational
Riprap Layer Bedding Layer water level. Experience has shown that dam
Thickness Thickness embankments built with interior or exterior bends or
at angles that are perpendicular to prevailing winds,
12-24” 9” can be more susceptible to erosion. Armoring of these
27-36” 12” areas should be a priority.
Over 36” 15” Placement
According to TR-69, for dumped rock, the placement of
Limit and Layout of Riprap Protection bedding and riprap on a dam embankment should be
as shown on Figure 5. This figure shows the riprap
According to TR-69, the lower limit of the riprap
supported by a level berm (also refer to Figure 4),
protection should be 1.5 times the significant wave
which facilitates placement.
height (Hs) below the reservoir normal water level at
the lowest ungated opening, or below the lowest
controlled outlet.
The upper limit of riprap is described by TR-69 as the
vertical distance above the reservoir still water flood
pool level equal to the sum of the wave runup (R) and

9
 Conclusion
Properly designed and installed riprap can provide
erosion protection from wave action that would
otherwise cause significant damage of earthen
embankment dams. For riprap to be effective the
designer must calculate the required riprap weight,
size and gradation, and specify acceptable material
properties. The designer must carefully consider
Figure 5: Dumped rock placement detail. bedding and/or filter requirements to ensure that they
are compatible with the embankment material and the
Where construction of a berm is impractical or on an
riprap itself.
existing slope, keying of the riprap into the slope is
recommended to prevent displacement of riprap down References (with Links where available)
the slope. A reference published by the Minnesota
To aid the designer through the process, the following
NRCS state office titled “Slope Protection for Dams and
is a list of design references that can be used:
Lakeshores” (USDA 1989) provides alternatives for
 USDA (1983), Technical Release No. 69: Riprap for Slope Protection
keying riprap into existing slopes where the riprap will against Wave Action.
not extend the full height or length of the dam.  USDA (1989), Minnesota Technical Note 2: Slope Protection for Dams
and Lakeshores.
The placement of riprap should be done by mechanical  Reclamation (1992), “Design Standards No. 13: Embankment Dams”,
means, such as a hydraulic excavator. Dumping riprap Chapter 7 – Riprap Slope Protection, U.S. Department of Interior,
from a truck down an embankment should be avoided Bureau of Reclamation, Technical Service Center, Denver, CO. Look for
the revised version of this reference at the following link soon:
as it can cause segregation of the rock by size and Reclamation Dam Design Standards
result in unsuitable gradation. Placement should be  Reclamation (1987), “Design of Small Dams”, U.S. Department of
performed to produce a well-graded, even mass of Interior, Bureau of Reclamation.
 FHWA (1989), “Design of Riprap Revetment”, Federal Highway
rock with uniform cover and minimal voids. Laborers Administration, McLean, Virginia.
should be provided during placement for
rearrangement of loose rock fragments, “chinking” of
void spaces, and hand-placement as needed to provide
a well-keyed and stable layer of riprap.
Figure 6 shows dumped riprap being placed over
bedding on the upstream slope of a small dam in South
Dakota.

Figure 6: Placing riprap bedding and riprap on the

upstream slope of a small dam in South Dakota.

10
Example #2: accordance with the aforementioned
Design embankment riprap protection for the dam reference, TR-69 and “Design of Riprap
described as Example #1 in the previous article of this Revetment” (FHWA 1989).
newsletter. The upstream dam embankment slope is
3H:1V, the significant wave height (HS) was calculated 5. Determine the limit and layout of the riprap
as 3.2 ft and the specific gravity ( ) of the riprap protection. Consider the limits described in
source is 2.65. this article and in TR-69.
Calculations:
1. Determine the required rock weight for
the riprap using Equation 1:

( )

2. Using the rock weight determine the

rock size for the riprap using Equation 2:

The riprap layer thickness and maximum rock

size is calculated as two times the D50 rock size.
Using the D50 rock size of 1.15 feet, DMAX is 2.3
feet.

3. Using the rock size estimate the gradation

limits using Figure 10 (TR-69). Gradation limits
for a riprap with a rock size of 1.15 feet
are shown in Table 2.
Table 2: Gradation limits for a D50 rock size of 1.15
feet.
Rock Passing Sieve Size
100% 21 to 28”
85% 19 to 26”
50% 14 to 20”
15% 2 to 9”

4. Using the guidance on the bedding layer

thickness provided in Chapter 7 of
“Embankment Dams” (Reclamation 1992),
adopt the bedding layer thickness as 12”.
Determine the gradation of the bedding and
any requirements for a filter layer in

11
Design Considerations for Outlet control gate/valve, is the most common consequence
of inadequate air vent design. As cavitation bubbles
Works Air Vents
are carried downstream from the gate into higher
Introduction pressure flow areas, they rapidly collapse (implode),
Outlet works air vent design is often a difficult, sending out high-pressure shock waves that can
misunderstood, or even unknown subject for many damage a conduit wall near the implosion. Cavitation
design engineers. This article introduces the subject of damage generally occurs downstream of the gate slots
air demand and air vent sizing, and discusses possible in the outlet works, but can also occur on the invert
consequences of inadequate air vent design. Important downstream of the control gate. Figure 1 shows typical
design criteria and guidelines are summarized, and a cavitation damage on an outlet gate and conduit walls.
conservative, generalized approach for estimating air
demand and sizing air vents is provided.

Several references containing alternate design

methodologies are presented in this article; however,
it is cautioned that there are limitations associated
with each design method. Designers should check
these limitations to ensure the specifics of their
projects are consistent with the methods being
employed.

Why Air Vents?

An important consideration in any closed conduit
design for an outlet works is the proper use of air
venting. An air vent simply allows air under
atmospheric pressure to flow into an outlet works
conduit, introducing (or entraining) air into the flow.
Specifically, a properly designed air vent serves the Figure 1: Typical cavitation damage on gate and conduit
following purposes:
Air blowback can occur as air collects on the crown of
 Reduces potential for formation of low the conduit downstream of a control gate and forms a
pressures within the flowing water; large pocket of air that can violently “blow back”
toward the control gate and intake structure, causing
 Reduces potential for unstable flow in the
damage to those structures.
conduit; and
 Allows bleeding of air from a conduit prior to Pipe collapse downstream from a gate can also occur if
operation. low pressure flow is extreme enough, as illustrated in
Air vents are typically installed downstream from a Figure 2.
control gate or valve, where formation of low flow
Excessive vibration in low pressure or unstable flow
pressures can occur. In the absence of adequate air
areas downstream of a control gate can lead to
venting, low flow pressures can lead to cavitation, air
structural damage of the conduit and gate, if severe
blow back, pipe collapse, excessive vibration, and
enough.
excessive noise. Each of these possible consequences is
discussed below. Excessive noise can occur at the air vent opening if the
air vent is designed too small. The noise can even be so
Consequences of Inadequate Air Vent Design
loud that it is damaging to hearing. At one dam, nearby
Cavitation, or the formation of vapor cavities (bubbles) residents complained of a popping noise coming from
in low pressure areas just downstream from the the air vent that was keeping them up at night.

12
 gate opening heights. Figure 3 illustrates the effect of
gate openings on air demand. At small gate openings
and when flow is not influenced by tailwater
conditions or by a hydraulic jump, “jet flow” occurs,
which entrains large quantities of air as the water jet
frays or breaks up. At large gate openings and free
surface flow conditions, air demand is caused by the
drag force between the water surface and the
overlying air column. Air demand for flow involving a
hydraulic jump has been shown by studies to represent
the lower bound of free surface flow air demand.
When the conduit flows full, or when the gate is at the
downstream end of the conduit (open to atmospheric
pressure), air demand is zero.

Figure 2: Outlet pipe liner collapse due to cavitation

Air Demand
Estimating air demand is the most important
component of adequate air vent sizing. Air demand
refers to the amount of air that the flowing water pulls
into the conduit (and entrains into the flow) through
the air vent and through the downstream exit portal (if
not submerged).

There are a number of variables that can influence air

demand, including:

 Gate opening height

 Head
 Volume flow rate and velocity of water
 Flow type (e.g., free surface flow, or hydraulic
jump that closes the conduit)
 Froude Number
 Gate geometry and roughness
 Conduit length, diameter, cross section shape,
Figure 3: Gate Opening, Flow Type, and Air Entrainment
and roughness
Demand
 Water surface roughness In addition to gate opening height and flow type, the
other variables bulleted above influence air demand to
 Outlet submergence
varying degrees. Accounting for these variables in air
 Air vent geometry (e.g., entrance, bends) and demand estimation can be challenging for the
head loss practicing design engineer, as there is currently no
 Altitude known comprehensive methodology applicable to the
Air demand is usually greatest at small (5 to 10 percent wide range of possible outlet works configurations and
open) and large (between 50 to 100 percent open) hydraulic conditions represented by these variables.

13
Fortunately, for small to medium size dams where air from Air-Water Flow in Hydraulic Structures may be
vents are likely not nearly as costly as for large dams, a used:
conservative design approach summarized below can
be employed, wherein the air vent is oversized, ( ) ( )
negating the need for rigorous hydraulic analysis or
model studies to account for all the variables. In cases
where:
where cost is a more significant issue, such as for low
budget projects or for larger or more complex dams, a Fr = Froude Number Upstream of the Hydraulic Jump
number of references describing alternate (Note: Fr is a dimensionless index of flow regime (i.e.,
methodologies are provided below. subcritical or supercritical)).
A Generalized, Conservative Design In a circular pipe, Fr can be calculated from the flow
Approach depth y by using the following equation:
For flow in gated closed conduits with free surface
open channel flow conditions (i.e., jet flow and air drag
flow), the following equation, obtained from the 1980 ( )
publication Air-Water Flow in Hydraulic Structures (See
where:
references for full citation.) may be used to calculate
maximum theoretical airflow rate: V = Mean Flow Velocity
g = Gravitational Constant
( )
ye = Effective Depth = A/T
A = Cross Sectional Area of the Water in the
where: Conduit
T = Top Width of Flow Passage = 2[y(D-y)]1/2
( ) = Air Demand Ratio
D = Conduit Diameter
Qa = Volume Flow Rate of Air
Y = Flow Depth
Qw = Volume Flow Rate of Water
After Qa is calculated, a maximum design air velocity
Ad = Cross Sectional Area of Conduit
can be selected, and the cross sectional area and
Awp = Maximum Cross Sectional Area of Water in diameter of the air vent can be calculated. An example
Conduit calculation using this design method is provided at the
end of this article.
Ideally, a conduit water surface profile should be
calculated for a range of gate opening heights to arrive As a side note, the Bureau of Reclamation
at Awp. Alternatively, Awp can be approximated from the conservatively designs their outlet conduits so that a
water surface profile corresponding to a gate opening hydraulic jump will theoretically never occur, while the
of 75 percent under maximum design head, as studies U.S. Army Corps of Engineers (USACE) allows hydraulic
have shown that maximum air demand typically occurs jumps in outlet conduits at their dams.
at/near 75 percent gate opening and maximum design
head. As a rough check, the design engineer should Alternative Design Methodologies
verify that the maximum volume flow rate of air is The 1980 USACE Engineering Manual Hydraulic Design
approximately equal to the maximum flow rate of of Reservoir Outlet Works (EM 1110-2-1602), together
water. with “Hydraulic Design Criteria” 050-1 and 050-2,
present a method of estimating air demand and sizing
For cases where the water surface profile indicates
the air vent based on an envelope design curve that
that a hydraulic jump will occur, the following equation
was developed from outlet works air demand data
from 5 different dams with heads ranging from 24 to

14
370 feet. The method relates Froude number and air  Limit maximum air flow velocity in the air vent
demand ratio and is generally applicable for slide and to approximately 100 feet/second by
tractor gates operating in rectangular gate chambers. increasing the vent size as necessary; above
The envelop design curve may underestimate air this velocity an objectionable, whistling noise
demand in some cases, such as for Beltzville Dam, occurs that can be damaging to hearing.
where actual air demand was 5 times higher than the  For safety reasons, keep children away from
air demand derived from the design envelop curve. vent openings, and place personnel barriers
This illustrates the necessity for the designer to check around vents if the air velocity is expected to
the limitations and applicability of a given method to exceed approximately 50 feet/second.
ensure the specifics of their projects are consistent
 A minimum air vent diameter of 4 inches
with the methods being employed. A spreadsheet that
should be used for all cases to facilitate vent
employs this design method is attached to this
cleaning and maintenance.
document.
 For valves, the air vent is typically located
The 2011 paper titled, Determining Air Demand for upstream from the point where the water jet
Small- to Medium-Sized Embankment Dam Low-Level impinges on the conduit walls.
Outlet Works presents a design method for estimating  If the air vent is of sufficient size to interrupt
air demand and sizing the air vent based on laboratory- rebar in the conduit wall, use a series of
scale low-level outlet tests with an inclined gated inlet smaller, side-by-side air vents.
on a 3H:1V slope. The design methodology presents a
series of design curves that relate gate geometry (and  Install an air vent through HDPE and CIPP pipe
corresponding discharge coefficient), driving head, liners if there is susceptibility to internal
gate opening (10, 30, 50, 60, 70, and 90 percent), and vacuum pressures and liner collapse.
air demand ratio. The design method uses an envelope  If steel vent pipes are used and will be in
curve of all the observed model data; with the contact with corrosive soils, design appropriate
limitation that parameters such as conduit length and cathodic protection, or use a protective
air vent geometry (and associated head losses) were coating or wrap.
not considered in the model, and the method may not  A typical configuration for the end (open to
be applicable for gates with inclinations different than atmosphere) of the air vent is to include a 90
3H:1V. degree elbow (see Figure 4) with an expanded
or bell-mouth opening oriented away from the
The 2008 thesis titled, Air Demand in Free Flowing
prevailing winds, with a stainless steel screen
Gated Conduits summarizes empirical design
over the opening, which will help prevent
methodologies developed by previous researchers, and
debris from entering the vent, and help
presents observations on significant parameters
prevent water from entering the pipe, which
developed from a laboratory model study. The
could result in freezing blockage during the
parameters studied included: Froude number, ratio of
winter.
head to gate opening, surface water roughness,
conduit length, and conduit slope. A possible limitation  Avoid air vent design features that could result
of this study is that the model air velocity in large head losses such as a small-mesh steel
measurements were not sufficiently detailed to draw screen, or an excessive number of vent pipe
conclusions. bends.
 Take precautions against small objects (e.g.,
Air Vent Design Criteria and Guidelines rodents, clipboards, etc.) getting sucked into
The following criteria and guidelines are commonly the vent and creating a potential blockage;
employed in air vent design practice: periodically inspect the air vent to ensure air is
flowing freely through it and that there are no

15
 blockages, corrosion, or structural damage that Hydraulic Design of Reservoir Outlet Works, EM 1110-2-1602, U.S. Army
Corps of Engineers, October 15, 1980 ; together with HDC 050-1 and HDC-
may affect performance. 050-2.
 For cases where it is not possible for an air
vent to have direct connection to the Air vent sizing example using method from
atmosphere, such as for control gates located the 1980 publication Air-Water Flow in
in outlet works tunnels, air demand must be Hydraulic Structures:
supplied by an air duct above the free surface
of the flowing water, and the hydraulic design Given:
should ensure flow never rises to the level of  Conduit diameter = 2 feet
the air duct.  Maximum water depth in conduit
corresponding to 75% gate opening = 1.5 feet
 Volume flow rate of water (Qw) = 50 ft3/s

Calculate:

= = 3.14 ft2

Awp = 2.53 ft2 (obtained from table typically found in

hydraulic textbooks that provides numerical values for
area, wetted perimeter, and hydraulic radius for a
partially filled circular pipe)

Figure 4: Typical outlet works air vent for a small dam

( ) = = 0.24
It is also important to point out that there are several
outlet works hydraulic flow issues that are commonly
misattributed to insufficient air vent size, but are Qa = 0.24 * Qw = 0.24*50 ft3/s = 12 ft3/s
actually associated with inadequate hydraulic design or
Setting maximum velocity at 100 ft/s,
operations errors. These include surging, structural
damage due to filling the pipe too rapidly, and bi- A=Q/V = (12 ft3/s)/(100 ft/s) = 0.12 ft2 =17.3 in2
stable flow in the conduit.

References (with Links where available)

 Air-Water Flow in Hydraulic Structures, A Water Resources Technical Dpipe = √ =√ = 4.7 inches
Publication, Engineering Monograph No. 41, United States Department
of the Interior, Water and Power Resources Service, by Henry T. Falvey,
Engineering and Research Center, Denver, CO, December 1980.
Increase Dpipe to commonly available pipe size of 6
 Cavitation in Chutes and Spillways, A Water Resources Technical inches.
Publication, Engineering Monograph No. 42, United States Department
of the Interior, Bureau of Reclamation, by Henry T. Falvey, Research
Engineer, Denver, CO, April 1990.
 Air Demand in Free Flowing Gated Conduits, D. Peter Oveson, A thesis
submitted in partial fulfillment of the requirements for the degree of
Master of Science, Civil and Environmental Engineering, Utah State
Engineering, Logan, Utah, 2008.
 Determining Air Demand for Small- to Medium-Sized Embankment Dam
Low-Level Outlet Works, Journal of Irrigation and Drainage Engineering,
American Society of Civil Engineers, B.P. Tullis, and J. Larchar, December
2011.

16