COMMONWEALTH OF KENTUCKY

DEPARTMENT OF HIGHWAYS
FRANKFORT

HENRY WARD January 21 , 1963 ADDRESS REPLY TO

COMMISSIONER OF HIGHWAYS
DEPARTMENT OF HIGHWAYS

MATERIALS RESEARCH LABORATORY

132 GRAHAM AVENUE
LEXINGTON 29, KENTUCKY

D. L 7,

MEMO TO: Ao 0, Nesier

Assistant State HighwayEngineer

As an outgrowth of our hydrological studies on rainfall runoff,

and our companion studies on the hydraulics of culverts , which began
about 1951 , and which were moreorless consummated in the Department's
current Drainage Manual, certain questions arose concerning the re lative
effects of var1ous types of entrances on the capacity of culverts, In
1954 , equipment for making model studies of culverts was designed , built,
and put into operation in the hydraulics laboratory at the College of
Engineering, University of Kentucky, Information was sought in three
general areas , which were as follows:

L Effect of obliquity of wing-walls on the capacity

of box culverts ,
2, Effect of hooded entrances on hydraulic capacity,

3 , Effect of drop inlets on the capacity of box-type

and pipetype culverts,

The first reports were concerned with item 1 , and a report was
made to the Department in 1956*. That report presented data obtained

"Hydraulic Model Studies of Culvert Operations ," byE, M. West,

November, l956 o

on 30- and 4Sdegree wing-walls, Somewhat concurrently , the information

from that report was combined with test results pertaining to item 2
(hooded inlets) and published in Bul letin No" 41 , EngineeringExperiment
A, 0, Neiser 2 - January 21 , 1963

Station, University of Kentucky, 1956*. For reasons which are now obscure

"The Development and Use of Hydraulic Models in a Study of Culvert

Perfonnance," by E, M. West and R, D, Hughes,

perhaps , the model was disassembled and was not operable for several years,
However , interest in drop inlets persisted; and , inasmuch as this was
one of the areas which we were more-orless committed to study , the model
was restored and adapted for the study of drop-inlets, The tests on this
phase were made over a year ago , and the model was then returned to stor
age. The reporting of this work has been delayed , so to speak , by other
matters,

The study was made by R. D, Hughes, and his report , "A Study
of the Hydraulics of Drop-Inlet-Type Culvert Models , " is attached,
hereto, His results and analyses are best summarized by the discharge
coefficient presented in the report, These coefficients, which are
dimensionless , may be used directly in the basic formulae to compute
discharges in full-scale culverts; whereas, the similitude analysis
merely yields a factor relating the discharge in the model to the dis
charge of its fullscale counterpart , The end results would be the
same , but the similitude analyses seem to be somewhat cumbersome.

The information may prove to be helpful should a need arise

where only a dropinlet structure will suffice or otherwise be preferred,
Should such a need arise , we would welcome an opportunity to assist in
its design and to observe its performance ,
Respectfully submitted,

'

W, B, Drake
Director of Research

WBD:dl
Enc,
cc : Research Committee Members
Bureau of Public Roads (3)
 Commonwealth of Kentucky
Department of Highways

A STUDY OF THE HYDRAULICS OF DROP-INLET-TYPE

CULVERT M)DELS

by
R. D. Hughes
Research Engineer Associate

Highway Materials Research Laboratory

Lexington , Kentucky

January , 1963
 INTRODUCTION

In certain unique situations , a dropinlet-type of culvert

might be employed to a greater advantage than the standard box- or pipe

type culvert, A dropinlet consists merely of a vertical drop-section

at the upstream end of the culvert, The drop section is connected to

the main barrel of the culvert by means of an elbow or merely a right

angle intersection, The entrance to the dropsection must collect and

funne1 the channe1 water into the drop" section, Thus , intercepting

dams and sidewalls may be needed to form the entrance, Figure 1 illus

trates , in a general way, some of the options which may be c?nsidered

in the design of a highway culvert, The diagram alludes to the parti

cular case where the difference between inlet and outlet elevations is

such that the slope-gradient is quite steep, Likewise , as envisioned

in Fig, 2 , drop"'inlets may find some application in situations where

there is not sufficient headroom beneath the pavement to permit the in

stallation of a culvert to the desired slope -- that is , on the uniform

grade between the upstream channel and the outfall channeL

Although the concept of drop-inlet culverts is not new , situa

tions have arisen in which this type of culvert might have been pre

ferred to other alternatives ,but was not used because reliable design

parameters and criteria for design were not available and appropriately

stylized, Previous information pertaining to drop-inlet structures

was reported by Kessler in 1934 (1) and by Huff in the early 1940"s (2),

- 1 -
 - 2 -

open or close d

--- -
=
- - -.:....-:-
ou tfall flume

::::.. --
- ,
____.t:..._ - ::::::
-

"'-
""""'--
--

I -- - - -
- -.-
' "'
I I - -
-\R CO U

-
IIIII
--

' -- '
1 l
--=-_-=:_-=:_
"-..... .. 0 aion
-:::- - - \
inlet

.::-'>,.
- ""' -
_
-

L- ---
- - -
I ow di!lchorg velocity
uniform grade

Fig. l, Optional Routes of Culvert Under High Hydraulic Gradients

L -
L---------

Fig. 2. DropInlet Culvert in Low-Head-Room Situation.

 - 3 -

Kessler's studies were concerned primarily with erosion-control struc

tures; whereas, Huff's studies were concerned more specifically with

the effects of the width of the approach channel and submergence of

the entrance of flow-capacity, In 1951, Blaisdell and Donnelly (3)(4)

reported data from 361 tests which involved free-flow, submerged-flow,

and outlet-control conditions, The aforementioned reports deal with

the hydraulics of drop-inlet structures, but none deals specifically

with the design of drop-inlet structures for highway culverts.

The purpose of this report is to present the theories related

to the hydraulic operation of drop-inlet culverts and present the re

sults and analyses from a series of tests conducted on various models

of highway culverts, The results are presented in the form of dis

charge coefficients which are recommended as interim design values and

which, of course, are subject to confirmation by full-scale, field

installations,

The design values presented herein are based on results of

tests conducted on drop-inlets having drop-sections which were 4 by 4

inches in cross-section and 6-1/2, 9 and 1 2 inches in length, The

barrel was 4 by 4 inches in cross-section and 7 2 inches in length, One

of the transition sections consisted of a 91-degree, box-type inter

section, and the other consisted of a modified box-type intersection.

The inlets used in all tests were similar in that each had sidewalls

parallel to the channel and a backwall -- all extending two inches

above the crest of the approach channel. Other features of the models

and the test procedures are included in the body of the report,
 THEORETICAL CONSIDERATIONS

Whereas bridges merely span the natural channels of large

streams , culverts usually constrict the natural channels of small

)
streams" Constrictions, of course, tend to impede flow" I f a culvert

is to carry the same volume of water as an unconstricted channel , the

velocity through the culvert must be greater than the velocity in the

approach channeL If the constriction is too abrupt, the water piles

up at the entrance, but , even so, the barrel of the culvert may not

flow full; hence, the capacity of the barrel exceeds the capacity of

the entrance" This condition is commonly termed entrancecontrol and

is typical of flow through orifices" In each culvert, there is some

dominant impedence or controlling factor, which largely governs its

hydraulic capacity , For instance, backwater in the downstream channel

creates backpressure in the barrel and decreased velocity in the barreL

The control may thus be at the outlet rather than at the entrance ,

Likewise , the control would be in the barrel only when the barrel itself

is the principal impedence,

Whereas Bernoulli"s equation suggests that all impedences are

accumulative and are equal to the total energy change , ioe, ,

2 2
(HW TW) + y 1
-- y 2 ; he + hf + ho + \) " 0'

Zg

some of the losses becomes negligible under certain conditions de

pending upon the location of the controL In the case where the outfall

<> 4
 0 5

is free, the outlet loss (h0 ) is insignificant, Thus, either he or

hf governs, Moreover, if the barrel of the culvert changes direction

abruptly, as in the transition from the dropsection of a dropinlet

type culvert to the main barrel, it might be desirable to introduce

a parameter , hb to describe the loss in the bend,

Inlet Acting as a Weir

The initial or low-head flow into a drop-inlet is typical of

flow over a horizontal weir, The general flow pattern is shown in

Fig, 3, The edge or top surface of the weir is called the crest, and

it may be either flush with the bottom of the approach channel or at

some elevation above ito Weirs are classified as either sharp- ,

broad-, square-, or round,ocrested, The over-falling stream is desig

nated as the nappe and takes the form as shown in Fig, 3 for the

square-crested weir, If the length of the weir is less than the width

of the approach channel , the nappe will be contracted at each side,

Directly upstream from the weir, there is a downward curvature of the

water surface, known as the draw-down curve, Head , as measured at the

crest, is slightly less than that actually causing flow; therefore ,

the velocity at some point upstream from the weir is taken to be the

velocity of approach and is so used in all calculations ,

Development of the basic weir formula is made in reference to

Fig, 4, Hw represents headwater elevation above the crest of the weir;

dh is the thickness of an elementary strip dh x L; h is the headwater

 - 6 -

drop inlet

crest
1/ orrel

Fig. 3. Weir-Type Flow into a Drop-Inlet.

-::-
_
I
L __
llJ l=+dll
L

Fig. 4. Elemental Section of a Weir.

 - 7 -

elevation above and o<... h is the approach velocity-head; and the effec

tive head is h h. The theoretical discharge through the elementary

strip may be found as follows:

dQ+ = L f dh zg (h + ' h) 1

= JL Vzg V
"
Q+ h + o(h dh 2

'{fi L
0

Q+ = 2/3 [ Cf\ + c:><.

3/
h)3/2 - (O(h) 2 l 3

The approach-velocity head is relatively insignificant in comparison

to depth of flow, thus the term"< h may be omitted from equation 3,

A more realistic value of discharge than that determined from equa-

tion 3 may be obtained through use of a coefficient with the theoretical

formula. This coefficient may be combined with constants in the weir

formula, and the resulting equation for actual flow over the weir is:

Qw = Lfiw 3/ 2
4

where represents the product of the weir-coefficient and the con

stants in the theoretical weir formula. Assuming no head-loss during

flow over the weir, has a value of 2/3 V Zg or 5, 35, Since head

losses do occur, has a value somewhat less than 5.35,

Inlet Acting as an Orifice

An orifice may be any closed perimeter through which a fluid

flows. The inlet may act as an orifice if the headwater floods the

weir. Flow through an orifice is governed by the area of opening as

 8

well as the total head acting at the center of the orifice, Theoreti

cally, discharge through the orifice is found to be the produce of

the area of the opening and velocity, The general equation is:

Go = Ao Vo = Ao 5

Where A 0, V0 and H0 represent the area of opening, velocity, and

headwater elevation above the center of the orifice, respectively,

Orifice flow is shown in Fig, 5,

As water passes the orifice, the area of flow is somewhat

less than the area of the orifice: due to convergence of the flow

paths as they enter the orifice, The ratio of the crosssectional

area of the jet at the contracta to the area of the orifice is

called the coefficient of contraction and is designated as Cc, The

ratio of the actual velocity to the theoretical velocity is termed the

coefficient of velocity and is designated as Cv, The actual discharge

passing the orifice is found to be the product of the actual area of the

jet times the actual velocity in the jet just beyond the orifice, or

Go = Ao C c Cv 6

The product of the coefficient of contraction, coefficient of velocity,

and {iii gives the orifice coefficient C0, The general formula for

actual flow through the orifice thus becomes:

l/2
Go = Ao C o Ho 7
 - 9 -

t_Ho
-

-
-=-
-
- - --_
-=- --... , .
.... . .-. .\\. ... ,.
0

1\. '.tll'
-
- -
_ _

" t
-
-
..
v.

., ("";". "
l}lll. - -:/-:. ---.::-
/0 ----,
-

Fig. 5. DropInlet Acting as an Orifice.

Short-Tube-Flow

In some instances flow may be great enough to fill the drop-

section but not the barrel, This condition is likely to produce surging

as control passes from the inlet to the barrel, and When

the drop section is running full, and the main barrel of the culvert is

not running full, discharge is again found to be the product of the area

of the vertical section and the velocity. Since the vertical section

is assumed to be completely full, the coefficient of contraction is

unity, and the short-tube coefficient is found to be the product of Cv

and y zg. The basic equation for short-tube flow is:

1/ 2
Qst - Ast Vst - t
- A Cst Hst 8
 - 10 -

Fig. 6. Drop-Section Flowing Full; Short-Tube-Flow.

where Hst is as shown in Fig. 6, est is found to vary with slope

of the approach channel, geometry, length and roughness of the drop-

section.

Pipe Flow; Barrel Control

When the headwater elevation becomes great enough to cause

full flow in both the drop-section and barrel, control switches to the

barrel and to the domain of ordinary pipe-flow. The total head causing

flow is assumed to be the difference between the headwater elevation

and centerline of the outlet. Theoretically, the Head, Hp, should be

the difference between headwater elevation and the point at which the
 - 11 -

Fig. 7. Drop-Section and Barrel Flowing Full; Pipe Flow.

hydraulic grade-line pierces the plane of the outlet , This point is

_
generally difficult to determine and is very nearly the same as the

centerline of the outlet, thus little error is introduced by use of the

assumption.

The total headwater elevation, as shown in Fig. 7, may be

expressed in terms of the various head-loss coefficients as well as

other factors governing flow. The equation thus becomes:

Hp = V2/Zg [1 + Ke + Kb + Ko + f c L c/4 R c + fr L r/4 Rr l 9

Ke, K b and Ko represent the head-loss coefficients at the inlet, elbow,

and outlet, respectively; V is the mean velocity throughout the struc

ture. The factors, fc Lc/4 Rc and f r L :r/4 Rr represent losses which

 12 -

are attributable to friction in the barrel and drop-section, respec

tively. The f's represent the Darcy-Weisbach coefficient of friction;

R's represent the hydraulic radii; and L's represent the lengths of the

sections. The discharge may thus be computed in terms of: 1) area

of flaw (when areas of all sections are equal), 2) total head causing

flow, and 3) head-loss coefficients. In this form:

ty/4 R;J
ll
Q =A 10
 APPARATUS AND TEST METHODS

The apparatus and test methods employed in this series of tests

were essential ly the same as those used in earlier studies and as re

ported previous ly (5) (6), A schematic and overallview are shown in

Figs , 8 and 9,

Water was pumped from a pit and into the approach channel,

from which it passed through the mode l culvert to the weir tank for

gaging and was then returned to the pit for reuse, Turbulence of in

coming water was stilled by means of an Htype outlet in a diffusing

tank and by means of baffles placed perpendicular to the flow from the

tank, Discharge was regulated by a valve above the Htype outlet,

The approach channe l 1as constructed of marine plywood and

was designed to simulate a trapezoidal stream-channel , A p lexiglass

end-section, having a 2:1 slope and representing a typical highway

embankment was connected to the downstream end of the approach channe l,

Provision for affixing various inlets was made through use of a flanged

opening in the plexiglass endsection,

Drop Inlet Mode ls

Each model consisted of four interchangeable units: the in

l et , drop section, elbow and barrel, All inlets were identical except

for the flange-angles which were varied according to the channel slope

for which each particular inlet was designed and in order that each

13
APPROACH CHANNEL

PLEXIGLAS S END SECTION

HOOK GAGE
......
...

BARREL
SUPPORT

"--- SCREW JACK FOR

ADJUSTING SLOPE

Fig. 8 - Model Apparatus

ST
 16

dropsection would be vertical when attached to the approach channel

at the designated slope, The inlets , one of which is shown in Fig, 10,

were designed so that the upstream wall of the dropsection would be

flush with the bottom of the approach channeL The headwal l extended

two inches above its juncture with the embankment portion of the end
section , The sidewalls were constructed to the same e levation as

the headwall, Four inlets were constructed one each for channel

slope 0, 3, 5 and 7 percent, The lower portion of each inlet was

flanged for connection to an e lbow or dropextension, The e lbows are

shown in Fig , 11, The center1ineintersection in each was 91 degrees,

The extnesionsections are shown in Fig, 12, These extensions were

simple , 4 x 4inch sections, 21/2 and 51/2 inches in length

giving a total drop of 9 and 12 inches , respectively, A drop of 61/2

inches was obtained with the e lbow connected directly to the inlet ,

The 4 x 4cinch barrel section , 72 inches in 1engtih ,was used

throughout these tests, Peiziometer tubes were attached along the

bottom of the barrel and e lbow , at Zinch intervals , and were connected

to the manometer board, One tube was attached at the top of the e lbow

in order to measure the pressure at the bend,

Test Procedure

The desired s lopes of the approach channel and barrel were

set by screw jacks and checked by means of a leve l, Tests were conducted

at headwater elevations ranging from approximate ly one inch to 15 inches

- LT -
....
.....

- 8T -
....
N

- 6I -
 20 -

and at intervals of approximate ly two inches, Headwater elevations

were controlled by a valve in the piping system, A lSminute period

was al lowed after final adjustment of the valve in order to permit the

f low to reach a steady state, Headwater elevations were then mea

sured at the inlet, and the discharge was determined by use of a hook

gage in the weir tank, Photographs were made of the manometer boards

to record the attendant pressures , Notes were made as to type of f low

in the approach channel and barreL

 RESULTS AND ANALYSES

The complete series of tests included model assemb lies consist

ing of 61/2, 9, and 12inch drops, approachchannel slopes of 0 , 3,

5 and 7 percent, and the two elbows, Twentyfour installations were

simulated through use of all combinations; thus comparative data were

obtained for each variable as wel l as the specific data needed to calcu

late head loss coefficients ,

Headdischarge curves were plotted from the raw data, TI1ese

curves proved to be insignificant inasmuch as the points were widely

scattered for headwater elevations of three inches above the bottom of

the approach channel, This was attributed to the control alternating

from orifice to shorttubeflow to pipeflow and thereb y , altering the

basic headdischarge relationships, Results were thus analyzed from

the standpoint of the coefficients in the basic formulas for various

types of flow through dropinlets, Ranges in headwater elevation for

control to occur in various sections were determined from observation

of flow patterns during testing, These heads and their respective

discharges were used in the calculations of coefficients for various

types of controL The resulting coefficients are listed in the fol low

ing table and are designated Cw, C0, Cst and Cp for the weir , orifice,

short tube, and pipeflow coefficients respectively,

Flow was controlled by the inlet acting as a weir at head

water elevations of zero to three inches , Values of Cw were obtained

21
 22

by substituting headdischarge values into equation 4, CW was found

to be constant for variable lengths of drop and for each e lbow since

control was upstream of the drop section, Values of CW were found to

increase with the slope of the approach channel , This is attributed to

an attendant increase in the approach velocity which was assumed to be

negligible and thus not taken into account in the basic weir formula,

D ISCHARGE COEFFICIENTS

Channe l Length Coefficients

Slope of Drop Weir fi
On ce Short 'i\i6e hpe
(%) (ins . ) cw Co Cst CE

0 61/2 2,78 3,61 3 .50 2.62

0 9 2.78 3,62 3,66 2,59
0 12 2 ,79 3 ,62 3.46 2.46

3 61/2 3,04 3.71 3.64 2 .59

3 9 3 . 03 3.71 3,83 2,53
3 12 3.05 3.72 4 .06 2,43

5 61/2 3,30 3.78 3.76 2 .56

5 9 3.31 3.80 4.04 2 .50
5 12 3 . 30 3, 79 4.30 2,43

7 61/2 3.55 3.86 3.93 2 .50

7 9 3,57 3 .87 4,23 2,43
7 12 3.56 3,86 4.51 2.37

Analysis for orificecontrol was similar to that for weir

control. No appreciable variation in C0 was noted for the various lengths

of drop . Values of C0 increased as the channels lope increased again

due to an increase in velocityhead, whid1 is considered negligib le in

the basic orifice formula, The variation in C0 with changes in slope

 - 23 -

were not as great as that for Cw because discharge under orificecorttrol

occurs at greater headwater e levations and thereb y , decreases the over

al l effect in variation of approach velocity,

The coefficients , Cst as determined for short-tube-flow,

vary more or less directly with the length of drop for a constant

approachchanne l s lope, Increases in Cst with greater lengths of drop

may be attributed to gain in head, The gain in head more than com

pensates for the increased friction,

Pipe-flow coefficients, decreases with increasing lengths

of drop, In reference to equation 10 , it may be seen that Cp decreases

as Q increases; thus , decreases with increasing drop , and this re

presents a gain in head over the loss due to friction,

Both Cst and increased for increased slopes. Here , again,

the increase is due to greater velocities of approach at the higher

slopes,

In all cases the effect of the type of e lbow was found to be

negligible in that the head-discharge relationships for the two were

a lmost identical, For this reason , data obtained from tests using

each e lbow were combined in the analysis for effect of s l ope and length

of drop, The combined head-discharge curves are shown in Figs, 13, 14

and 15, The continuous portion of these curves were plotted from the

test results; whereas , the dashed portions represent theoretical

extrapolations which are based upon discharge coefficients determined

from tests, The headwater elevations are shown as Hw and H0, The

datum for short-tube-flow was taken to be the break-point between the

 - 24 -

3% SLOPE
1---
- O"'o SLOPE
6.5" DROP 6.5" DROP

" H0,11,..
' 1.6 1.4 '
1.8 1.8
"" .,;'\.
1.6
1.4 1.2 I I 1 .6
1.4 1.2 II I

II II VI II

!A

I1/
"'

..
1.2 1.0 . 1.2 1.0
::;: 1.4

/;
c
"' ..

1.0 .6 "'
1.0 .6 --------
...--
Ill l
1.2 :1: 1.2

/; 7 ..-----
...--

p Jj/fl
.6- .6 .6 .6
1.0 1.0

/-
1
.6

I/ //
.4 .6 .4

//I ?
.6 .6

.4- .2 /
[/ 1/ 1/
.4 .2
.6

JL-
. 6

/ olZ- [? 17 1/

1/
.2- 0 /
[/
.2
.4 .4

.2
0-'-- ,_.../
/
l/ .2
0
./

/
0
____ ,/ 0
-
0 .2 .3 .4 .5 .6 .7 .6 0 .I .2 .3 .4 .5 .6 .7 .6
0 ISCHARGE-(c.f.s.l OISCHAAGE-(c.f.s.)

I I
1----
7% SLOPE 5% SLOPE
-
6.5" DROP 6.5" DROP

Hst Ho, " ""'""

" 1.6 1.4 ' 1.6 1.4
1.8 1.6

1.6
1.4 1.2 I 11 1 1.6
1.4 1.2 I I
I I;II IllII

1.2 1.0 c 1.2 1.0

1.4 1.4

I fl

"'
.a
2
" 1.0 1.0 .6

I vi 1----
1. 1.2

1.0
.8 .s
IIjI[-v
.6 .6 /; I -----

t'!-kl-----
1.0

v I v 7/
.6 .4 .6 .4
.a .6

/
l/ Gl
.4 .2 /

v / [/
.4- .2
.6 .6

2
v
-
/
0 .2 0
.4 .4
v /
v
_v
0 -
.2 0__[_

I/
.2

/ 0
0
0 .2 .3 !4 .5 .6 .7 .8 0 .I .2 .3 .4 .5 .6 .7 .8
DISCHARGE -(c. f.s.) DJSCHARG E- (c.f.s.)

Fig. 13. Combined Head-Discharge CUrve for 6 .5-inch Drop.

 - 25 -

I I I 1
0% SLOPE 3% SLOPE
-
9" DROP
,1--
He. 9"0ROP liM
H..,M.,
1.4
liM
"
1.8 1.4 ' 1.8
2.0 2.0

1.6
1.2 I I; 1.8
1.6
1.2
I I
1.8

I I
f/ I
_...-l. . I _...-
1.0 1.0
1.4 1.4
1.6 2t.6

7 I r7/ /

II 1 -----
II
.8
.
1.2 .8
1.2
0
"' 1.4

1/.....-: II 7 -----
',21.4 "'

'/

1.0 .6

/
1.0 .6

j.v
0

..--!-VI
"'1.2 1.2
"'

I/
.8- .4

v:::
8 .4
1.0 1.0

v 1/ r7 I 1/

)/ I
.2

\{--
.2 .6
6 / .8

6
.8

0 I / .6
.4 0

/ 1/
I/
.
I;
1/
' .2-
.2

I./ /
.4
.4
/'
o_L_-

0

l../
.2
.2

-v 0
0 0 .1 .2 .4 .8
.8
.3 5
0 .2 .3 .4 .5 .6 .1
DISCHARGE- (c.f.s.l
DISCHARGE(c.f.sJ

1 I I I
.
5% SLOPE 7% SLOPE
r--
-
9" DROP
9" DROP

.a '"'
",J\, " ......
1.8 1.4
It' '
" 2.0
2.0

I 1.6 1.2 I ,I
1.6 1.2 I 1.8

7
\I /,
1.8

1.6
1.4
1.0
\ II 1.6
1.4 1.0

I 'I ; \II
I jl
.8 1.2 .8
0
<c:l.4
1.2 1.4
"'
.6
r7 /, ]... "'
1.0
.6

-;
%
1.0
1.2

w
1.2 /

r;
.4 // 8 4 /
1/1
.a- 1.0

f:1
1.0

I
!7
v-:: ......:.

1/j
.2 .6 .2 /
-
/
.6 .8

0
.8
l;

1/ I
.4 0
0
.4 .6

I/ 1/
.6

7 v 1/
.2
.2 .4
.4
v
_v-
0
0 .2
.2

0
v 0
.I .2 .5 .6 ,7 .8
0

0 .2 .3 A .5 .6 .7 .e .3 A
DISCHARGE-( c.f.s.)
DISCHARGE ( c.f.s.)

Fig. 14. Combined Head-Discharge Curves for 9-inch Drope

 - 26 -

"ollw
H.,.l!
"t 1.4- " I :4

f-- r---
"

1;1
2.0 O'Yo SLOPE " 2.0 3% SLOPE
2.2 22

I; I
12' DROP 12" DROP
12

I I
1.2
18 1.8
2.0 2.0
10

I J 1.0
IlL
I
I
I
16 1.6

I I
1.8 1.8
1---
$
.8 -
.8
/ ----

'--t
14 14

I
1.6

I
16
.6

:Jt.4
.6

/
0
12- / 12
1.4
4
v _,/-
v

V/
I 4

'/
10 / 10

/
12 1.2

/ /
L
.2 .2
/

jL--
/ ll
.8 .8

/_ 1/
1.0 10
0 f'--- 0

/
.6 .6
8 .8

/ --t-- /
v 1/
.4 .4
.6 .6
/ I
/ 1/
.2 .2
.4 .4

/v v

/
0 0
.2 .2

_v 0
0
0 .2 .3 A .5 .7 0 .I .2 .3 .4 .5 .6 .7 .B
D!SCHARG E (c.f.s.) DISCHARGE- (cJ.s.l

I l
,
-
H00H, ....

-
11ot 1.4
... . "

I II
2.0- 7% SLOPE
" 2.0 5 % SLOPE 2
2.2

j_
12" DROP
1.2
12" DROP
18
12
II
I
1.8 2.0

II II II
2.0
10 1.0

I /, I
16
1.6
1.6

..

1.6 14 __.-I 1.4

I I /;---
1.6

)/ ------ " .6 -----

.6
12

/-/
0 1.2 1.
.;
1/
cXt.4
'I v
/v
.4
.4

/
10
10 t.:

1----/
1.2
/ .2 ?
.2
/;
jL
.8
/

/ (I
.8 10

r/
10
0 L-- 0

//
.6 6
8

I//
..

7
.4
.4 6

L
6
I
v I//
.2
.2

/
4

/7
.4

v
__
0
0 .2
.2

0
---/ 0 .I .2 .3 .4 .5 .6
0 .z .3 A .s .s .7 .a
DISCHARGE-{ c.f.s.)
DISCHARGE ( c.f.s.)

Fig 15. Combined Head-Discharge Curves for 12-inch Drop.

 27

dropsection and barrel and is below the weir" and orifice-datum, a

distance equivalent to the length of the drop-section minus the

height of the barreL The datum for pipeflow was taken to be the

centerline of the barrel at the outlet and is thus plotted below the

weir and orifice-datum by a distance equivalent to the difference in

the elevation of the bottom of the approach channel and the centerline

of the out let,

1-Iead"discharge curves for the 6-l/Z"inch droposection and

channel slopes of 0 , 3 , 5 and 7 percent are shown in Fig, 13 , and

curves for the 9- and 12"inch drop"sections are shown in Figs , 14 and

15 respectively, Flow was observed to be controlled by the inlet acting

as a weir at headwater elevations from zero to approximately three

inches, Above three inche's, the weir became flooded and the control

switched to orifice"flow, Orifice"control persisted through heads of

seven inches ,

Shorttube"flow prevailed at headwater elevations of seven to

eight inches, Shortotubeoflow is a residual-type flow which occurs

as a transition between orifice" and pipe-flow, Flow of this nature

is most difficult to predict and generally persists for only a short

period of time, Pipeoflow prevailed at headwater elevations of eight

to 15 inches ,

Figures 16 and 17 are head"discharge curves for weiro and

orifice"flow, respectively, for the various slopes , Each curve , for a

specific slope, represents the combined results from the three drop

lengths 00 length of drop did not affect the headodischarge relationships,

 - 28 -

WEIR FLOW
0-3-5-7
percent channel
s I opes I
I

-
....:
-

0
I

<t
w
:::r:
1.0

O%
.8

3%
5%
7%

c;:-
__..-

--
.6

/ :::---

.4

.2

0
0 .I .2 .3 .4 .5 .6 . .7 .a
DISCHARGE- (c.f.s)

Fig 16. Head-Discharge Curves for Weir-Flow.

 - 29 -

2.4
ORIFICE FLOW
0-3-5-7 0,3,5,7 percent
2.2
percent channel
!I
s l opes

1/
2.0

1/$
1.8

1.6

/;w

..;
-

'

0
1.4

1/j
<(
w
J:
1.2

V/11
A
1.0

.8

II
Av
.6

.4

/
.2

0
.
0 .I .2 .3 4 .5 .6 .7 .8
DISCHARGE- (c.f. s l.

Fig. 17. Head-Discharge Curves for Orifice-Flow.

 30

Increases in slope increased the approach velocity and the discharge

at a given headwater e levation ,

Headdischarge curves for shorttube and pipe-flow are shown

in Figs. 18 and 19, respectively, The curves are grouped to show

effects of lengths of drop, It may be noted that the effect of length

of drop is greater for short-tube-flow than for pipe-flow. This may be

explained by the fact that the headloss due to ve locity is greater in

the case of pipe flow than in short-tube-flow,

 - 31 -

2.4
SHORT TUBE
- SHORT TUBE

f!!
1--
FLOW 6. "9"12" drop FLOW 6 5"9"12 " drop
2.2

I,u
2.2
0- percent 3-percent
s Io p e slope

II[/
2.0 2.0

I II;
1// 1//1
1.8 1.8

-
!J'I r!l
1.6 ....., 1.6

0 !.4 *1.4

Ill p;

w w

1.2 1.2

1/1 1.0 Ill

If
1.0

.8
/!, .8

.6 l&_ .6 II
.4 h .4
I!w
,
v v
.2 .2

-"" 0
__.
0
0 .2 .3 .4 .5 .6 .7 .8 0 .I .2 .3 .4 .5 .6 .7 .8
OISCHARGE-(c.f.s.l DISCHARGE- (c.f.s.)

2.4

2.2 - SHORT TUBE

FLOW
5-percent
6.59': 12'
2.2 - SHORT TUBE
FLOW
7- percent
6.5 12
'

sI o pe
I 1/ s I ope
I I
;I/I
2.0 20

I
II II
1.8 1.8

21
.,.

.6

1.4

/II
. .
f----- :21.6

w

!.4 II;1/
rl!
II 'I
1.2 1.2

10
.
I/; 1.0

w lj;
;;r;
.8 8

.6 !J 6

/.ljl . .4
kP .I
_,.,/ /
.2 .2

0
O
0 .I .2 .3 .4 .5 6 .7 8 0 2 .3 .4 .5 6 7 .8
DISCHARGE -(c.f.s.) DIS CHARGE- (c.f.s.}

Fig. 18. Head-Discharge Curves for Short-Tube- rlow.

 - 32 -

I I
PIPE FLOW
l PIPE FLOW

1--
Op&rcent 6.5;Ef
' l:i'dtop 3- percent 6.if,12''drop
2.2 - 2.2
1-- s I o pe s l o pe

II
VII
2.0 2.0

1.8 1. 8

1.6 1.6 J

1.4 1.4

I

h
I
1.2 1.2

1.0
I
I
1.0

I
.a .8

J! .6

f/
.6
f

]/ /
.4 .4

_.... /
.2 .2

_.....
- 0
0 .I .2 .4 .5 .6
0
0 .I .2 .s .4 .a .6 .7 .3 .7 .B
DISCHARGE-(c.b) DISCHARGE- (c.f.s.)

I I I l
PIPE FLOW PIPE FLOW

If
I--
6.5;'(12" 6. rt
&-percent
2.2 -
2.2 slope 7-percent
II ope
2.0

VI!
2.0

l#
d
1. 8 1.6

2t.G
1/!
..., 1.6
'
1
VI
1.4 1.4
Q

/!;

1.2 1.2

II!

1.0
1,0

lf .B

/)
.8

j
I'
.6

Jl
6

v
.4

v
.4

y
.2 .2

0
0
_....V .2 .3 .4 .5 .6 .7 .B
0
0 2 .3 .4 .5
DISCHARGE -{c.f.s.l
.6
-
.7 .B

DISCHARGE- (c.f.s.J

Fig. 19. Head-Discharge Curves for Pipe-r:low.

 APPLICATIQ OF DATA

Prediction of flow through ful lscale structures may be made

by dimensional analysis and hydraulic similitudeo Basic relationships

used in predicting flow through the prototype culvert (the full-scale

equivalent of the model) from model studies were reported previously (5)

(6)o There are three types of similarity to be considered: 1) geometric,

2) kinematic, and 3) dynamic; they refer to simi larity of form, motion ,

and forces acting within and upon the structureo

Since it is seemingly impossib le to control the forces acting

upon the fluid masses, complete similarity is never attained; thus,

dimensional analysis must be appliedo

All dimensional units of the model are 1/lZth those of its

corresponding ful lscale prototype , and thus the corresponding headwater

e levations for the prototype are given by: 1 inch 1 foot, since a
=

scale modulus of 1:12 was used in construction of the modelo An ap

proximate conversion factor for determining the discharge through the

prototype under conditions of weircontro l may be found as fol lows :

Q(p) =
(p) L(p) Hw(p) 3/2
1
Q(m) (m) L(m) Hw(m) 3 2

where (p) and (m) subscripts denote prototype and model values respec

tivelyo The similitude relationships between L(p) and L(m) and

Iiw(p) and Iiw(m) are: L(p) = 12 L(m) and Iiw(p) = 12 Iiw(m); thus, the

- 33 -

34

equation becomes:

3/ 2
Q(p) _ Cw(p) 12 L(m) (12 I-Iw(m)

3 /Z
Q (m) --rw(m) L(m) Cf\v(m) )

Q(p) Cw(p)
=
(12)5/2 X

Q(m) Cw(m)

The factor becomes unity since Cw(p) and Cw(m) are dimensionless

/
and are nearly equa l , Q(p) is thus (12)5 2 times Q(m) or approximately

500 Q (m) ' Similar relationships for orifice, shorttube , and pipe

control may be developed from their respective basic eq11ations ,

As noted above, the coefficients, Cw(p) and Cw were dimen

(m)
sionless and were assumed to be nearly equal; hence, the ratio of Q (m)

to Q(p) is who lly dependent upon the scale modulus, Alternatively, then

the use of the respective coefficients in the basic equations and in

conjunction with fullscale dimensions would be a means whereby full

scale design calculations may be made from res11lts of the study,

 REFERENCES

L Kessler, L, H.; Experimental Investigation of Drop Inlets and Spil l

wys for Erosion Control Structure;L, Madison: University of
Wisconsin Exper1ment Station Bullretin No , 80, 1934 ,

2, Huff, A, N,: The Hydraulic Design of Rectangular Spillways, U. S.

3, B laisde ll, F , W , and Donnel ly, Co A.; Capacit{' of Box Inlet Spill
ways Under Free Flow and Submered Flow COnctitlons, St , Anthony
Falls Hydrauhc Laboratory ; Umversity of Minnesota Technical Paper
No, 7, Series B,, 1951 ,

4, B laisde ll, F, W, and Donnel ly, C. A.; Htdraulic Desi!\!3 of Drpp

S2i llways, St , Anthony Falls Hydrau1ics aboratory, Un1vers1ty of
":;nnesota, Technical Paper No, 8 , Series B , , 1951,

5, West, E, M, and Hughes , R, D.; "The Development and Use of Hydraulic

Mode ls in a Study of Culvert Performance," Bulletin No, 41, Engineer
ing Experiment Station, Col lege of Engineering, Uri1vers1ty of Kentucky,
1956 , (Also see "Report No, 1 on Hydraulic Model Studies, ,", Reports
of the Hi hwa Materials Research Laborato , VoL 10, Part I I I, 1956,
an u es , , , ; xper1menta esearc on Highway Box( ) Cu lvert(,)
Through Use of Mode ls," M,S, in C. E, Thesis, University of Kentucky,
1957 (unpublished) ,

6, West , E. M,; "Model Study of Flow Through Culverts,"llulletin No , 40

(Some Technical Papers, Kentucky Highway Conference, March 28 Z9,
1956), Engineering ExpeTiment Station, University of Kentucky, 1956,

35 "
 BIBLIOGRAPHY

L Addison, H, L,; HydE_aulic Measurements, New York, John Wiley &

Sons, 194L

2, Allen, J,, Jr.; Sc[!le 1odels in" Hydraulic Engineering, New York:
Logmans, Green & CO, 1941

3, American Soceity of Civil Engineers; Hydraulic Models. Manual of

Engineering Practice No, 25, 1942,

4. Birkhoff, G.; Hydrodynamics, New York: Maple Press, 1950.

s. (?
Blaisdell, F. W. and Donnelly, C. A.; C acitr of Box Inlet Spill
wa s Under Free F low and Submer ed Flow ondit1on, St, Anthony
Fal s y rau 1cs a oratory: 1vers1ty o nnesota Technical
Paper No. 7, Series B, 1951.

6. t
Blaisdell, F. w. and Donnelly, c. A.; H draulicDesign of Drop
Spillway,. St, Anthony Falls Hydraulic aboratory, Univers1ty of
Minnesota, Technical Paper No. 8, Series B, 1951.

7. l
Blaisdell, F. W,; !;Jl::draulics of Closed Conduit S illways - Part I -
Theory and A ation, St. Anthony Falls Hydrau ics Laboratory;
ITn1versity o nllesota, Technical Paper No, 12, Series B, 1952,

8, Bridgman, P, W,; Dimensional Anal_ys, New Haven: Yale University

Press, 1943.

9 Han_dpoo_of Applied Hydraulics. New York: McGrawo

trt, lSZ:.;
10, Daugherty, R, L,; and Ingersoll, A. C,; Fluid Mechanics. New
York: McGraw-Hill, 1954.

lL Daugherty, R, L,; rJydraulics - Text on Practical Fluid Mechanics.

New York: McGraw-"Fhll, 1937.

12, Donnelly, C. A., and Blaisdell, F, W.; Straight Drop}?Pillway

Stillin& Basin, St, Anthony Falls Hydrauhcs Laboratory:
University of Minnesota Technical Paper No, 15, Series B, 1954,

13. Huff, A, N.; The Hydraulic Desil[! of Rectangularillways. U. S,

Department otAgriculture, Soi1Conservat1on Service, Washmgton,
D, C,, SCS-TP-71, Feb. 1944,

14. Kesler, Lo Ho; !perimeal Investigation of Drop Inlets and

SJ;.!_llwa.rs for Eros10n "Control Struct11res,. Madison: Univers1ty of
\hsconsin Expenment Station Bulletin No, 80, 1934o

- 36
 Bibliography (Continued)

15, King, H , W, ; !i_andbook of Hydraulic_:;_, New York : 1cGrawHi l l , 1939 ,

16 , King, H , W , , Wisler , C . o , , and Woodburn, J, c. ; !:!t.d;aulic1_.
New York : John Wiley & Sons , 1949 .

French , J , L , ; First Progress Retort on Hdraulics of Short Pipes

and Hydraulic Charactenstlcs of oimllonl sed P1pe Entrances ,
17 ,

l'1asinngton : Bureau of Standards Report o , 4444 ,

18, Haimllond' , H , P , ; Hydraulics o Scranton : International Textboosk ,

1942,

19, rphy , G, ; chanics of Fluids , Scranton : International Text

books , 1941 o

20, Pickels , G, W, ; Drainage and Flood Control Engineering, New York :

McGrawHil l , 1941 .

21 , Powe l l , R , W , ; An E lementary Text in Hydraulics and Fluid chanics ,

New York : McMillan, 1951.

22 , Vennard , J , K , ; E lementary Fluid chanics ,

- -
New York : John Wiley &
Sons , 1954,

23, West , E , Mo and Hughes , R , D . ; "The Development and Use of Hydrau

lie Models in a Study of Culvert Performance , " Bul letin No, 41 1
Engineering Experiment Station , College of Engineenng , Dmvers1ty
of Kentucky, 1956 (Also see "Report No, 1 on Hydraulic Model
Studies , , , ," Reports of the Hi fflw a Materials Research Laborato, ,
VoL 10, Part 111 , 1956 ; and ii es , , , ; xpenmenta Researc
on Highway Box()Culverts ( ,) Through use of Models , " M, S, in CEo
Thesis , University of Kentucky, 1957 (unpub lished) ,

24, Wes t , E , M, ; "Model Study of Flow Through CUlverts ," Bulletin No,
(Some Technical Papers , Kentucky Highway Conference;" March 28
ZY , 1956) , Engineering Experiment Station , University of Kentucky ,
1956 ,

37