Module
3
Irrigation Engineering
Principles
Version 2 CE IIT, Kharagpur

Lesson
10
Distribution and
Measurement
Structures for Canal
Flows
Version 2 CE IIT, Kharagpur

Instructional objectives
On completion of this lesson, the student shall learn:
1.
2.
3.
4.
5.

What are flow distributing and measurement structures in a canal system
What is the function of outlets and what are their classifications
What are modules
How are notches and weirs used to measure flow in a canal
How are flumes used to measure flow in a canal

3.10.0 Introduction
The flow of a main canal bifurcating into a branch canal with the rest flowing
downstream is controlled with the help of a cross regulator across the parent
canal and a head regulator across the branch canal. At times, the flow of a canal
divides into two or three smaller branch canals without any regulating structure
by designing the entrance of the canals IN such a way that the flow enters each
branch canal proportionate to its size. Again, from a canal, outlet structures may
take out water for delivery to the field channel or water courses belonging to
cultivators. These outlet works, of course, are generally not provided on the main
canal and branches, but are installed in the smaller distributaries. Apart from
these, there could be a need to measure the flow in a canal section and different
structures have been tried, mostly based on the formation of a hydraulic jump
and calibrating the discharge with the depths of flow. Typical structures of these
kinds are graphically represented in Figure 1 and this lesson deals with each
type in detail.

Version 2 CE IIT, Kharagpur

3.10.1 Flow distributing structures
The flow of a canal can be distributed in to smaller branches using a variety of
structures which have been developed to suit a wide variety of conditions. The
flow being diverted in to each branch is usually defined as a proportion of the
total flow. Thus, these flow distributing structures differ from the flow regulating
structures discussed in Lesson 3.9 since the latter are designed to draw off any
amount of discharge irrespective of the flow in the parent channel. The flow
distributing structures require a control section in both the off-take channel and in
the parent channel. Flow distributors of fixed proportion type are generally used
in India, whereas in some countries a flow splitter with a mechanical arrangement
is used to change the flow distribution proportions.
The Punjab type proportional distributor has each opening or offtake constructed
as a flume or free overfall weir and is dimensioned so as to pass a given
fraction of the total flow. The controlling section consisting of the flume, elevated
floor or weir crest is located in the individual offtakes, and not in the supply
channel. A typical plan of a proportional distributor with two offtakes is shown in
Figure 2.

Version 2 CE IIT, Kharagpur

Proportional distributors may have only one offtake as shown in the typical plans
shown in Figure 3. Generally, all offtakes should be designed to bifurcate at 600
or 450. The crest of all the offtakes and the flume in the parent channel should be
at the same level and at least 0.15m above the downstream bed level of the
highest channel. The parent channel flume may have provisions for a stop log
insertion for emergency closures.

Version 2 CE IIT, Kharagpur

3.10.2 Canal outlet
Canal outlets, also called farm turnouts in some countries, are structures at the
head of a water course or field channel. The supply canal is usually under the
control of an irrigation authority under the State government. Since an outlet is a
link connecting the government owned supply channel and the cultivator owned
field channel, the requirements should satisfy the needs of both the groups.
Version 2 CE IIT, Kharagpur

Since equitable distribution of the canal supplies is dependent on the outlets, it
must not only pass a known and constant quantity of water, but must also be
able to measure the released water satisfactorily. Also, since the outlets release
water to each and every farm watercourse, such structures are more numerous
than any other irrigation structure. Hence it is essential to design an outlet in
such a way that it is reliable and be also robust enough such that it is not easily
tampered with. Further the cost of an outlet structure should be low and should
work efficiently with a small working head, since a larger working head would
require higher water level in the parent channel resulting in high cost of the
distribution system. Discharge through an outlet is usually less than 0.085
cumecs.
Various types of canal outlets have been evolved from time to time but none has
been accepted as universally suitable. It is very difficult to achieve a perfect
design fulfilling both the properties of ‘flexibility’ as well as ‘sensitivity’ because of
various indeterminate conditions both in the supply channel and the watercourse
of the following factors:
•
•
•
•

Discharge and silt
Capacity factor
Rotation of channels
Regime condition of distribution channels, etc.

These modules are classified in three types, which are as follows:
(a) Non-modular outlets
These outlets operate in such a way that the flow passing through them is
a function of the difference in water levels of the distributing channel and
the watercourse. Hence, a variation in either affects the discharge. These
outlets consist of regulator or circular openings and pavement. The effect
of downstream water level is more with short pavement.
(b) Semi-modular outlets
The discharge through these outlets depend on the water level of the
distributing channel but is independent of the water level in the
watercourse so long as the minimum working head required for their
working is available.
(c) Module outlets
The discharge through modular outlets is independent of the water levels
in the distributing channel and the watercourse, within reasonable working
limits. This type of outlets may or may not be equipped with moving parts.
Though modular outlets, like the Gibb’s module, have been designed and
implemented earlier, they are not very common in the present Indian
irrigation engineering scenario.

Version 2 CE IIT, Kharagpur

The common types of outlets used in India are discussed in the next sections.

3.10.3 Pipe outlets
This is a pipe with the exit end submerged in the watercourse (Figure 4). The
pipes are placed horizontally and at right angles to the centre line of the
distributing channel and acts as a non-modular outlet.

Discharge through the pipe outlet is given by the formula:
Q = CA (2gH) 1/2

(1)

In the above equation, Q is the discharge; A is the cross sectional area; g is the
acceleration due to gravity; H is difference in water levels of supply channel and
watercourse and C is the coefficient of discharge which depends upon friction
factor (f), length (L) and diameter of the outlet pipe (d) related by the formula:

C=

1
2 × 10 5

d
⎛
1.5d
f ⎜⎜ L +
400 f
⎝

⎞
⎟⎟
⎠

(2)

Version 2 CE IIT, Kharagpur

The coefficient f is the fluid friction factor and its value may be taken as 0.005
and 0.01 for clear and encrusted iron pipes respectively. For earthenware pipes,
f may be taken as 0.0075. All other variables are in SI units, that is, meters and
seconds.
It is a common practice to place the pipe at the bed of the distributing channel to
enable the outlets to draw proportional amount of silt from the supply channel.
The entry and exit ends of the pipe should preferably be fixed in masonry to
prevent tampering. Since the discharge through this type of outlet can be
increased by lowering the water surface level of the watercourse (thus increasing
the value of H in the discharge equation), it is possible for the irrigator to draw
more than fair share of water. A pipe outlet may also be designed as a semimodular outlet, that is, one which does not depend upon the water level in the
watercourse by allowing it to fall freely in to the watercourse (Figure 5).

Pipe outlets require minimum working head and have higher efficiency. It is also
simple and economical to construct and is suitable for small discharges.
However, these outlets suffer from disadvantages like the coefficient discharge
which varies from outlet to outlet and at the same outlet at different times apart
from the possibility of tampering in the non-modular type.

Version 2 CE IIT, Kharagpur

3.10.4 Open flume outlets
This is a smooth weir with a throat constricted sufficiently long to ensure that the
controlling section remains with in the parallel throat for all discharges up to the
maximum (Figure 6). Since a hydraulic jump forms at the control section, the
water level of the watercourse does not affect the discharge through this type of
outlet. Hence this is a semi-modular outlet.

This type of structure is built in masonry, but the controlling section is generally
provided with cast iron or steel bed and check plates. The open flumes can either
be deep and narrow or shallow and wide in which case it fails to draw its fair
share of silt. Generally, this type of outlet does not cause silting above the work,
except when supplies are low for a considerable length of time. The silt which
gets accumulated gets washed away during high supplies.
The open flume outlet is also cheaper than the Adjustable Proportional Module
(APM), discussed below. The discharge formula for the open flume outlet is
given as:

Version 2 CE IIT, Kharagpur

Q = C Bt H3/2

(3)

Where Q (given in l/s) is related to the coefficient of discharge, C, as given in the
table below; Bt is the width of the throat in cm; and H is the height of the full
supply level of the supply channel above the crest level of the outlet in cm.
B

B t (cm)

C

6 to 9

0.0160

> 9 to12

0.0163

> 12

0.0166

The minimum head required to drive the outlet is about 20 percent of H.

3.10.5 Adjustable Proportional Module (APM)
There are various forms of these outlets but the earliest of them is the one
introduced by E.S. Crump in 1992. In this type of outlet, a cast iron base, a cast
iron roof block and check plates on either are side are used to adjust the flow and
is set in a masonry structure (Figure 7). This outlet works as a semi-module since
it does not depend upon the level of water in the watercourse.

Version 2 CE IIT, Kharagpur

The roof block is fixed to the check plates by bolts which can be removed and
depth of the outlet adjusted after the masonry is dismantled. This type of outlet
cannot be easily tampered with and at the same time be conveniently adjusted at
a small cast.

Version 2 CE IIT, Kharagpur

The roof blocks may also be built of reinforced concrete. The face of the roof
block is set 5 cm from the starting point of the parallel throat. It has a lamniscate
curve at the bottom with a tilt of 1 in 7.5 in order to make the water converge
instead of a horizontal base which would cause it to diverge. The cast iron roof
block is 30cm thick.
As such, the APM is the best type of outlet if the required working head is
available and is the most economical in adjustment either by raising or lowering
the roof block or crest. However, it is generally costlier than the other types of
outlets and also requires more working head.
The discharge formula for this type of weir is given as:
Q = C Bt H 1(H2)1/2

(4)

Where Q (given in l/s) is related to the coefficient of discharge, C, which is taken
equal to around 0.0403; Bt is the width of the throat in cm; H1 is the depth of
head available, that is the difference between the supply channel full supply level
and the outlet bed (crest) level; and H2 is the difference between the supply
channel full supply level and the bottom level of the roof block (Figure 8) .
B

The base plates and the roof block are manufactured in standard sizes, which
with the required opening of the orifice are used to obtain the desired supply
through the outlet.

Version 2 CE IIT, Kharagpur

3.10.5 Tail clusters
When the discharge of a secondary, tertiary or quaternary canal diminishes
below 150 l/s, it is desirable to construct structures to end the canal and
distribute the water through two or more outlets, which is called a tail cluster.
Each of these outlets is generally constructed as an open flume outlet (Figure 9).

3.10.6 Flow measurement in canals
The available water resources per person are growing scarcer with every passing
day. Although a region may not face a net reduction in water resources, the
increasing population of the area would demand increased food production and
consequently, agricultural outputs. Such that an equitable distribution of water is
ensured as far as possible with a command area, it is required to measure water
at important points in the canal network. Measurements may also help in
estimating and detecting losses in the canal.

Version 2 CE IIT, Kharagpur

Further, at the form level, advanced knowledge of soil properties and soil
moisture / plant relationships permits irrigation systems to be designed so that
water can be applied in the right amount and at the right time in relation to the
soil moisture status thereby obtaining maximum efficiency of water use and
minimum damage to the land. This knowledge can be utilized most effectively
only by reasonably accurate measurement of the water applied.
The amount of water being delivered to a field of an irrigator should also be
measured in order to make an assessment of water charges that may be levied
on him. If the charge to the user of canal water is based on the rate flow, then
rate-of-flow measurements and adequate records are necessary. Charges on the
basis of volume of water delivered necessitate a volumetric measuring device.
Ideally, water flow should be measured at intakes from storage reservoirs, canal
head works, at strategic points in canals and laterals and at delivery points to the
water users. The most important point for measurement is the form outlet which
is the link between the management authority of the canal system and the user.
The degree of need for a measuring device at the outlet varies according to the
delivery system employed. Delivery on demand usually relies up on the
measurement of water as a basis for equitable distribution as well as for
computing possible water charges. Where water is distributed by rotation among
farmers along a lateral (or distributary or minor canal) and the where the amount
of water supplied to each farmer may be different, a measuring device at the
turnout is required. On the other hand, if farmers along a lateral receive water on
the basis of area of land or crops irrigated measurement is not entirely
necessary, but may still be desirable for other purposes, such as improvement of
irrigation efficiency.
Similarly in all systems based on constant flow,
measurement is not entirely necessary but may be advantageous.
Where several farmers share the water of each outlet and the flow in the canal
fluctuates considerably, each such outlet should be equipped with a measuring
device, even if equitable distribution among outlets is practiced, so that each
group of farmers will know the flow available at any one time from their
respective outlet.
Amongst the methods and devices used for measuring water in an irrigation
canal network, the weir is the most practical and economical device for water
measurement, provided there is sufficient head available. Measuring flumes are
also used in irrigation networks and their advantage are smaller head losses,
reasonable accuracy over a large flow range, insensitivity to velocity of approach,
and not affected much by sediment load. Propeller meters are used in many
countries and are particularly suited to systems where no head loss can be
permitted for water measurement and where water is sold on volumetric basis.
For water measurement in small streams, particularly in field ditches and furrows
and where head losses must be small, the deflection or vane meter has proved

Version 2 CE IIT, Kharagpur

to be a useful device. Only the weir and the standing wave (hydraulic jump) type
flume are discussed in this lesson as these are most commonly used.

3.10.7 Weirs
Weirs have been in use as discharge measuring devices in open channels since
almost two centuries and are probably the most extensively used devices for
measurement of the rate of flow of water in open channels. Weirs may be divided
in to sharp and broad crested types. The broad crested weirs are commonly
incorporated in irrigation structures but are not usually used to determine flow.
The types of sharp crested weirs commonly used for measuring irrigation water
are the following:
3.10.7.1 Sharp crested rectangular weir
A general view of this type of weir is shown in Figure 10.

Version 2 CE IIT, Kharagpur

Amongst the many formulae developed for computing the discharge of
rectangular, sharp crested weirs with complete contraction, the most accepted
formula is that by Francis and is given as:
Q = 1.84 (L – 0.2H) H3/2

(5)

Where Q is the discharge in m3/s; L is the length of the crest in meters; and H is
the head in meters, that is, the vertical difference of the elevation of the weir crest
and the elevation of the water surface in the weir pool.

Version 2 CE IIT, Kharagpur

3.10.7.2 Sharp crested trapezoidal (Cipolletti) weir
A general view of this type of weir is shown in Figure 11.

The discharge formula for this type of weir was given by Cipoletti as:
Q = 1.86 L H3/2

(6)

Where Q is the discharge in m3/s; L is the length of the crest in meters; and H is
the head in meters. The discharge measurements using the above formula for
the trapezoidal weir are not as accurate as those obtained from rectangular weirs
using the Francis formula.

Version 2 CE IIT, Kharagpur

3.10.7.3 Sharp sided 900 V-notch weir
A general view of this type of weir is shown in Figure 12.

Of the several well known formulae used to compute the discharge over 900 Vnotch weirs the formula recommended generally is the following:
Q = 8/15 (2gCd)1/2 H 5/2

(7)

Where Q is the discharge in m3 /s; g is the acceleration due to gravity (9.8m/s2);
Cd is a coefficient of discharge; and, H is the head in meters. The value of Cd
varies according to the variation of H and can be read out from (Figure 13).

Version 2 CE IIT, Kharagpur

Version 2 CE IIT, Kharagpur

Each of these weirs has characteristics appropriate to particular operating and
site conditions. The 900 V-notch weir gives the most accurate results when
measuring small discharges and is particularly suitable for measuring fluctuating
flows. Weirs require comparatively high heads, considerable maintenance of the
weir or stilling pool and protection of the channel downstream of the crest.

3.10.8 Flumes
Flumes are flow measuring devices that works on the principle of forming a
critical depth in the channel by either utilizing a drop or by constricting the
channel. These two forms of flumes for flow measurement are described below.
3.10.8.1 Flume with a vertical drop
This type of structures has already been discussed in Lesson 3.9 where it was
shown to be utilized to negotiate a fall in the canal bed level. One of these, the
standing wave flume fall developed at the Central water and Power Research
Station (CWPRS), Pune, has been standardized and documented in Bureau of
Indian Standard code IS: 6062-1971 “Method of measurement of flow of water in
open channels using standing wave flume-fall” and shown in (Figure 14)
Because of the inherent free flow conditions the measurement of flow requires
only one gauge observation on the upstream side.

Version 2 CE IIT, Kharagpur

The discharge equation for this structure is given by the following equation:
Q = 2/3 (2g)1/2 CBH1.5

(8)

Where Q is the discharge in m3/s; g is the acceleration due to gravity; C is the
coefficient of discharge (=0.97 for 0.05 < Q <0.3 m3/s and = 0.98 for 0.31< Q
<1.5m3/s); B is the width of the flumed section, also called the throat and H is the
total head, that is, the depth of water above crest plus the velocity head.
The other type of flume type of fall is the one called the Central Design Office
(CDO) Punjab type fall, which is simple and robust in construction. Up to 1m
drop, a glacis is used on the downstream side (Figure 15) and if the drop
exceeds 1m, the crest ends in a drop wall (Figure 16). The structure is often
combined with a bridge, an intake of a third degree canal or both.

Version 2 CE IIT, Kharagpur

3.10.8.2 Flume with a constricted section
This type of structures for measuring water discharge creates a free flow
condition followed by a hydraulic jump by providing a very small width at some
point with in the flume. These are also further divided in to two types: Long and
short-throated flumes. In the former, the constriction is sufficiently long to
produce flow lines parallel to the flume crest, for which analytical expression for
discharge may be obtained. In the short-throated flumes, the curvature of water
surface is large and the flow in the throat is not parallel to the crest of the flume.
Hence, due to the non-hydrostatic pressure distribution, there is not analytically
derived expression for discharge but has to be calibrated from actual
measurements. However, these flumes require small lengths and are economical
than long throated flumes. One of the commonly used short-throated flumes is
the Parshall flume (Figure 17).

Version 2 CE IIT, Kharagpur

The flume consists of a short parallel throat preceded by a uniformly converging
section and followed by a uniformly expanding section. The floor is horizontal in
the converging section, slopes downwards in the throat, and is inclined upwards
in the expanding section. The control section, at which the depth is critical,
occurs near the downstream end of the contraction. There are standard
dimensions of Parshall flumes which are available commercially and may be had
from the reference “Design of Small Canal Structures” of USBR (1978).
One of the advantages of this type of flume is that it operates with a small head
loss, which permits its use in relatively shallow channels with flat grades. For a
given discharge, the loss in head through a Parshall flume is only about one
fourth that required by a weir under similar free flow conditions. The flume is
relatively insensitive to velocity of approach. It also enables good measurements
with no submergence (that is free flow with a hydraulic jump downstream) or with
submergence (that is the jump is drowned by the downstream water level) as
shown in (Figure 17). The velocity of flow within the flume is also sufficient to
eliminate any sediment deposition within the structure during operation. A
disadvantage of the flume is that standard dimensions must be followed within
close tolerance in order to obtain reasonable accuracy of measurement. Further,

Version 2 CE IIT, Kharagpur

the flumes cannot be used close to an outlet or regulating devices. Parshall
flumes can be constructed in a wide range of sizes to measure discharges from
about 0.001m3/s to 100m3/s.

References
•

FAO Irrigation and Drainage paper 26/1: Small Hydraulic Structures, Volume
1 (1982)

•

FAO Irrigation and Drainage paper 26/2: Small Hydraulic Structures, Volume
2 (1982)

•

Garg, S K (1996) “Irrigation engineering and hydraulic structures”, Twelfth
Edition, Khanna Publishers

•

IS 12331 : 1988 General Requirements for Canal Outlets

•

IS 7986 : 1976 Code of practice for canal outlets

•

IS: 6062-1971, “Method of measurement of flow of water in open channels
using standing wave flume-fall”

•

USBR (1978), “Design of Small Canal Structures”

•

Varshney, R S, Gupta, S C and Gupta, R L (1993) “Theory and design of
irrigation structures”, Volume II, Sixth Edition, Nem Chand Publication

Version 2 CE IIT, Kharagpur