Culvert Design

Prof. A. Taigbenu
School of Civil & Environmental Engineering
University of the Witwatersrand

Culverts
 A culvert is a short underground conveyance system that allows
water to flow through an obstruction such as a highway or
railway embankment or a large superstructure

1
 Culvert entrance and end treatments
 Entrance and end treatments are required to stabilise the
embankment slopes around the entrance and exit of the culvert.

 They help to blend the culvert with the embankment slope.

P6 - Taigbenu 3

Culvert entrance and end treatments (II)

 They also influence the hydraulic efficiency at the entrance

(achieve lower headwater elevation at the entrance with
increased discharge through the culvert).

P6 - Taigbenu 4

2
 Culvert entrance and end treatments (III)

Projecting barrel Headwall (Endwall) & wingwalls

Mitered end
5
P6 - Taigbenu

Culvert hydraulics
 Flow in culverts is complex and takes place within its short length;

 Flow is a combination of gradually varied flow and rapidly varied flow;

 Using varied flow calculations for the hydraulic analysis is tedious, and
generally not followed in design;

 Two flow control assumptions that are followed in culvert design are Inlet
Control and Outlet Control assumptions.

3
 Inlet Control
 Culvert’s slope is hydraulically steep;

 Flow is controlled at the inlet, therefore flow in the culvert is

supercritical;
 Critical depth occurs in the vicinity of the entrance of the
culvert;

 Water surface profile in the culvert approximates the S2;

 With this condition, culvert barrel is capable of conveying a

greater discharge than the inlet will accept;

 Two situations that could occur are presented below

Inlet control cases

Submerged Inlet Control

Unsubmerged Inlet Control

4
 Inlet control hydraulics
The inlet of the culvert behaves as a weir when the head water surface
elevation is less than the height of the culvert barrel, D. It is described
by the relationship;

M
HW  1.811Q 
 K 1/ 2 
D  AD 
Where
HW = head water elevation above the invert at entrance of culvert (m)
D = height of the culvert barrel (m)
A = full cross-sectional area of the culvert barrel (m2)
Q = discharge (m3/s)
K and M are constants that are presented in the Table below
The above equation applies when Q/AD1/2 < 1.93

Constants for inlet control

10

5
 Submerged Inlet control
When the head water surface elevation is greater than the height of the culvert
barrel, D, so that its entrance is submerged, the flow into the culvert is
considered as orifice flow that is described by the relationship
2
HW  1.811Q 
 c 1/ 2 
 Y  0.5  S0
D  AD 

Where
HW = head water elevation above the invert at entrance of culvert (m)
D = height of the culvert barrel (m)
A = full cross-sectional area of the culvert barrel (m2)
Q = discharge (m3/s)
c and Y are constants that are presented in the Table above
The above equation applies when Q/AD1/2 > 2.2

Replace the slope term (-0.5S0) with (+0.7S0) if the culvert entrance is mitered.

11

Outlet Control
 Outlet control occurs when the culvert slope is hydraulically mild or
when the tailwater elevation is high enough to affect the head water
elevation;

 Flow is controlled at the outlet, therefore flow in the culvert is subcritical;

 Analysis of flow in the culvert is carried out using energy loss

relationships for a pipe flowing full;

 With this condition, inlet is capable of accepting a higher discharge than

the downstream flow condition will permit;

 Two situations that could occur are presented below

12

6
 Outlet control cases

Submerged inlet and outlet

submerged inlet and unsubmerged outlet; Culvert flows full

and flows out freely.

13

Outlet control case 1

 Submerged inlet and outlet

 2 gn 2 L  V 2
hl   ke  1  4 / 3 
 R  2g
HW  hl  TW  S0 L
Where ke is the entrance loss coefficient (see Table below)
n = Manning’s coefficient; R = hydraulic radius = A/P (m);
V = average velocity in the culvert (m/s);
L = length of culvert (m)
14

7
 Entrance loss coefficient for various entrances

15

Outlet control case 2

 Submerged inlet and unsubmerged outlet

 2 gn 2 L  V 2
hl   ke  1  4 / 3 
 R  2g

HW  hl  D  S 0 L
Where ke is the entrance loss coefficient (see Table below)
n = Manning’s coefficient; R = hydraulic radius = A/P (m);
V = average velocity in the culvert (m/s);
L = length of culvert (m), D = height of the culvert barrel (m)

16

8
 Culvert Example 1
A circular culvert (L=30m long, n = 0.014, ke = 0.5; S0 = 0.003) under a
roadway is expected to carry a flow of 5m3/s. If the maximum permissible
headwater is 4m above the top invert, what size of pipe culvert would you
select ? The outlet discharges full and freely.

Only outlet control is examined because the question indicates that the
outlet discharges full and freely
 2 gn 2 L  V 2
hl   ke  1  4 / 3 
 R  2g
17

Culvert Example 1 (II)

ke = 0.5, R = D/4, n=0.012, L = 30m, hl = 4m+0.003×30m = 4.09m

 19.62  0.0122  30  V 2  0.538  V 2

4.09  1.5  4.09  1.5  4 / 3 
 ( D / 4) 4 / 3  2g  D  2g
 

4Q 45 6.3662 V 2 2.066

V   
D 2
D 2
D 2
2g D4

 0.538  2.066
4.09  1.5  4 / 3  4
 D  D

By trial-and-error, D = 1.007m. Consider using a standard culvert size of 1200mm.

18

9
 Culvert Example 2
A circular concrete culvert (L=18m long, D=600mm, n = 0.014, ke = 0.5; End conditions:
K=0.018, M=2, c=0.0292, and Y=0.74) is laid on a slope of 0.005 and expected to allow a
flow of 1.5m3/s to pass through it.
(a) For a constant tailwater of 0.2m, what is the headwater depth for both inlet and
outlet control conditions?
(b)Is the culvert flowing under inlet or outlet control conditions?
(c) What would be the result if the tailwater depth were 0.8m?

2
HW  1.811Q 
Submerged Inlet control:  c 1/ 2 
 Y  0.5  S0  HW  3.14m
D  AD 
Q
 6.85
AD1 / 2
P6 - Taigbenu 19

Culvert Example 2 (II)

Outlet control:  2 gn 2 L  V 2
hl   ke  1  4 / 3 
 R  2g

ke = 0.5, R=D/4=0.15m, n=0.014, L=18m, V=4Q/πD2=5.31m/s,

V2/(2g)=1.434m
HW  hl  D  S 0 L  3.40  0.6  0.09  3.91m
hl = 3.40m

HW from outlet control is higher than that from inlet control, so outlet control
prevails, and HW=3.91m
20

10
 Culvert Example 2 (III)

With the new outlet condition that is submerged, inlet submerged control
produces HW=3.14m as before.

Outlet control gives:

HW  hl  TW  S 0 L  3.40  0.8  0.09  4.11m
The head loss, hl from outlet control is as before, hl = 3.40m

Outlet control still prevails, and HW=4.11m

21

Culvert Example 3
A circular culvert (L=11m long, n = 0.013, ke = 0.5; End conditions: K=0.0098,
M=2, c=0.0389, and Y=0.67) is laid on a slope of 0.03 and expected to allow a
flow of 1.4m3/s to pass through it with a maximum allowable headwater
elevation, HW=1.8m. Determine the size of the culvert barrel to convey this flow
with TW=0.15m.
Examine unsubmerged Inlet control:
M 2
HW  1.811Q  1.8  1.8111.4  4 
 K    0.0098   D  0.00567  D  0.49m
4
D  AD1/ 2  D  D 2.5 
Unsubmerged inlet condition cannot be valid because HW > D; and this is
Q
further supported by the fact that  10.6  1.93
AD1 / 2

Examine Submerged Inlet control:

2 2
HW  1.811Q  1.8  1.8111.4  4 
 c   Y  0.5  S0   0.0398   0.67  0.015
D  AD1 / 2  D  D 2.5 

1.8 0.415
  0.67  0.015
D D5
Q
The value of D that satisfies the above equation is 0.75m, and  3.7  2.2
AD1 / 2
22

11
 Culvert Example 3 (II)

Examine Outlet control:  2 gn 2 L  V 2

hl   ke  1  4 / 3 
 R  2g
ke = 0.5, R=D/4, n=0.013, L=11m, Q=1.4m3/s V=4×1.4/πD2=1.783/D2,
V2/(2g)=0.162/D4

HW+S0L = hl+D → 1.8+0.03×11 = D + (1.5+0.232/D4/3)0.162/D4

2.13 = D + (1.5+0.232/D4/3)×0.162/D4

The value of D that satisfies the above equation is 0.68m

23

Culvert Example 3 (III)

Comparing the value of D=0.75m obtained from inlet control and the
value of D=0.68m from Outlet control, design considerations require that
the more conservative value of D=0.75m be used as the size of the barrel
of the culvert.

Recommended culvert barrel diameter = 0.75m.

24

12
 Culvert Example 4
Twin 1220mm by 910mm box culverts (L=31m long, n = 0.013, ke = 0.5; 0 ×
wingwall flares: K=0.061, M=0.75, c=0.0423, and Y=0.82) is laid on a 1% slope
and expected to allow a flow of 8.5m3/s to pass through it. TW=0.61m
a) What is the headwater depth?
b) Are the culverts flowing under inlet or outlet control conditions?

Unsubmerged Inlet control should not be evaluated because the critical value of HW
should be the largest of all the possible control conditions.

25

Culvert Example 4 (II)

Examine Submerged Inlet control:

2 2
HW  1.811Q  HW  1.811 8.5 
 c   Y  0.5  S 0   0.0423   0.82  0.005
D  AD1 / 2  1.22  2.2204  1.220.5 
HW = 3.03m

 2 gn 2 L  V 2
Examine Outlet control: hl   ke  1  4 / 3 
 R  2g

ke = 0.5, n=0.013, L=31m, Q=8.5m3/s, A = 2.2204m2, P = 8.52m, R = A/P = 0.2606m,

V = Q/A = 8.5/2.2204 = 3.828m/s
 2  9.81 0.0132  31  3.8282
hl   0.5  1   1.582m
 0.26064 / 3  2  9.81
 
26

13
 Culvert Example 4 (II)

HW+S0L = hl+D HW = hl+D-S0L

HW = 1.582+1.22 – 0.01×31 = 2.492m

Inlet control has a larger value of HW, and as such it is the control condition
that prevails.

HW = 3.03m

27

End of Presentation

28

14