'
RESEARCH REPORT H-70-2
EROSION AND RIPRAP REQUIREMENTS AT
CULVERT AND STORM-DRAIN OUTLETS
l-lydraulic Laboratory Investigation
by
J. P. Bohan
January 1970
Sponsored by
Conducted by
Office, Chief of Engineers, U . S. Army
U. S. Army Engineer Waterways Experiment Station, Vicksburg, Mississippi
AR M Y · M RC V I CKSBURG . M ISS.
This document has been approved for public release and sale; its distribution is unlimited
w
J 11-- ,_
IVIJ• fi-7D-
~
~,'J
FOREWORD
The experimental investigation reported herein was authorized by the Directorate of Military
Construction, Office, Chief of Engineers, 1 July 1963.
The study was conducted in the Hydrau-
lics Division of the U. S. Army Engineer Waterways Experiment Station during the period July
1963 to June 1969 under the direction of Mr. E. P. Fortson, Jr., Chief of the Hydraulics Division, and Mr. T. E. 1\\urphy, Chief of the Structures Branch.
Mr. J. L. Grace, Jr., Chief of the
Spillways and Conduits Section, provided supervision during the formulation and conduct of the
tests, analysis of data, and preparation of this report.
The tests were conducted by Messrs. G. A.
Pickering, E. C. Maschmeier, J. 0. Farrell, and C. R. Styron III, with the assistance of Engineering
Aides B. Perkins and C. Wright.
The data were analyzed by Messrs. Styron and J. P. Bohan.
This report was prepared by Mr. Bohan.
Directors of the Waterways Experiment Station during the investigation and preparation and
publication of this report were COL Alex G. Sutton, Jr., CE, COL John R. Oswalt, Jr., CE, and
COL Levi A. Brown, CE.
Technical Directors were Messrs. J. B. Tiffany and F. R. Brown.
Ill
CONTENTS
'
Page
FOREWORD
111
NOTATION
Vll
..
CONVERSION FACTORS, BRITISH TO METRIC UNITS OF MEASUREMENT .
SUMMARY
PART 1:
.
IX
XI
INTRODUCTION
1
Background . . . .
Purpose and Scope of Study
1
PART II:
EXPERIMENTAL FAC ILITIES
Description
.
Scale Relations
PART Ill:
2
.
.
.
.
5
.
.
5
5
TESTS AND RESULTS
8
Tailwater Conditions . . . .
Scour Hole Geometry . . . .
Riprap Protection Requirements
Effect of Culvert Shape on Scour Hole Geometry
PART IV:
DISCUSSION AND DESIGN RECOMMENDATIONS
8
8
11
13
14
LITERATURE CITED
18
SELECTED BIBLIOGRAPHY
18
TABLES 1 and 2
PHOTOGRAPHS 1-4
PLATES 1-20
v
NOTATION
A
Cross-sectional area of flow m a channel, ft2
B
Channel bottom width, ft
ds
Stone diameter required to prevent failure, ft
D
Depth of flow in channel, ft
00
Culvert diameter, ft
Ds
Depth of scour, ft
Dsm
Maximum depth of scour, ft
Fch
Froude number of flow in a channel, Fch
F0
g
Ls
Froude number of flow at the culvert
Acceleration due to gravity, ft/sec2
Length of scour, ft
Lsm
Maximum length of scour, ft
Lsp
Length of stone protection, ft
n
Manning's roughness coefficient
Q
Channel discharge, cfs
Uc
Discharge at which stone was displaced on the nprap blanket, cfs
S
Channel bottom slope, ft/ft
t
Test duration, min
T
Qj'\) gA3jT
outlet, F 0 = V 0/~
=
Top width of flow in a channel, ft
Vs
Volume of scour, ft3
V0
Average velocity of flow at the culvert outlet, fps
W5
Width of scour, ft
Wsm
Maximutn width of scour, ft
..
Vll
CONVERSION FACTORS, BRITISH TO METRIC UNITS OF MEASUREMENT
British units of measurement used in this report can be converted to metric units as follows:
Multiply
By
To
- - ---
Obtain
inches
2.54
centimeters
feet
0.3048
meters
square feet
0.092903
square meters
cubic feet
0.0283168
cubic meters
feet per second
0.3048
meters per second
cubic feet per second
0.0283168
cubic meters per second
feet per second per second
0.3048
meters per second per second
.
IX
SUMMARY
Investigations were conducted in experimental facilities to determine the characteristics of
scour below a culvert outlet discharging onto a horizontal blanket of cohesionless soil and to develop guidance for designing riprap protection downstream of culvert and storm-drain outlets.
Scour holes, produced by the discharge of various flows of several durations through culverts of
various shapes and sizes, were observed and contoured.
These data were obtained for several tail-
water conditions and were used to develop generalized expressions describing the maximum length,
width, depth, and volume of scour as functions of the flow duration and the Froude number of
flow at the culvert outlet.
Dimensionless scour profiles and cross sections, for low and high tail-
water conditions, were also developed.
Riprap blankets of various configurations and stone stzes were tested with several culvert
stzes and shapes and various tailwater conditions to determine when displacement or failure occurred.
The results were used to develop generalized relations to describe the stone size and
appropriate blanket configuration required to prevent blanket failure and soil erosion, respectively,
at a culvert or storm-drain outfall.
.
Xl
EROSION AND RIPRAP REQUIREMENTS AT CULVERT
AND STORM-DRAIN OUTLETS
Hydraulic Laboratory Investigation
PART I:
INTRODUCTION
BACKGROUND
1.
The lack of adequate means for estimating the extent of erosion to be anticipated
downstream of culvert and storm-drain outlets and the size of stone and extent of a riprap blanket sufficient to prevent such erosion can be readily verified by a cursory inspection of field installations.
Rather extensive and noteworthy surveys of conditions at culvert outlets have been
accomplished by Keeleyl in Oklahoma and Scheer 2 in Montana. During the course of the subject
model investigations, observations were made of the drainage and erosion control facilities of over
SO military installations throughout the continental United States in connection with a separate
but related study concerning field performance of drainage and erosion control facilities sponsored
by the Office, Chief of Engineers. The conditions of erosion observed throughout the field performance study support the observation reported by Keeley 1 that two general types of channel
instability can develop downstream from a culvert and/or storm-drain outlet, i.e. either gully scour
or a localized erosion referred to as a scour hole.
Distinction between the two conditions of
scour and prediction of the type to be anticipated for a given field condition can be made by a
comparison of the original or existing slope of the channel or drainage basin downstream of the
outlet relative to that required for stability as illustrated in fig. 1.
ORIGINAL GROUND (STABLE SLOPE)
~--.,..--rt--- - - - - - - - - - - : : : - - - - - - - - - - - - - -
SCOUR HOLE
h
FLOW.._
STABLE
SLOPE
GULLY SCOUR
Fig. 1.
Types of scour at culvert outlets
1
2.
Gully scour is to be expected when the Froude number of flow in the channel exceeds
that required for stability.
and progresses upstream.
It begins at a control point downstream where the channel is stable
If sufficient differential in elevation exists between the outlet and the
section of stable channel, the outlet structure will be completely undermined as shown in fig. 2.
'
--
"
-t\
..
Fig. 2.
Failure of outlet structure due to gully scour
Erosion of this type may be of considerable extent depending upon the location of the stable
channel section relative to that of the outlet in both the vertical and downstream directions.
3.
A scour hole or localized erosion is to be expected downstream of an outlet even if
the downstream channel is stable as shown in fig. 3a.
The severity of damage to be anticipated
depends upon the conditions existing or created at the outlet.
In some instances, the extent of
the scour hole may be insufficient to produce either instability of the embankment or structural
damage to the outlet.
However, in many situations flow conditions produce scour of the extent
that embankment erosion (fig. 3a) as well as structural damage of the apron, end wall, and culvert
(fig. 3b) is evident.
4.
The observations and empirical methods developed by Keeley,l,3,4 which provide spe-
cific guidance relative to the conditions that produce gully scour or only a localized scour hole
as well as those required for stable channels in several Oklahoma soils, merit consideration and
application in general.
However, there still is a lack of guidance relative to the extent of the
scour hole to be anticipated downstream of culvert and storm-drain outlets.
for initiation of the subject study.
This is the basis
PURPOSE AND SCOPE OF STUDY
5.
The research reported herein has as its purpose the development of guidance for esti-
mating the extent of localized scour to be anticipated downstream of culvert and storm-drain
2
a.
Embankment instability
'
•
...
•
b.
Structural damage of apron, end wall, and culvert
Fig. 3.
Damage resulting from localized erosion
•
3
outlets as well as protective stone requirements sufficient to prevent such erosion.
A hydraulic
model was used to determine (a) the geometry of the scour hole formed by the discharge of
various rates and durations of flow from various sizes and shapes of culverts onto a horizontal
bed of sand under various depths of tailwater, (b) the conditions of discharge and tailwater
through and downstream of these culverts required to displace various sizes of crushed stone, and
(c) the configuration and extent of riprap blanket required
ditions.
to
prevent erosion under similar con-
Test results for various tailwater conditions were analyzed and are presented in terms of
dimensionless parameters involving the average velocity of flow and diameter of the outlet so that
the design engineer can readily and easily predict the geometry of the scour hole to be expected
and/or the size of stone and extent of riprap blanket required to prevent localized erosion.
4
PART II:
EXPERIMENTAL FACILITIES
DESCRIPTION
6.
Circular culverts with diameters of 0.125, 0.224, 0.333, and 1.000 ft* located at the
upstream end and center of a 16-ft-wide, 5.5-ft-deep, and 40-ft-long test flume (fig. 4) were used
t-----;;::::
------f
-
-
-
-
..
-
'
~
ION2
I•
40'
PLAN
~
~
ClJLVERT
BAFFLE- ._
:i:
.::·
s
'
II\
")
A
N
D ':;[; :,
i:I:
=·~;:;~;
·;_-·.::.
I ..
BRICK
END
SILL
);AILGATE
<ri
r--
•I
20'
'
II\
..A:
DRAIN
ELEVATION
Fig. 4.
Experimental facilities
to investigate scour downstream of a culvert outlet.
Circular, arch, rectangular, and box-shaped
sections of equal areas (cross-sectional area equal to that of the 0. 3 3 3-ft-diam circular culvert)
were used to investigate the effect of shape upon the scour characteristics.
7.
A horizontal sand bed, 3.5 ft deep, was provided for a length of 20 ft downstream of
the culvert.
Gradation of the material used for the sand bed was as shown in fig. 5.
The nprap
tests were conducted using uniformly graded crushed limestone in sizes ranging from 114- to 6-in.
diameters.
The stone was kept uniformly graded for ease in handling and storage.
was placed between the sand and riprap to prevent slumping of the riprap blanket.
A filter cloth
Standard de-
vices were used to measure discharge, velocity, and water- and soil-surface elevations.
SCALE RELATIONS
8.
*
In order to obtain some idea of the model scale effects, the 1.000-ft-diam culvert was
A table of factors for converting British units of measurement to metric units is presented on page ix.
5
U. S. STANDARD SIEVE SIZE
NO. 4
100
-~
NO. 10
NO. 40
I
I
I
I
I
I
i
!
eo
i
I
'
I
-w
>-
I
60
I
!
0:
w
I
I
I
I
I
lL.
40
I
I
ll.
I
20
I
I
I
I
I
I
I
i
I
I
I
I
I
li
I
I
I
I
I
I
I
i
I
0
10
6
I
ti
I
1:
I
I
I
I
I'
I
I
I
'I
!
i
II
I
I
\
I
I
I
I
II
II
I
I
II
I
I
I
I
I
I
I
I
I
I
I
[I
i
'
I
I
I
I
I
I'
w
u
0:
w
I
I
I
-z
z
It
It
I
aJ
1-
I
I
I
1J:
C>
~
I
I
\
NO. 200
I
I
4
2
1 .0
0.6
0. 4
0.2
I
I
I
\
I
I
It
0. 1
0.06
0.02
0.0 1
G RAIN SI Z E IN M ILLI M ETERS
Fig. 5.
Grain-size distribution of bed material
considered the prototype and the 0.3 3 3-, 0.224-, and 0.125-ft-diam culverts were referred to as
1 :3-, 1:4.46-, and 1 :8-scale models, respectively.
Th e 1:4.46 scale resulted from the fact that com-
mercially available tubing was used for the 0 .224-ft-diam culvert.
9.
The accepted equations of h ydraulic similitude , based upon the Froudian criteria, were
used to ex press mathematical relations between dimensio ns and h ydraulic quantities of the models
and prototy pe.
follo ws:
General relations for t ransference of model data to prototype equivalents are as
Dimension
10.
Ratio
Scale Relations
Length
Lr
1 :3
1:4.46
1:8
Area
Ar - L2r
1:9
1:19.89
1:64
Velocity
vr
-
Ll/2
r
1 :1.732
1:2.112
1:2.828
Discharge
<lr -
L5/2
r
1 : 15.6
1:42
1:181
The experimental data indicated that the proto type-t o-model scale ratws for the 0.3 3 3-
and 0.224-ft-diam culverts were valid; however, the data for the 0.125-ft-diam culvert did not correlate using the above relations.
For example, displacement of 1/ 2- to 3/4-in. stone downstream
of the 0.125-ft-diam culvert that simulated 5- to 6-in. st one d o wnstream of the 1.000-ft-diam
6
prototype culvert to a scale of 1:8 required simulated discharges 1.25 times those observed with
the prototype. The Reynolds number of flow at the culvert outlets for all of the circular culverts was calculated and was found to be 1 X 105 or greater for all tests of the 1.000-, 0.3 3 3-,
and 0.224-ft culverts. The Reynolds number of flow in tests of the 0.125-ft culvert was found
to be less than 1 X 105. It was concluded that the viscous effects were significant and that the
Reynolds number of flow in the model culverts should be of sufficient magnitude (apparently
:2: 1 X 105) to produce a resistance coefficient of the magnitude anticipated and attributed to
the prototype stone in a fully rough flow regime. This was accomplished with all but the 0.125ft-diam culvert, and therefore the unconservative results of tests obtained with the smallest (1 :8scale model) culvert were disregarded.
7
PART III:
TESTS AND RESULTS
TAILWATER CONDITIONS
11.
Tests indicated that all of the tailwater conditions investigated could be grouped into
two categones. Tailwater conditions of less than 0.5 0 0 above the culvert invert produced approximately the same flow pattern and scour hole geometry and will be referred to as minimum
tailwater conditions in this report; all tailwater conditions of 0.5 D0 ft and greater above the
culvert invert produced approximately the same flow pattern and scour hole geometry and will
be referred to as maximum tailwater conditions. Variations in the pattern of scour observed
with various tailwaters are shown in photograph 1.
12. These results agreed very well with those presented by Seaburn and Laushey5 which
indicate that for a constant discharge the velocity just downstream of a culvert outlet remains
constant for tailwater conditions from 0 to 0.5 D0 ft above the culvert invert. The velocity increases with increasing tailwater and reaches a constant maximum velocity again at a tailwater
approximately 1.0 D0 ft above the culvert invert.
SCOUR HOLE GEOMETRY
Test Procedure
13.
The sand scour tests were initiated usmg a 1.000-ft-diam culvert and a horizontal sand
bed molded to the invert of the culvert (fig. 6).
Fig. 6.
Several different discharges were passed through
Horizontal sand bed
8
the 1.000-ft-diam culvert, and it was observed that the scour hole extended to the model limits
for the higher discharges. For this reason and to check model scale effects, three smaller culverts
(0.333-, 0.224-, and 0.125-ft diam) were tested. A low flow and corresponding scour hole are
shown in photograph 2.
14.
The first phase of the sand scour investigation consisted of testing all of the circu-
lar culverts with several different discharges and test durations of 20 and 30 min for minimum
and maximum tailwater conditions, respectively. Typical flows and resulting scour holes for minimum and maximum tailwaters are shown in photograph 3. The test durations were arrived at by
conducting several tests with times varying up to 1 hour and observing that most of the scour
took place within the first 20 min of each test for minimum tailwater and the first 30 min of
each test for maximum tailwater.
15.
The scour hole was contoured after each test.
These data were used to obtain the
volume of scour and the maximum length, width, and depth of scour as shown in table 1.
After the scour hole was contoured, the sand bed was remolded prior to the next test.
16. The second phase of the sand scour investigation was conducted to determine the
effect of the duration of the test on the scour hole geometry. These tests were conducted
using the 0. 3 3 3-ft-diam culvert.
Two series of tests, with Froude numbers of flow at the cul-
vert outlet equal to 1.43 and 3.05, were conducted with minimum tailwater and a third test
series, with a Froude number at the culvert outlet of 3.05, was conducted with maximum tailwater.
During each test series the flow was stopped at various intervals and the scour hole con-
toured. The volume of scour and the maximum length, width, and depth of the scour hole were
obtained and are tabulated in table 1. The total duration of each of the test series was approximately 24 hours.
As indicated by the tabulated data, a stable condition was not reached
in this time period.
Data Analysis
17. Several attempts were made to generalize the data using dimensionless parameters.
The
plots which best satisfy the data are those of the Froude number versus the ratios of maximum
length of scour to the culvert diameter
diameter
Wsm/D0
,
Lsm/D 0
,
maximum width of scour to the culvert
maximum depth of scour to the culvert diameter
Dsm/00
,
and the vol-
Vs/D~ . The test duration data were best satisfied by plots of these same ratios versus the duration of the tests. These plots are shown in
ume of scour to the culvert diameter cubed
plates 1-5 for minimum tailwater conditions and in plates 7-11 for maximum tailwater conditions and will henceforth be referred to as the Froude number relations and the test duration
relations.
18.
Further analysis indicated that the Froude number relations and corresponding test
duration relations could be combined, yielding the following equations:
Minimum tailwater:
(1)
(2)
(3)
9
(4)
Maximum tailwater:
(5)
(6)
(7)
(8)
The technique used to develop these equations will be demonstrated with the graphs shown in
plate 3.
19. The test duration relation indicates that the data for Froude numbers of 1.43 and
3.05 are best fit by straight, parallel lines. It was assumed that the best fit for any other
Froude number would also be a straight line parallel to those for F 0 = 1.43 and 3.05 . The
equation of a straight line on this plot is
where a is the slope of the line and b is the Dsm/D 0 intercept at
the lines on this plot will be parallel, the slope is the same for every line.
t
= 1 .
Since all of
However, the
D5 m/D0 mtercept at t = 1 is variable and depends upon the Froude number. The Froude
number relation indicates that Dsm/D0 varies as the logarithm of the Froude number. The
curve shown on the upper plot of plate 3 is for a test duration of 20 min and the curve required is for a duration of I min.
The 1-min curve was obtained as follows:
Two points on the Froude number relation were selected (F0 = 1 and F 0 = 10)
and plotted on the test duration relation at t = 20 min. Since all curves on this
plot will be parallel, lines were drawn through the two plotted points, parallel to
the existing lines and through t = 1 min. At this point (t = 1 min) the Dsm1D 0
values were observed and plotted on the upper plot of plate 3 at F 0 = 1 and 10 .
A straight line was drawn between these two points and the equation of this line is
- 1.3 + 1.4 log F 0
This equation shows how Dsm/D 0 varies with the Froude number at
can be substituted into the equation
for
b , yielding,
10
t - 1 min
and therefore
The test duration plot indicates that
a - 0.1 .
The final equation is
0 sm
Do
or
(3 bis)
This same procedure was used to obtain equations 1 to 8.
These relations can be used to cal-
culate the expected length, width, depth, and volume of a scour hole for a known discharge, culvert size, flow duration, and tailwater condition.
20.
The next objective in the sand scour analysis was to develop a dimensionless scour
hole using the contour data.
A dimensionless profile along the center line of the culvert was
first constructed by plotting the ratio of the depth of scour to the maximum depth of scour
DsiDsm
versus the ratio of the length of scour to the maximum length of scour
Ls/Lsm .
This was done for both minimum (plate 6) and maximum (plate 12) tailwater conditions.
These
two plots indicated that the maximum depth of scour occurred at a distance 0.4 of the maximum
length of scour downstream of the culvert outlet for both tailwater conditions.
Dimensionless
cross sections were then plotted at this location for both tailwater conditions and are shown
in plates 6 and 12 for minimum and maximum tailwater, respectively.
The data satisfy these
plots quite well, especially considering the fact that the data were gathered by several different
individuals.
RIPRAP PROTECTION REQUIREMENTS
Types of Failure
21. Three types of nprap failure were considered:
(a) individual displacement of stones
on the blanket caused by inadequate stone size, (b) raveling of the ends of the blanket caused
by inadequate blanket size and scour in excess of the blanket thickness, and (c) movement of
material from below the blanket allowing the blanket to settle.
This latter type of failure is
caused by an inadequate filter below the blanket.
22. The first two types of failure have been investigated and the results are presented
below.
The third type of failure, blanket slumping, was not investigated but was prevented
during the testing by using a filter doth beneath the riprap blanket.
Stone Size
23. Tests were conducted to determine the stone size required to prevent failure (ds) or
displacement for various culvert sizes, discharges, and tailwater conditions.
The first problem in
this series of tests was to define this failure so that some type of standard could be referred
to for each test.
It was decided that the condition at which several stones were displaced from
the upper layer of the two-stone-diameter-thick blanket would be termed failure.
In several
cases only one or two stones were displaced and no other stone movement occurred.
was not considered failure.
was displaced.
This
On the other hand, in some cases a major portion of the blanket
The conditions termed failure herein did not permit displacement of the entire
11
blanket thickness for it is considered that it would be appropriate to be conservative in such a
laboratory investigation, smce a greater degree of turbulence is to be anticipated in the prototype
or field installations.
24. The tests that were conducted resulted in a list of stone sizes and corresponding discharges at which failure occurred for the different culvert sizes and tailwater conditions (see
table 2). Plots of the ratio of the stone diameter to the culvert diameter versus the Froude
number of flow at the culvert outlet for both minimum (plate 13) and maximum (plate 16)
tailwater conditions resulted in the best representation of the data. The relations obtained from
these plots are:
Minimum tailwater:
(9)
and
Maximum tailwater:
(10)
The stone size, therefore, depends upon the culvert diameter, Froude number of flow at the culvert outlet, and the tailwater condition.
Blanket Configuration
2 5. An appropriate rip rap blanket configuration was arrived at by a trial-and-error procedure. Several different configurations were tested, using the scour hole geometry data as a
guide and engineering judgment to determine whether the blanket was adequate. Once again a
standard had to be developed to use as a guide in determining whether a blanket was adequate.
The standard used was as follows: if the scour beyond the edges of a blanket was two or less
stone diameters deep, the blanket was adequate; if this scour depth exceeded two stone diameters, the blanket was inadequate. Tests of 20-min duration were run for several different
discharges and tailwater conditions.
26.
The blanket configurations were quite different for mm1mum and maximum tailwater
conditions. The jet from the culvert dispersed rapidly with minimum tailwater and required a
short, wide blanket; however, with maximum tailwater, the jet traveled a considerable distance
downstream and required a long, narrow blanket to prevent erosion.
With maximum tailwater
the jet also changed position from side to side, as shown in photograph 4.
27.
Tests indicated that the only parameter of the blanket configuration that changed sig-
nificantly with discharge was the blanket length.
A blanket width of three culvert diameters was
required at the culvert outlet for all conditions tested.
A flare of 1 on 2 (plate 15) was re-
quired to contain all discharges with minimum tailwater, while a 1-on-5 flare (plate 18) was
sufficient to contain all flows with maximum tailwater.
28.
For each test conducted, the blanket length was classified as adequate or inadequate
and the results were plotted with the blanket length, expressed as a number of culvert diameters,
12
versus the Froude number of flow at the culvert outlet for both tailwater conditions. The
adequate and inadequate blanket lengths can be divided quite well with the lines shown in plate
14 for minimum tailwater and plate 17 for maximum tailwater. The equations of these lines,
which define the required length of blanket to prevent raveling as defined in paragraph 21, are:
Minimum tailwater:
(11)
Maximum tailwater:
Lsp
=
D0 (8 + 55 log F 0 )
These relations are valid for Froude numbers greater than one.
(12)
A limited number of tests were
conducted with Froude numbers less than one and these indicated that a blanket length of approximately eight culvert diameters should be provided to contain the flow for both tailwater
conditions with Froude numbers less than one.
EFFECT OF CULVERT SHAPE ON SCOUR HOLE GEOMETRY
29.
The effect of culvert shape on the scour hole geometry was not investigated in detail.
A quick comparison was made by passing a discharge of 0.87 cfs through four differently shaped
culverts (circular, square, rectangular, and arch) with the same cross-sectional area (0.087 sq ft)
and by observing the scour hole. This was done for both minimum and maximum tailwater conditions, and the results indicated that the culvert shape had no significant effect on the scour
hole geometry.
30. The oscillating jet that was observed for flow from the circular culvert into maximum
tailwater (photograph 4) was also observed with flow from the square, rectangular, and arch
culverts. This oscillation was random and quite slow for all conditions except that in which
flow passed through the arch culvert into maximum tailwater after a scour hole had been
developed with minimum tailwater.
position about every 15 sec.
For this condition the oscillation was periodic and changed
Thus, it appears that the jet discharged from an arch culvert is less
stable than that from the other culverts investigated.
13
PART IV:
31.
DISCUSSION AND DESIGN RECOMMENDATIONS
In designing protection against erosion below culvert outlets, one must first know what
type of erosion to expect.
previously in this report.
Keeley 1 classifies erosion as gully scour or scour hole, as discussed
He states that gully scour may be expected in a cohensionless soil if
the Froude number of flow in the channel below a culvert outlet is 0.35 or greater.
This value
varies for different types of soil.
32. If one determines that gully scour will occur, the Froude number of flow in the
channel must be decreased.
This can be best accomplished by using one or more drop structures
to achieve considerable change in elevation and maintain a stable slope to the downstream conNo type of stilling device or blanket protection will prevent gully scour if the
trol point.
Froude number of flow in the downstream channel exceeds the critical value.
As indicated by
Keeley,! the ultimate depth of channel degradation to be anticipated at the outlet can be approximated.
The elevation of the invert of the control section (upstream limit of the stable
channel downstream) plus the product of the stable gradient and the distance from the control
section to the outlet will yield the elevation of the degraded channel to be anticipated at the
outlet.
This elevation compared with that existing at the outlet is a measure of the scour depth
which may ultimately develop.
3 3. Even if the Froude number of flow in a channel below a culvert outlet is less than
'
the critical value required for gully scour, a scour hole can be expected to develop.
The results
presented in this report can be used to determine the size of the scour hole expected for either
minimum or maximum tailwater conditions.
A decision must be made either to allow all or a
portion of this scour hole to form and provide the necessary structural protection or to prevent
erosion entirely.
34. If the location of the culvert is such that a scour hole is not objectionable, allowing
the scour hole to form may be a very economical solution to controlling erosion.
The scour
hole acts as an excellent energy dissipator; however, the culvert still must be protected against
undermining.
A cutoff wall will normally be used for this purpose and the dimensionless scour
holes presented in this report can be used to estimate the required depth to which a cutoff wall
must extend to prevent undermining.
The dimensionless scour hole for minimum tailwater, shown
in plate 6, indicates that erosion directly under a culvert outlet would be approximately 0. 7 as
deep as the maximum expected scour depth.
Thus it would be necessary to extend a cutoff wall
down to, at least, this depth to prevent undermining.
35.
A "plunge pool" (fig. 7) is another effective device which can be used to prevent
undermining of a culvert.
The plunge pool should be as deep and wide as the maximum ex-
pected depth and width of scour with the upstream slope protected with riprap.
This type of
protection requires much less riprap than a full blanket and will prevent erosion only near the
culvert (fig. 8).
36.
If a scour hole is undesirable, complete protection can be provided with a riprap blan-
ket, concrete apron, or a combination of these.
The blanket or apron configuration required
to provide protection can be determined with the technique outlined in this report.
The choice
between a riprap blanket and a concrete apron can be made after determining the stone size required for a riprap blanket.
The required stone size can be determined from equations 9 and
10 for minimum and maximum tailwater, respectively.
14
Examination of equation 9 indicates
Fig. 7.
Plunge pool or riprapped scour hole
Fig. 8.
Eroded plunge pool
that the stone size required for a Froude number of 2 is half the diameter of the culvert. Of
course, as the culvert size increases the stone size becomes unrealistically large and impractical,
in which case a concrete apron may be desirable with the same configuration as that required
for the riprap blanket. It would be uneconomical to extend the concrete apron or a riprap
blanket with the stone size determined from equations 9 or 10 the full length of the required
protection, since the degree of attack becomes less severe downstream. Thus, it is advisable to
gradually decrease the protection in a downstream direction. The dimensionless scour holes discussed herein can be used as a guide for determining the degree of attack along a channel.
These dimensionless scour holes (plates 6 and 12) indicate that the area of the most severe
attack is from the culvert outlet to a point approximately 0.6 or 0. 7 of the maximum expected
length of scour downstream; therefore, protection can be decreased downstream of this area. For
example, if a concrete apron is used below a culvert, grouted riprap can be placed beyond the
area of the most severe attack. A transition of loose riprap, decreasing in size in a downstream
direction, can be used between the grouted riprap and the natural channel. If a riprap blanket
is used, the stone size downstream of the area of the most severe attack may gradually decrease in size downstream to the end of the blanket.
37. For conditions in which the stone size of a nprap blanket is unrealistic or impractical and a concrete apron is exceedingly large, a stilling device may be the most economical
means of providing protection. Some commonly used energy dissipators are the St. Anthony Falls
stilling basin,6 the stilling well,7 and the U. S. Bureau of Reclamation Type VI stilling basin.8 A
laboratory study is presently getting under way at the U. S. Army Engineer Waterways Experiment
Station to develop simple stilling devices to be used at culvert outfalls for intermediate flow conditions where protection other than riprap and/or the most commonly used energy dissipators is
required.
38.
Design of culvert outlets, subjected to both mtmmum and maximum tailwater condi-
tions, requires good engineering judgment.
One suggestion of a sufficient blanket configuration
for this condition follows. The width of the blanket at the culvert outlet should be 30 0
wide and should flare 1 on 2 until the total width is equal to the greater of either the blanket
width required for minimum tailwater or the blanket width required for maximum tailwater.
From this point the blanket should extend downstream parallel to the culvert to a length equal
to the blanket length required for maximum tail water. This is illustrated in plate 19.
39. One very important factor of drainage design is deposition in the channel and culvert.
This condition may be a more severe problem than erosion, because the capacity of the channel
or culvert is reduced and maintenance (effort and cost) is required. The problem of deposition
can be reduced by providing more .stable channels, thus reducing erosion upstream of a drainage
facility. Keeley4 has observed some prototype drainage facilities in Oklahoma at which deposition has occurred. He suggests that deposition will occur if the Froude number in a channel is
less than a certain critical value, depending upon the sail type. Again, it is considered that the
results of Keeley 1,3 ,4 can be applied to design drainage channels in cohesionless soils such that
a Froude number of flow from about 0.15 to 0.35 is maintained and the channel downstream of
an outlet will not be subject to either deposition or erosion.
The graph shown in plate 20 was
developed (using the Manning formula) to provide guidance in selecting a satisfactory channel.
The parameters for a given channel may be calculated and plotted to determine whether the
Froude number for the channel is in the acceptable range. The plot shown is for a channel
16
side slope of 1 on 2. The same type of graph can be developed for other side slopes.
40. The results involving depth, width, length, and volume of scour; stone diameter;
outlet diameter; average velocity of flow; and tailwater depth were purposely presented in
terms of dimensionless parameters so that they can be readily and easily applied to predict
the geometry of the scour hole to be expected downstream of an outlet and to select the
size of stone and extent of riprap blanket required to prevent development of a scour hole.
judgment can and must be exercised to evall;late the need for and/or degree of protection
required at an outlet, such as no protection, prefabricated end section, end wall, small apron
and cutoff wall, protective stone blanket, grouted riprap blanket, concrete apron, or an energy
dissipater with either a protective stone blanket or concrete apron. Further, the outlet with
or without appurtenances must be provided with a stable channel downstream to prevent the
likelihood of gully scour and structural damage.
17
LITERATURE CITED
1.
2.
Keeley, J. W., "Soil Erosion Studies in Oklahoma; Part III, Culvert Oudet Conditions and
Downstream Channel Stability," 1963, U. S. Bureau of Public Roads, Oklahoma Division,
Oklahoma City, Okla.
Scheer, A. C., "Large Culvert Studies in Montana," 1968, Department of Civil Engineering
and Engineering Mechanics, Montana State University, Missoula, Mont.
3.
Keeley, J. W., "Soil Erosion Studies in Oklahoma; Part I, Water Erosion in Narrow Ditches
and Channels; Part II, Erosion Control Devices for Ditches and Channels," 1961, U. S.
Bureau of Public Roads, Oklahoma Division, Oklahoma City, Okla.
4.
, "Soil Sedimentation Studies in Oklahoma; Deposition in Culverts and
Channels," 1967, Federal Highway Administration, Bureau of Public Roads, Oklahoma Division, Oklahoma City, Okla.
5.
Seaburn, G. E. and Laushey, L. M., "Velocity of Culvert Jets for Incipient Erosion,"
Proceedings, Twelfth Congress of the International Association for Hydraulic Research, Vol 3,
Sept 1967, pp 1-8.
6.
7.
8.
Blaisdell, F. W., "Development and Hydraulic Design, Saint Anthony Falls Stilling Basin,"
Transactions, American Society of Civil Engineers, Vol 113, Paper No. 2342, 1948,
pp 483-520.
U. S. Army Engineer Waterways Experiment Station, CE, "Impact-Type Energy Dissipator
for Storm-Drainage Outfalls, Stilling Well Design," Technical Report No. 2-620, Mar 1963,
Vicksburg, Miss.
Beichley, G. L., "Progress Report No. XIII-Research Study on Stilling Basins, Energy Dissipators, and Associated Appurtenances-Section 14, Modification of Section 6 (Stilling
Basin for Pipe or Open Channel Outlets-Basin VI),'' Report No. HYD-572, june 1969,
Hydraulics Branch, Division of Research, U. S. Bureau of Reclamation, Denver, Colo.
SELECTED BIBLIOGRAPHY
Bradley, J. N. and Peterka, A. ]., "Hydraulic Design of Stilling Basins: Small Basins for Pipe
or Open Channel Outlets-No Tailwater Required (Basin VI)," Journal, Hydraulics Dil•ision,
American Society of Civil Engineers, Vol 83, No. I-IYS, 1957, pp 1406-1 - 1406-17.
Laushey, L. M., Ulrich, K., and Ofwona, M. P. , "Magnitude and Rate of Erosion at Culvert
Outlets," Proceedings, Twelfth Congress of the International Association for Hydraulics
Research, Vol 3, Sept 1967, pp 338-345.
Stevens, M. S., Scour in Riprap at Culvert Outlets, Ph. D. Dissertation, Colorado State
University, Fort Collins, Colo., Jan 1969.
18
Table 1
Summary of Sand St:our Tests
Culvert
Diameter
D0 , ft
Tail water
Condition
Do
Test
Uurauon
t , min
Dtscharge
Q , cfs
l•roude
Number at
Cuhert
Outlet, F0
Maxtmum
Length of
Scour
l"Sm • ft
Maxtmum
\\ tdth of
Scour
\\sm . ft
Maxtmum
Depth of
Scour
Dsm , ft
.us
\ olume
of cour
\'s ' ft3
Test Durations, 20 and 30 mm
1.000
20
0
0.333
0
20
4.710
5.500
6.280
7 070
7.850
9.420
11.800
1.06
1.23
1.41
1.59
1.76
2.12
2.64
13 .00
14 00
16 .00
17.00
17.00
20.00
EM I.
S.10
8.00
E.\11.
2.40
2.50
2.85
2.90
3.05
3.00
0.3 51
0.401
0.452
0.503
0.870
1.30S
1.740
1.23
I .41
1.58
1.76
3.05
4.58
6.10
4.00
3 so
4.00
5.00
7.50
10.00
12.00
3.30
2.80
2.50
3.50
5.50
8.00
8.00
0,68
0.70
0 .69
0.70
0.90
1.0!1
1.11
so.ss
F.\\1 . ·
EM I.
1..\11.
1.\\L
92.00
118.00
I S4.00
211\.00
260 00
3 76.00
465 .00
2.40
2.76
3.14
4 .H8
12.33
34.40
0.224
0
20
0.112
0.131
0.150
0.168
0.187
0.224
0.280
1.06
1.24
1.41
1.59
1.77
2.12
2.65
2.20
2.50
2.70
2.80
3.30
4.00
4 .50
1.10
1.25
1.10
1.15
1.40
1.90
2.15
0 .-1-8
0 .50
0.50
0.53
0.57
0.58
0.63
() 71
0.98
0 94
1.22
1.69
2.5S
3.32
0.333
1.0
30
0.410
0.450
0.870
1.300
1.740
1.-H
1.58
3.05
4.58
6.10
6.50
7.00
13 00
15.50
15.20
3 00
2 so
3.50
3.50
8.00
0.64
0.63
0.92
1.01
S.H
0.112
0.131
0.150
0.327
1.06
1.24
1.42
3.10
3.30
3 70
4.30
7.00
060
0.80
0.90
1.55
0.26
0.39
0.43
0.70
5.20
14.64
22.12
55.69
().28
0.67
0.89
4. 11
0.224
1.0
30
l.H
Extended Test Duration with 0.333 -ft·d•am Culvert
0.33 3
0
10
15
20
25
30
45
60
90
120
180
360
0.41
1.43
3.50
3.50
3.60
3.15
4.00
4.10
4 50
4.SO
5.00
5.00
5 50
9 00
1000
10 00
1.93
2.00
2.00
2.00
2.20
2 50
3 10
3.00
2.50
3.50
4.00
4.50
4.50
5.50
0.70
0.70
0.70
0.74
0.73
0.79
0.80
0.73
0.84
0.70
0.91
0.82
1.19
1.18
2.28
2.47
2.76
2.92
3.18
3.64
3. 70
4.23
4.58
5.62
6.57
9.08
12 89
14.48
0.87
3.05
7.00
7.50
7.50
7.50
8.00
10.00
12.00
18.00
18.00
18.00
4.00
5.00
5.50
5 50
5.50
8.00
0.84
0.88
0.90
0.94
1.00
1.06
1.22
1.44
1.47
1.55
I 0.42
12.19
12.33
I 4 .1 1
16.H
24 79
42.00
64 77
71.59
81.94
0.82
0.90
0.96
1.03
1.16
1.19
1.20
1.29
8.50
14.04
18.13
23.72
28.43
33.27
32.95
37.61
HU
900
1260
0
0.333
0.333
1.0
10
15
20
30
45
120
360
720
1020
1425
10
30
60
120
360
720
1080
1-HO
•
EML
=
exceeded model limits.
0.87
3.05
9.50
11 .50
13 so
14 50
15 50
16 so
15.50
16 50
EML
EML
EML
1:..\\L
3.50
3.00
3.00
4.50
4.00
4.00
4.00
~.00
Table 2
Stone Size Data
D0
,
ft
d5
,
ft
<lc , cfs
Tailwater Condition, 0 D0
0 .224
0.0625
0.0833
0.1250
0.1667
0.111
0.194
0.201
0.400
1.32
1.83
1.93
3.78
0.3 3 3
0.0833
0.1250
0.1667
0.2500
0.33 33
0.360
0.510
0.710
0.825
1.160
1.26
1. 79
2.49
2.90
4.06
1.000
0.0625
0.0833
0.1250
0.1667
0.2500
0.3333
0.4167
0.5000
0.240
0.790
1.700
3.210
4.090
7.470
8.980
9.100
0.53
0. 75
0.93
1.10
1.15
1.68
2.02
2.04
Tailwater Condition, 1.0 D 0
0.224
0.0625
0.0833
0.1250
0.1667
0.207
0.272
0.323
0.523
1.99
2.61
3.10
4.93
0.3 3 3
0 .083 3
0.1250
0.1667
0.2500
0.3333
0.580
0.810
0.925
1.000
1.310
2.04
2.84
3.24
3.50
4.59
1.000
0.0625
0.083 3
0.1250
0.1667
0.2500
0 .3333
0.4167
0.5000
4.190
4.600
5.710
7.880
8.380
9.100
10.200
12.100
0.94
1.03
1.28
1.77
1.89
2.04
2.29
2.71
a.
b.
Photograph 1.
Tailwater el 0 D0
Tailwater el 0.25 D 0
Variation of scour with tailwatcr. discharge 0.25 cfs (sheet 1 of 3)
c.
Tailwater el 0.50 D 0
d.
Tailwater el 0.75 D0
Photograph 1.
(sheet 2 of 3)
e.
Tailwater el 1.00 D0
Photograph 1.
(sheet 3 of 3)
•
a.
Flow from culvert, discharge 3.14 cfs
•
•
•
•
• ••• • •
•
•
•
•
•
• ••
• ••
•
'
I
..
•
• •
b.
Photograph 2.
•
•
....
Resulting scour hole
Typical flow condition and resulting scour, 1.000-ft-diam culvert
a.
Flow condition, m1mmum tailwater
b.
Photograph 3.
Scour hole, mm1mum tailwater
Flow condition and resulting scour hole for 0.224-ft-diam culvert,
discharge 0.131 cfs (sheet 1 of 2)
66646
c.
Flow condition, maxtmum tailwater
1
d.
Scour hole, maxtmum tailwarer
Photograph 3.
(sheet 2 of 2)
a.
b.
Flow concentrated at left (time - t)
Flow concentrated at right (time
Photograph ~.
=t
+ ~t)
Oscillating jet with max1mum tailv.:ater
40
I
I
I
I
I
'
I
I
I
l
I
I
~
_........ ~
30
~
~
--...
f--
20
0
0
~_;~
f--
0
10
0
~
~
~
(_--
!.---"
-
~
-
LEGEND
l .000-FT-DIAM CULVERT
0.333-FT-DIAM CULVERT
0.224-FT-DIAM CULVERT
0
0
~
6
-
~
I
I
0
2
I
I
l
I
5
4
FROUDE NUMBER, Fo= Vo/VgDo
3
I
I
7
I
I
B
9
10
100
50
ro== 3.os
0
Lsm
I 0.15
I
Do
b
0
D
fo::: /. 43
u
0
-u
0
1000
l-'
10
[)"
10
0
I!
0
I
0
50
100
500
TEST DURATION , t
1
M IN
5000
I 0 1 0 00
LENGTH OF SCOUR A SA
FUNCTION OF FROUDE NUMBER AND
TIME FOR MINIMUM TAILWATER
lJ
r
40
~
I
fTl
T
I
I
I
'
I
1
r
I
,__
1\)
-
30
~
1-
0
Wsm
Do
o j!:!_----
2 M
't :::: ::.;.-
20
,__
n
10
:
0
~
~
0
4
!::.
~
-
l--""
-
LEGEND
!::.
1.000-FT-DIAM CULVERT
0.333-FT-DIAM CULVERT
0.224-FT-DIAM CULVERT
0
0
!::.
a.
-
I
0
~
~ -
I
I
I
2
4
3
FROUDE NUMBER,F0
60
I
I
6
1
i
I
I
Fo:: 3.05
--,
I
I
'
Wsm
20
- !o.2o
Fo:: /.43
1-
Do
-
-""""
0
0
h
-
-
10
6
I
10
8
I
I
0
I
T
I
I
l
7
I
= V 0 /V9 Do
I
40
I
I
-
0
u
0
....
1000
10,0 00
0
~
50
5000
100
500
TEST DURATION,
t, MIN
WIDTH OF SCOUR AS A
FUNCTION OF FROUDE NUMBER AND
TIME FOR MINIMUM TAILWATER
4
I
l
I
l
I
l
I
I
I
--
1-
In
0
Q
-
"----
3
0
I-
0
6
2
:u
-
6
6
t::: 1 MIN
0
0
0
~
··-
t::: 20 MIN
0
6
6
6
0
0
6
--
0
-
~
I-
Dsm
Do
--- - -
-
I-
-
1.3+1.4 LOG F 0
LEGEND
-
-
--
1.000-FT-DIAM CULVERT _
0.333-F" T-DIAM CULVERT
0.224-FT-DIAM CULVERT
- -- 0
0
6
-
1-
I
I
0
I
3
2
I
I
I
4
FROUOE NUMBER,Fo = V 0
I
I
5
7
6
I
I
8
/.Jg Do
10
+ -- r-
--
-- I -
--
1-
5
----- r- - - f- - - - -- -
--
f.-
--
-f-
,..
-
v
)
-
f.0 :.:1 -
-
1-
·-
-
0
-
- -
-
Fo =~.OS
0
..... - 1-
I
r- r-
-
Fo:: J.43
- f-:_
n
I
0./0
-- -
--
---- --
-
-
-
-- -
1- -
- -
-
+--t-1000
I
10
-
-
.
0
f-
-
v
0
-
0
-
--- ~
1-
-
-
0
-
I
0
n
...:
- f.-
I-
-
- -- 1-- - 1-
u
v
n
-- -
I--
.
-
Do
- --
-
-
t-
F0 =10
Dsm
---
10
9
50
100
500
TEST DURATION, t
1
MIN
5000
10,0 00
DEPTH OF SCOUR AS A
FUNCTION OF FROUDE NUMBER AND
TIME FOR MINIMUM TAILWATER
2000
I
[
I
1000
1/
0
I
800
I
$f_
~,
600
~'J
p/
0
400
0
v
2
1\
I
7
0
6
0
200
0~
0
100
I
6p
80
I
of
6
60
40
I
I
-
LEGEND
I
0
1,000-FT•DIAM CULVERT -
0
0.333-FT·DIAM CULVERT _
6
0 . 224-FT-DIAM CULVERT
20
1
2
FROUDE NO.,
4
F0 :=
V
6
0
/1 go
8
10
0
VOLUME OF SCOUR AS A
FUNCTION OF FROUDE NUMBER FOR
MINIMUM TAILWATER
PLATE 4
10,000
5000
0
p
1000
o5
F; ~ 3- _.
...........
v
v
l?
I-'
....... I--'
/
~.40
v
v
I
0~
..,. I-'.....
500
L.,..../
~
~~
I..- ~
l.--6'
__./
I>
,t3 v
~
100
rr
h
_.jY
v
I-'
~
I..- 1---
p
~
~
v v0
v
.....
...,........ v
v
v-'" 10. 40
~
....... ~
I
0
0
0
../
~
~--"'
50
0
10
50
500
100
TEST DURATION , t. 1 MIN
1000
5000
VOLUME OF SCOUR AS A
FUNCTION Of TIME FOR
MINIMUM TAILWATER
. PLATE 5
FLOW ..,
Lo
_,-ORIGINAL GROUND LIN£
0
..
0.2
0
~ 0
0 0
v
0.4
Ds
Ds m
Oo
I
0.6
o¢
¢
0.8
0
00
6€l
~
,...
oo
0
,...
.......
6
0
0
0
1\
0
rv ......_ ~-....,..,
0
1.0
1.2
J -0
()f\
'"'
0.1
0
~
I
- [..7
/
~
6
6
I
0
6
6
6
0.5
0.4
0.3
0.2
lo _......
v
v
00
/
v
0./
vv
0
c--
0.7
0.6
0.8
0.9
1.0
LsfLsm
DIMENSIONLESS CENTER- LINE PROFILE
Ch
0
u
0.2
(J
6
r--_
0
0
,0.
I
--o....
0
~
i
0.4
~0
6
~ 0.6
Dsm
I
0.8
0
\
~
I
'
v
,___..,.
~
n
0
0
..o-- ~
6
f/
./
1\
1.0
I
rV
II
v
(')
I
0.8
LEGEND
0
0
6
~ [
-
I
0
1.0
€
L--cuLV£RT
~~
1.2
~ORIGINAL GROUND LIN£
1.000-FT-DIAM CULVERT
0.333-FT-DIAM CULVERT
0. 224·FT-DIAM CULVERT
0.6
0. 4
0.2
0
0.2
0. 4
Ws/Wsm
DIMENSIONLESS CROSS SECTION AT 0.4 L 5 m
0.6
0.8
1.0
DIMENSIONLESS
SCOUR HOLE GEOMETRY
FOR MINIMUM TAILWATER
80
I
I
I
I
I
I
I
I
I
-
f-
L---
60
-30~
._
40
~
~
20
0
I"'
~
------
~0
-
LEGEND
t.
0 . 333-fT- DIAM CULVERT
0.224-FT-DIAM CULVERT
0
t.
-
I
0
~
-~
l_..--- f..--
I
I
2
I
I
I
5
4
3
FROUDE NUMBER,Fo = V0
I
I
6
7
I
8
I
10
/V9 Do
100
r
Fo:: 3. 05
50
0
0
0
v
I
0./0
Lsm
Do
.
1000
10
10
50
100
500
TEST DURATION, t
1
MIN
5000
ro,o 00
LENGTH OF SCOUR A SA
FUNCTION OF FROUDE NUMBER AND
TIME FOR MAXIMUM TAILWATER
20
I
T
I
I
T
MIN
f-
10
0
f-
~
5
I
I
--------l.---
-
15
Do
I
-
f-
Wsm
I
..il"
-----
--~
~
~-
~
-
0
-
A
I
LEGEND
0
A
A
0.333-F"T-DIAM CULVERT
0.224-FT- DIAM CULVERT
A
-
f-
I
0
I
I
I
3
2
I
I
5
4
FROUDE NUMBER,Fo = V 0 /vg 0 0
I
I
6
7
I
8
I
10
40
zo
F0 =3.05
0
"
0
I
0.05
10
8
e
1000
4
10
50
100
500
TEST DURATION, t
1
MIN
5000
10, 0 00
WIDTH OF SCOUR AS A
FUNCTION OF FROUDE NUMBER AND
TIME FOR MAXIMUM TAILWATER
3000
v
I
..
2000
v
v
j1.70
~I
1000
v
1
~'l
It
800
I
I~
600
I
vs
1~/
400
o3
I
0
V'
6.
v
200
LEGEND
Do
0
6
6.
100
6.
80
I
60
Iv
I
0.333-FT•DIAM CULVERT
0.224-FT·DIAM CULVERT
I
"
!&.-
40
1
2
4
6
8
10
VOLUME OF SCOUR AS A
FUNCTION OF FROUDE NUMBER FOR
MAXIMUM TAILWATER
PLATE 10
10, 000
5 0 00
L.,..
1000
__.
0
500
-
~~
~
-l.---"
~
-
I<"
~ ---
v
v
v
-U'
1---"
...---ro.Jo
~
I
lb
100
50
0
10
50
100
TEST DURAT ION , t
1
500
MI N
1000
5000
VOLUME OF SCOUR AS A
FUNCTION OF TIME FOR
MAXIMUM TAILWATER
P L AT E II
lJ
r
~
Lo
fTl
I\)
0.2
FLOW.,.
~ORIGINAL
~I'\.
I
'\
0.6
6
\.
.A
I
I'
l
1.0
1.2
6
I
0
6
D""' ~
I
0
to..
0
~L
0
6_n
~
I
I
I
I
i
0.1
0
I
-
I
I ~ ""-
I
0.8
I
~
I
v
AD~
I
0 .4
Ds
Ds m
GROUND LINE
0.2
0.3
0.4
_g.. p- ~
-
v
/
/
y
v
v
6
Ll
t:.
6
~
0.5
0.7
0.6
0.8
0.9
1.0
0.8
1.0
LsfLsm
DIMENSIONLESS CENTER - LINE PROF ILE
0
.n
0
0.2
'
0.4
I
o.6
-
"'·~
I
Dsm
l-.--
\
'
'
I
Ds
CD
\
'
I
I
1.2
1.0
CULVERT
~
~I
_l
I
I
i
I
~
I
~
~
~
v
i
/
I
.
l
0.8
~
~
/
I
~
6
I
'
1.0
0
GROUND LINE
0
0
I
0.8
~ORIGINAL
0. 4
0.2
0
0.2
0.4
0.6
Ws/Wsm
LEGEND
0
6
0.333-f"T-DIAM CULVERT
0. 224-FT-DIAM CULVERT
DIM EN SIONLESS CROSS SECTION AT 0. 4 L sm
DIMENSIONLESS
SCOUR HOLE GEOMETRY
FOR MAXIMUM TAILWATER
1.0
I
M
l
I
I
I
I
I
0.9
I
I
I
I
I
cis= 0.25 Do F0 0.8
0.7
I
v
I
t::.
Jozs
II
0.6
t::.
d
~ 0.5
~
I""
Do
.....
0
0.4
Ot::.
0
v6
0.3
I
0.2
00
LEGEND
0
0
0
0
t::.
1.000-F"T-DIAM CULVERT
0. 333-FT-DIAM CULVERT
0. 224-FT-DIM.A CULV ERT
0.1
0
0
0
0
2
3
F ROUDE NUMBER,
r0
=V
0 /
4
5
e
,J,-g-=D-0
STONE DIAMETER AS A
FUNCTION OF CULVERT DIAMETER AND
FROUDE NUMBER OF FLOW FOR
MINIMUM TAILWATER
PLATE 13
50
0 0
(X)
00
~
(X)
00
aD
--- - -
- -o
~
40
00
-
30
00
0
SUGGESTED CURVE
L sP =D0 (8+ 17 LOG FQ}-
aD
)
~
20
0
)
6
"J~
~
0
0
~
0
10
I
0
2
I
3
5
4
FROUDE NUMBER AT CULVERT OUTLET7 F0
= V0
6
7
8
9
10
/~
LENGTH OF STONE PROTECTION
0
0
6
ADEQUATE
INADEQUATE
ALMOST ADEQUATE
I
LENGTH OF STONE PROTECTION AS A
FUNCTION OF CULVERT DIAMETER AND
FROUDE NUMBER OF FLOW FOR
MINIMUM TAI LWATER
S::
0
-.J
IJ...
'
PLAN
0
0
2d5
ELEVATION
RECOMMENDED CONFIGURATION
OF RIPRAP BLANKET FOR
Ml NIMUM TAILWATER
PLATE 15
1.0 ~-------.--------.--------.---------r----~---.--------,
0.9 ~-------+--------4-------~--------~v~------+-------~
~
I
I
ds=0.25FQ 0 0 -0.150<) ~
0.8 ~-------+--------+-------~------~~--------+-------~
0
0.7~------+-------~-------+--~~--~--------+-------~
~0.25
0. 6 ~-------+--------+-------~~------~--------+-------~
ds 0.5
Do
~-------+--------+-----kr~~uf-----~--------+-------~
p
0.4
I
0.3
v
6
0
0
,l
0
lo
0.2 ~-------+--~----4-------~~-------+--------+-------~
LEGEND
0
0
0
1.000-FT-DIAM CULVERT
0
6
0. 333-FT-DIMv1 CULVERT
0. 224-FT-DIM.A CULVERT
0.1 ~------~--------~-------+--------4---------~-------4
0 ~--~~~--------~------~~-------L--------~------~
0
2
3
4
5
e
FROUDE NUMBER, F0
=V0 /
,JgD 0
STONE DIAMETER AS A
FUNCTION OF CULVERT DIAMETER AND
FROUDE NUMBER OF FLOW FOR
MAXIMUM TAILWATER
PLATE 16
so
0
v
0
0
00
/
40
~
\
SUGGESTED CURVE
ro
Lsp=Do (8+5SLOGFo)~
0
0
30
0
/
20
0
0
0
0
10
0
~
~
I.J
0
0
0
Do
DO
0
0
2
I
0
~
0
3
4
FROUDE NUMBER AT CULVERT OUTLET, F0
5
= V0
6
7
e
9
10
/-v'9Do
L ENGTH OF STONE PROTE C TION
0
0
1:i.
ADL: Q UAT E
INADEQUATE
ALMOST ADEQUATE
LENGTH OF STONE PROTECTION AS A
FUNCTION OF CULVERT DIAMETER AND
FROUDE NUMBER OF FLOW FOR
MAXIMUM TAILWATER
I
-.------
i30o
PLAN
ELEVATION
RECOMMENDED CONFIGURATION
OF RIPRAP BLANKET FOR
MAXIMUM TAILWATER
PLATE 18
\
\
\
\
\
I
I
I
I
I
I
I
'"'\
~
1-.:
~
~
......
\...
\
l't
II)
\
I
I
I
I
I
\
-..1
I
\
\
\
$
'"'\
~
1-.:
\
------r----+----I
\
\
\
\
\
\
\
I
I
I
I
I
I
I
~
\.:.
'
II)
-..1
\
\
I
I
\
\
\
I
I
I
I.
~
Wsp (MAX. T. W.)
Wsp (MIN. T.W.)
RECOMMENDED CONFIGURATION
OF RIPRAP BLANKET
SUBJECT TO MAXIMUM AND
MINIMUM TAILWATERS
PLATE 19
04
01
008
/ vv
/
V
-LV
/
,,
//
v
vv
v
/
/
v
0
B
_Q_
~gas
0001
/
o.oooaV
/
V
0.0006 1/
v
v
v
v
v
1/
v
v
I
I
v
V
v
o'
bv++-r~~~~~--~
/
0.0004 /
v
0
v
~
oooo2 r-r-/-+v-+v-t--1--t-t -t-:rf-V-+V-Hvl--+
1-+--+--t--r--t
v
vv
vv I
i
!
j
0 000 I L----l..L..--t~--L-L...u...J...-'--L..L.....J....l-l.-...L......l.---1.---L.--.L-...J
0 06 0.08 0 .10
04
06
08 10
CHARACTERISTICS OF
A TRAPEZOIDAL CHANNEL WITH
1- ON -2 SIDE SLOPES AS A
FUNCTION OF FROUDE NUMBER
PLATE 20
Unclassified
Security Claaairication
DOCUMENT CONTROL DATA- R & D
(Security cleadflcatlon of title, body of abatract and lnde1Cin
I. ORIGINATING ACTIVITY (Corporate author)
Q annotation muat be entered when the overall report Ia claaelfled)
Z.. REPORT SECURITY CLASSIFICATION
Unclassified
U. S. Army Engineer Waterways Experiment Station
Vicksburg, Mississippi
2b, GROUP
3. REPORT TITLE
EROSION AND RIPRAP REQUIREMENTS AT CULVERT AND STORM-DRAIN OUTLETS·
Hydraulic Laboratory Investigation
'
4. DESCRIPTIVE NOTES (Type of report and lnclualve datee)
Final report
e. AU THO RCSI (Firet name, middle lnlllal, Ia a I name)
Joseph P. Bohan
e. REPORT DATE
7a. TOTAL NO . OF PAGES
7b. NO. OF REFS
ll
53
January 1970
... CONTRACT OR GRII.NT NO.
Ga. ORIGINATOR'S REPORT NUM'BERCSI
Research Report H-70-2
b. PROJECT NO.
c.
Gb. OTHER REPORT NOCSI (Any other numbera that may be aeei(;Jed
lhl• report)
d.
10. QJSTRIBUTION STATEMENT
This document has been approved for public release and sale; its distribution is unlimited.
II· SUPPLEMENTII.RY NOTES
12. SPONSORING MILITARY ACTIVITY
Office, Chief of Engineers, U. S. Army
Washington, D. C.
11. ABSTRACT
Investigations were conducted in experimental facilities to determine the characteristics of scour below
a culvert outlet discharging onto a horizontal blanket of cohesionless soil and to develop guidance
for designing riprap protection downstream of culvert and storm-drain outlets. Scour holes, produced by the discharge of various flows of several durations through culverts of various shapes and
s1zes, were observed and contoured. These data were obtained for several tailwater conditions and
were used to develop generalized expressions describing the maximum length, width, depth, and
volume of scour as functions of the flow duration and the Froude number of flow at the culvert
outlet. Dimensionless scour profiles and cross sections, for low and high tailwater conditions, were
also developed. Riprap blankets of various configurations and stone sizes were tested with several
culvert sizes and shapes and various tailwater conditions to determine when displacement or failure occurred. The results were used to develop generalized relations to describe the stone size and
appropriate blanket configuration required to prevent blanket failure and soil erosion, respectively,
at a culvert or storm-drain outfall.
DD .'!-:••1473
IIIIC~LAC&a DD ~OIIIIM 1411, I JAN t4, WHICH . .
OaaGLIKTK ~01111 AIIIIMV ua&.
Unclassified
Security Claniflcatlon
I• .
LINK A
LINK B
LINK C
KEY WORDS
ROLE
WT
ROLE
WT
Culverts
Erosion
Rip rap
Storm sewers
.
.
•
c
.
•
Unclassified
Security Classification
ROLE
WT