Flumes: Theory & Application

(.doc) (.pdf)

ABSTRACT ................................................................................................................................................................ 2
PRELUDE ................................................................................................................................................................... 3
SELECTING A PRIMARY MEASURING DEVICE: WEIR VS. FLUME ......................................................... 4
CRITICAL FLOW THEORY................................................................................................................................... 5
Energy Equation Applied to Open-Channel Flow
Specific Energy
Characteristics of Critical Flow
THEORY APPLIED TO FLUMES ........................................................................................................................ 13
General Discharge Vs. Head Formula
Submergence
FLUME CLASSES .................................................................................................................................................. 16
Long-Throated Flumes
Short-Throated Flumes
OTHER SPECIAL FLUMES .................................................................................................................................. 18
Parshall Flumes
Background
Operation
Advantages
Disadvantages
Design
Discharge Calculation
Parshall Flume in Biosolids Project
H-Flumes
Cutthroat Flumes
Palmer-Bowles Flumes
Trapezoidal Flumes
Quick Selection Guide
SITE CHARACTERISTICS RELATED TO LOCATING & SETTING FLUMES ......................................... 31
Approach Conditions
Channel Flow Characteristics and Operational Needs
Erosion and Scour
WORKMANSHIP .................................................................................................................................................... 34
ACKNOWLEDGEMENTS ..................................................................................................................................... 35
REFERENCES ......................................................................................................................................................... 36
2

Abstract

The primary objective of writing a paper that describes the theory underlying flow in
flumes is acquainting the reader with the flow-measuring device upon which most our samplers
are based. A basic derivation of the formulas governing critical flow in flumes is presented and
followed by a concise description of the different classes of flumes and their modes of design,
operation, and installation. Finally, a case study of the PVC flumes constructed for the C-11653
Biosolids Project is presented as a sample application of the published empirical dimensions and
discharge equations listed earlier.
3

Prelude

Flumes are open-channel flow sections that force flow to

accelerate. Acceleration is produced by converging the sidewalls,
raising the bottom, or a combination of both. When only the
bottom is raised with no side contractions, the flume is
commonly called a broad-crested weir. When the downstream
depth is shallow and enough convergence exists between the
upstream and downstream channels, the flow passes through
critical depth. Therefore, flumes are sometimes called critical-
depth flumes. When flow passes through critical depth, a unique
water surface profile occurs within the flume or broad-crested
weir for each discharge. This condition is known as free flow.
For this case, upstream heads at one location relative to the control bottom elevation near the
region of critical depth can be used to determine a usable head versus discharge relationship for
flow measurement.

Flume head loss is less than about one-fourth of that needed to operate a sharp-crested
weir having the same control width, and in some long-throated flumes, may be as low as one-
tenth. Another advantage compared to most standard weirs is
that for a properly designed and installed flume, the velocity of
approach is a part of the calibration equations. Unauthorized
altering of the dimensions of constructed flumes to obtain an
unfair share of water is difficult and, therefore, not likely.
Velocity of flow can usually be designed to minimize sediment
deposition within the structure. Gradual convergence sections at
the entrance tend to improve velocity distribution of approach
flow and the passage of floating debris. Some flumes can be
more expensive than sharp-crested weirs or submerged orifices
in unlined channels.
4

Selecting a Primary Measuring Device: Weir vs. Flume

The selection of a primary measuring device for a particular flow measurement

installation usually involves a series of three decisions:

§ Which primary measuring device to use: a weir or a flume?

§ Which specific type of primary device to use?
§ What is the exact size of the primary device to be installed at the location in
question?

Weirs and flumes each have decided advantages and disadvantages. A weir is the
simplest device that can be used to measure flow in open channels. It is low in cost, relatively
easy to install, and quite accurate when properly used. However, it normally operates with a
rather significant loss in head, and its accuracy can be affected by variations in the approach
velocity of the liquid in the flow channel. Moreover, it must also be periodically cleaned to
prevent deposits of sediment or solids in the upstream side of the wear, which will adversely
affect its accuracy.

On the other hand, a flume tends to be self-cleaning since the velocity of flow through it
is high and there is no obstruction across the channel. It can also operate with a much smaller
head loss than a weir, which can be important for many for many applications where the
available head is limited. In addition, a flume is relatively less sensitive to varying approach
velocity. Nonetheless, a fume is much more costly than a weir, and its installation is more
difficult and time consuming. Flumes are also generally less accurate that weirs.

It should be noted that the initial installation cost is not the only expense that should be
considered when choosing a primary device. As mentioned earlier, flumes tend to be self-
cleaning, whereas weirs must be periodically cleaned. Lower maintenance costs associated with
a flume may eventually outweigh the higher initial cost.
5

Critical Flow Theory

In reference to J. A. Robertson & C. T. Crowe, “Engineering Fluid Mechanics – 6th Edition”, 1997

q Energy Equation Applied to Open-Channel Flow

The one-dimensional energy equation for open channels (see Fig. 1-a) is:

p1 v2 p v2
+ α 1 1 + z1 = 2 + α 2 2 + z 2 (1)
γ 2g γ 2g

Figure 1-a
6

It’s evident from Fig. 1-a that the following equalities hold:

p1 p
+ z1 = y1 + S 0 ∆ x and 2 + z 2 = y 2
γ γ

Where S0 is the slope of the channel bottom, and y is the depth of flow. Assuming á1 = á2
= 1, we can write Eq. (1) as:

v12 v2
y1 + + S 0 ∆ x = y 2 + 2 + hL (2)
2g 2g

Considering the special case where the channel bottom is horizontal (S0 = 0) and head
loss is negligible (hL = 0), Eq. (2) becomes:

v12 v2
y1 + = y2 + 2 (3)
2g 2g

q Specific Energy

The sum of the depth of flow and the velocity head is defined as specific energy:

v2 (4)
E = y+
2g

Thus Eq. (3) states that the specific energy at section 1 is equal to the specific energy at
section 2. The continuity equation between section 1 and 2 is:

A1V1 = A2V2 = Q (5)

Accordingly, Eq. (3) can be expressed as:

7

Q2 Q2
y1 + = y 2 + (6)
2 gA12 2 gA22

Since A1 & A2 are both functions of the depth y, the magnitude of the specific energy at
the section 1 or 2 is solely a function of the depth at each section. If, for a given channel
and given discharge, depth versus specific energy is plotted, a relationship such as that
shown in Fig. 1-b is obtained.

Figure 1-b

For a given value of specific energy, we can see that the depth may be either large or
small. Physically, this means that for the small depth, the bulk of the flow energy is in the
8

form of kinetic energy (Q2/2gA2); whereas for a larger depth, most of the energy is in the
form of potential energy.

Flow under a sluice gate (Fig. 1-c) is an example of flow in which two depths occur for a
given value of specific energy. The large depth and low kinetic energy occurs upstream
of the gate, while the low depth and large kinetic energy occurs downstream. The depths
are called alternate depths.

Figure 1-c

If we maintain the same rate of flow but set the gate with a larger opening, the upstream
depth will drop, and the downstream depth will rise. Thus we have different alternate
depths and a smaller value of specific energy than before. This is consistent with the
diagram in Fig. 1-b.
9

Finally, it can be seen in Fig. 1-b that a point will be reached where the specific energy is
minimum and only a single depth occurs. At this point, the flow is termed critical. Thus
one definition of critical flow is the flow that occurs when the specific energy is minimal
for a given discharge. The flow for which the depth is less than critical is (velocity is
greater than critical) is termed supercritical flow, and the flow for which the flow is
greater than critical is termed subcritical flow. In Fig. 1-c, subcritical flow occurs
upstream and supercritical flow occurs downstream of the sluice gate. (It should be noted
that some engineers refer to subcritical and supercritical flow as tranquil and rapid flow.)

q Characteristic of Critical Flow

It has been proven that critical flow occurs when the specific energy is minimum for a
given discharge. The depth for this condition may be determined if we solve for dE/dy
from E = y + Q2/2gA2 and set dE/dy equal to zero:

dE Q 2 dA
= 1− 3 ⋅ (7)
dy gA dy

However, dA = Tdy, where T is the width of the channel at the water surface. Then Eq.
(7), with dE/dy = 0, will reduce to:

Q 2Tc Ac Q2
= 1 and = (8,9)
gAc3 Tc gAc2

Introducing the hydraulic depth D, which is defined as A/T, then Eq. (9) becomes:

Q2 v2
Dc = = (10,11)
gAc2 g

Upon dividing Eq. (11) by Dc and taking the square root of it, we get
10

v
1= (12)
gDc

In open channel hydraulics, the Froude number is a very important nondimensional

parameter. The Froude number, Fr, is the ratio of inertia force to gravity force, which
simplifies to:

v
Fr = =1
gDc

(The Froude number is equal to unity when critical flow prevails. If the Froude number is
less than unity, the flow is subcritical and has a low velocity, which is often described as
tranquil and streaming. On the other hand, if the Froude number is greater than unity,
the flow is supercritical and has a high velocity, which is usually described as rapid,
shooting, and torrential.)

If a channel is of rectangular cross section (as in the case of a Parshall Flume), then A/T
is the actual depth, and Q2A2 = q2/y2, so the formula for critical depth, Eq. (9), becomes:

1
 q2 3 (13)
y c =  
 g 

Were q is the discharge per unit width of the channel.

Critical flow may also be examined in terms of how the discharge in a channel varies
with depth for a given specific energy. For example, consider in a rectangular channel
where

Q2 Q2
E = y+ = y+
2 gA 2 2 gy 2 B 2

If we consider a unit width of the channel and let q = Q/B, then the above equation
becomes
11

q2
E = y+
2gy 2

If we determine how q varies with y for a constant value of specific energy, we see that
the critical flow occurs when the discharge is maximum (see Fig. 1-d)

Figure 1-d

Originally, the term critical flow probably related to the unstable character of the flow for
this condition. Referring to Fig. 1-b, we notice that only a slight change in specific energy
will cause the depth to increase or decrease a significant amount; this is a very unstable
condition. In fact, observations of critical flow in open channels show that the water
12

surface consists of a series of standing waves. Because of the unstable nature of the depth
in critical flow, designing canals so that normal depth is either well above or well below
critical depth is usually best. The flow in canals and rivers is usually subcritical; however,
the flow in steep chutes or over spillways is supercritical.
13

Theory Applied to Flumes

Flumes are designed to force a transition from sub-critical to super-critical flow. The
transition is caused by designing flumes to have a narrowing at the throat, raising of the channel
bottom, or both. Such a transition causes flow to pass through critical depth at the flume throat.
At the critical depth, energy is minimized and there is a direct relationship between water depth
and flow rate. However, it is physically very difficult to measure critical depth in a flume
because its exact location is difficult to determine and may vary with flow rate. Through mass
conversion, the upstream depth is related to the critical depth. Therefore, flow rate can be
determined by measuring the upstream depth, which is a highly reliable measurement.

q General Discharge Vs. Head Formula

This equation and the specific energy equation are the basic critical flow relationships for
any channel shape.

2 3
Qc A v2
= c & E= +y
g Tc 2g

In order to derive the basic form of the discharge equation for various flumes, assume that
we have a channel of rectangular cross section. As mentioned earlier, all flumes are designed
with the intent of producing critical depth in the flume throat and thereby creating a unique
depth-discharge relationship. If this critical depth is measured, the rate of flow can easily be
computed from:

q = gy c ⇒ Q = B gy c
3 3

where B is the width of the channel normal to the flow direction.

Because yc/2 = v2c /(2g), it is easily shown that yc = 2/3E, where again, E is the total head, y +
v2approach /(2g); hence the preceding equation can be written as:
14

3
 22
3
Q = B g   E 2 ⇒ Q = 0.385 B 2 g E c
32
(13)
 
3

Had the velocity of approach been neglected, the equation can be rewritten as:

QTheoretical = 0.385 B 2 g H 3 2

If the height of the flume is relatively small, then the velocity of approach may be significant
and the discharge produced will be greater than that given by the preceding equation. To account
for these effects, a discharge coefficient C is defined as:

Q
C=
QTheoretical

Then

Q = 0.385CB 2 g H 3 2 (14)

This sums up the basic derivation of the discharge vs. head formula for most flumes.
However, extensive experimental research has been done and most flumes have a published set
of empirical formulas that are function of channel widths or other key dimensions. In the case of
the Parshall flume, for example, different flume widths (with the corresponding different
dimensions) had their discharge vs. head results plotted and fitted to power functions with K and
n as the variable parameters. Replacing the constant parts of Eq. (14) by K (which also
encompasses the throat width), and n as the head power (a value that circulates around 1.5) the
general formula becomes:

Q = KH n (15)
15

The various contributing variables to this formula are the velocity of approach, channel
width, non-horizontal channel bottom, and the desired units. This generalized equation is to be
elaborately discussed in the to the Parshall flume section of this paper.

q Submergence

All flumes have a minimum needed head loss to assure that free flow exists and that only an
upstream head measurement is needed to determine discharge rate. This required head loss is
usually expressed as a submergence limit defined by the ratio of the downstream head to the
upstream head, both referenced to the flume throat bottom. The term "modular limit" is defined
as this limiting submergence ratio for a particular flow module, which causes no more than a 1%
deviation in the upstream head reading for a given discharge. When these limits are exceeded, an
additional downstream head measurement is sometimes used to extend the measurement range of
a flume, particularly for Parshall and cutthroat flumes, but at considerable loss of accuracy.
Submergence also increases upstream channel depth, decreasing the upstream velocity, which
may aggravate sedimentation problems.
16

Flume Classes

As mentioned earlier, the two basic classes or forms of flumes are long-throated flumes
and short-throated flumes; they are discussed in depth below:

q Long-Throated Flumes:

Long-throated flumes (Fig. 2-a) control discharge rate in a throat that is long enough to
cause nearly parallel flow lines in the region of flow control. Parallel flow allows these
flumes to be accurately rated by analysis using fluid flow concepts (energy principle, critical
depth relationships, and boundary layer theory). Long-throated flumes can have nearly any
desired cross-sectional shape and can be custom fitted into most canal-site geometries. The
modified broad-crested weirs, also called ramp flumes, are styles of long-throated flumes.

Long-throated flume (broad-crested weir) under The long-throated flume (broad-crested weir)
3
construction. with approximately 1,200 ft /s.
Water Resources Research Laboratory, “Water Measurement Manual – 3rd Edition” Figure 2-a
17

q Short-Throated Flumes:

Short-throated flumes are considered short because they control flow in a region that
produces curvilinear flow. Although they may be termed short throated, the overall specified
length of the finished structure, including transitions, might be relatively long. The Parshall
flume is the most common example of this type of flume (Fig. 2-b). These flumes would
require detailed and accurate knowledge of the individual streamline curvatures for
calculated ratings, which is usually considered impractical. Thus, calibrations for short-
throated flumes are determined empirically by comparison with other more precise and
accurate water measuring systems.

“Water Measurement Manual – 3rd Edition” Figure 2-b

Figure 2

The long-throated flume (broad-crested weir)

3
with approximately 1,200 ft /s.
18

Other Special Flumes

Most tables and diagrams are from Grant & Dawson, “ISCO Open Channel Flow Measurement Handbook – 5th Edition”, 1998.

Most flumes in common use today can be traced to one of three early design sources:
rectangular English flumes based upon early work in India around 1908-1914; the Parshall
flume whose forerunner, a Venturi flume developed by Cone, was extensively modified and
tested by Parshall; and flumes of the type first developed by Palmer and Bowles. The follow
section will concisely discuss some of the more popular flume designs currently in use. Included
are discussions of the H-flume, Cutthroat flume, Palmer-Bowles, the trapezoidal flume, and a
detailed discussion of the Parshall Flume.

q Parshall Flumes (in depth):

§ Background:

A very common critical-flow

flume is the Venturi Flume (see
Figure 1.), which was developed and
calibrated by Dr. Parshall of the U.S.
Soil Conservation Service in1922.
The essential change introduced by
Parshall was a drop in the floor,
which produced supercritical flow
through the throat of the flume. This
seemingly perfected device was
named the Parshall Measuring Flume,
which is not patented and the
discharge tables are not copyrighted.
19

§ Operation:

The constricted throat of the flume produces a head that is related to discharge. The level
of converging section followed by the downward sloping floor in the throat gives the
Parshall flume its ability to withstand relatively high degrees of submergence without
affecting the rate of flow. The converging upstream portion of the flow accelerates the
entering flow, helping to eliminate deposits of sediments, which would otherwise reduce
measurement accuracy. The approaching flow should be relatively free of turbulence,
eddies, and waves if accurate measurements are expected.

§ Advantages:

The principal advantages of the Parshall flume are its capabilities for self-cleaning
(particularly when compared with weirs), its relatively low head loss, and its ability to
function over a wide operating range while requiring only a single head measurement.
These characteristics of the Parshall flume make it particularly suitable for flow
measurement in irrigation canals, certain natural channels, and sewers.

§ Disadvantages:

Parshall flumes have gone out of favor due to their construction complexity and
likelihood to trap sediment compared to newer flume designs. The rectangular and
trapezoidal flumes, for example, are simpler to construct, can be more easily fit into an
existing channel, and can trap less sediment than a Parshall flume. However, the
methodology relating discharge to measured head is more complex.

Although Parshall flumes are in extensive use in many western irrigation projects,
they are no longer generally recommended. Some states specify the use of Parshall
flumes by law for certain situations.
20

§ Design:

Parshall flume sizes are designated by the throat width, W, as shown in Fig. 3-a.
Dimensions are available for flumes with throat widths ranging from 1” to 50ft. The
configuration and standard nomenclature is also given in Fig. 3-a. For a given throat
width W, all other dimensions are rigidly prescribed.

ISCO Open Channel Flow Measurement Handbook – 5th Edition Figure 3-a

Since the discharge tables for Parshall flumes are based on extensive research, faithful
adherence to all dimensions is necessary to achieve accurate flow measurement. The
flumes must be constructed according to the dimensions listed in Table. 3-a for each
flume, because the flumes are not geometrically similar.
 21

§ Discharge Calculation:

The general discharge (head vs. flow rate) equation of free flow through a Parshall
flume takes the form:

Q = KH n
th
ISCO Open Channel Flow Measurement Handbook – 5 Edition Table 3-a

Where Q is the flow rate, H is the head measured at point Ha (see Fig. 3-a), K is a
constant that depends on throat width & units, and n is a constant power that depends on
throat width. Table. 3-b presents the discharge equations for a number of common
Parshall flumes.

§
§ Parshall Flume in Biosolids Project:

ISCO Open Channel Flow Measurement Handbook – 5th Edition Table 3-b
22

Parshall flumes are not patented; therefore we custom designed our own flumes for
the C-11653 Biosolids Project. However, it is absolutely necessary to abide by the
published dimensions.

Our custom-built flumes are about 1 feet high and 3 feet long. Both the sidewalls and
the bottom parts are made of PVC and joined with stainless steel screws. The sidewalls
are made of ¾” PVC sheets, whereas the bottom parts are made of ½” PVC sheets. As to
the most integral dimension, the throat is 2" wide, as illustrated in Fig 3-b.

2”

Figure 3-b
As for the corresponding discharge equations, Table. 3-c sums up the equations to be
used in different discharge units:
 23

Q in cfs & h in ft Q in gpm & h in ft Q in mgd & h in ft Q in l/s & h in m Q in m3/hr & h in m
Units

Formula Q=0.676h1.55 Q=303.4h1.55 Q=0.4369h1.55 Q=120.7h1.55 Q=434.6h1.55

Table 3-c
24

q H-Flumes:

H-flumes, developed by the Natural Resources Conservation Service (former known as

the Soil Conservation Service), are made of simple trapezoidal flat surfaces. These surfaces
are placed to form vertical converging sidewalls. The downstream edges of the trapezoidal
sides slope upward toward the upstream approach, forming a notch that gets progressively
wider with distance from the bottom. These flumes should not be submerged more than
30 percent. This group of flumes, including H-flumes, HS-flumes, and the HL-flumes has
been used mostly on small agricultural watersheds and has not found extensive use in
irrigation flow measurements.

Dimensional proportions of the H-type flumes are shown in Fig. 3-c. (Note that the
dimensions are proportional to the maximum depth, D.)

Depth D
feet meters
0.5 0.152
0.75 0.229
1.0 0.305
1.5 0.457
2.0 0.610
2.5 0.762
3.0 0.914
4.5 1.37

ISCO Open Channel Flow Measurement Handbook – 5th Edition Figure 3-c
 25

q Cutthroat Flumes:

Cutthroat flumes are so named because they

resemble Parshall flumes with the throat "cut out."
They are formed by directly connecting a 6:1
converging section to a similar diverging section.
Thus, they consist of a converging level inlet
section with vertical sidewalls and a diverging level
outlet section also with vertical sidewalls. They do
not have any parallel walls forming a straight throat
and, thus, belong to a class of throat less flumes.
The converging and diverging walls do not
necessarily match those of other flumes in either
converging or diverging slope or length. The
primary objective of their development was
construction simplicity compared to Parshall
flumes.

However, the prescribed head

measuring location, which may be in a
zone of separation, and conditions of
the upstream channel in which it is
placed, along with variable conditions
of the sharp connection of the
convergence and the divergence, have
caused considerable variability in
calibrations. Because of these
complexities in hydraulic behavior,
several authors do not recommend their
use.

A characteristic length, L, and a

throat width, W, dimensionally define
the rectangular cutthroat flume, shown
in Fig. 3-d. As shown, all other flume
dimensions can be derived from the two
dimensions.

ISCO Open Channel Flow Measurement Handbook – 5th Edition Figure 3-d
26

q Palmer-Bowles Flumes:
ISCO Open Channel Flow Measurement Handbook – 5th Edition
Palmer-Bowles flumes are
frequently made as inserts with circular
bottoms that conveniently fit into U-
shaped channels or partially full pipes.
These flumes make a transition from a
circular bottom section to a raised
trapezoidal throat and transition back
to a circular bottom section. These
flumes are of the long-throated type
and can be calibrated by theoretical
analysis.

The Palmer-Bowles flume is

essentially a restriction in the channel
designed to produce a higher velocity
critical flow in the throat. The flume is
most often used in manholes or open
round or rectangular bottom channels
to measure flow rate. It is useful in Figure 3-e
Figure 5
temporary installations to provide flow data for determining flume size and equipment
requirements for permanent installation. Typical shapes of Palmer-Bowels flumes are shown
for installation in round and rectangular conduits are shown in Fig. 3-e.

Some of the flume’s advantages include accuracy of measurement (comparable to

Parshall flumes), low energy loss, and minimal restriction to flow. A principal advantage of
the Palmer-Bowles flume is the comparative ease with which it can be installed in existing
conduits, since it does not require a drop in the conduit invert as would be required with a
Parshall flume. A disadvantage of Palmer-Bowles flumes is that they have a smaller useful
range of flow rates than a Parshall flume, with a range that seldom exceeds twenty to one.
Also, the resolution of the Palmer-Bowles flume is not as good as that of the Parshall flume.
For a given change in flow rate, that Parshall flume produces a greater change head that does
the Palmer-Bowles flume.

Palmer-Bowles flumes are usually purchased prefabricated from one of the many flume
manufacturers. They are normally made of fiberglass, a reinforced plastic, or stainless steel.
Palmer-Bowles flumes are available in a number of installation configurations as shown in
Fig. 3-f.
27

ISCO Open Channel Flow Measurement Handbook – 5th Edition Figure 3-f

Figure 6
28

q Trapezoidal Flumes:

The trapezoidal flume was developed primarily to measure flow in irrigation channels
and has been used for many years by the Agricultural Research Service, U.S. Department of
Agriculture. For agricultural applications, it is superior to Parshall-type flumes for a number
of reasons, particularly for measuring smaller flows. The trapezoidal shape conforms to the
normal shape of ditches, especially those that are lined. This minimizes the amount of
transition section needed as compared to that required from a trapezoidal shape to a
rectangular on and back to the trapezoidal. The trapezoidal one is also desirable since the
sidewalls expand as the depth increases. This means that a trapezoidal flume can convey a
larger range of flow rates since an incremental increase in flow produces a relatively small
increase in depth because of the trapezoidal shape. An isometric view of a typical isometric
flume is presented in Fig 3-g.

ISCO Open Channel Flow Measurement Handbook – 5th Edition Figure 3-g
Figure 7

Another primary characteristic of the trapezoidal flume is that it can operate under a
higher degree of submergence than the Parshall flume without the need of corrections.
Furthermore, the straight through bottom of the flume permits the flume to pass trash quite
readily and reduces the problem of silt build-up upstream of the flume. A plan, profile, and
end/throat section schematic are shown in Fig. 3-h.
29

ISCO Open Channel Flow Measurement Handbook – 5th Edition Figure 3-h
Figure 8

q Quick Selection Guide:

Advantages of Flumes
• Self-cleaning to a certain degree
• Relatively low head loss
• Accuracy less affected by approach velocity than weirs

Disadvantages of Flumes
• High cost
• Difficult to install
30

Type of Flume General Comments

Most widely known and used flume for permanent installations. Available in
o Parshall Flume
throat widths ranging from 1” to 50ft to cover most flows. Fairly difficult
installation requiring a drop in the conduit invert.
Developed to measure agricultural runoff. Principal advantage is ability to
o H-Flume
measure wide range of flows with reasonable accuracy. Construction is
fairly simple and flume is easily installed.
Similar to Parshall flume, except that flat bottom does not require drop in
o Cutthroat Flume
conduit invert. Can function well with high degree of submergence. Flat
bottom passes solids better than Parshall flume.
Flume designed to be easily installed in existing conduit. Good for portable
o Palmer-Bowles Flume
or temporary installations, as not drop in conduit invert required. Widely
used in sanitary field for measuring flow in manholes.
Developed to measure flows in irrigation channels. Principal advantage is
o Trapezoidal Flume
ability to measure wide range of flows and also maintain good accuracy at
low flows.
31

Site Characteristics Related to Locating & Setting Flumes

In reference to Water Resources Research Laboratory, “Water Measurement Manual – 3rd Edition”

Proper location of the flume is important from the standpoint of accuracy and ease of
operation. For convenience, the flume should be located near the diversion point and near the
regulating gates used to control the discharge. Flumes should be readily accessible by vehicle for
both installation and maintenance purposes. All structures for measuring or regulating the rate of
flow should be located in a channel reach where an accurate head can be measured. The survey
of a channel to find a suitable location for a structure should also provide information on a
number of relevant factors that influence the performance of a future structure.

q Approach Conditions:

Flumes should not be installed too close to turbulent flow, surging or unbalanced flow, or
a poorly distributed velocity pattern. Poor flow conditions in the area just upstream from the
measuring device can cause large discharge indication errors. In general, the approaching
flow should be tranquil. Tranquil flow is defined as fully developed flow in long straight
channels with mild slopes, free of curves, projections, and waves.

Studies of approach requirements for closed conduits have led to the acceptance of 10
diameters of straight pipe as sufficient for pipe meters claiming to be accurate to within 0.5
to 1 percent. By the usual hydraulic analogy, open channel flow would require 40 times the
hydraulic radius of straight, unobstructed approach channel. The hydraulic radius is the area
of flow section divided by the wetted perimeter, which becomes d/4 for full pipes; hence, the
suggested 40 times hydraulic radius approach distance. These requirements can probably be
relaxed because open channel measuring flumes claim accuracy to a wider margin of 2 to
5 percent. However, for a rectangular channel that is twice as wide as it is deep, 40 times the
hydraulic radius is numerically equal to 10 top widths.

Approach velocities less than 1 foot per second (ft/s) encourage aquatic pests, insects,
and sediment deposition, so the approach velocity should exceed 1 ft/s if at all practical. To
prevent wave interference of head measurement, the Froude number of the approaching
channel flow should be less than 0.5 for the full range of anticipated discharges and should
32

not be exceeded over a distance of at least 30 times the measurement head before the
structure.

It is recommended that a check be made of the approach velocity condition by current

meter measurements, especially when using baffles. In any case, approach condition should
be verified visually. Visual inspection should be made for obvious boils and backflows and
unstable surface conditions.

q Channel Flow Characteristics and Operational Needs:

For accurate measurements, sufficient head loss must be created to obtain a unique
discharge versus head relationship. This relationship assures that submergence limits have
not been exceeded and modular flow exists. To prevent submergence altogether or to assure
that excessive submergence does not occur, the designer needs to know whether the
downstream water surface elevation relationships are consistent and do not change with
season or whether they are influenced by operation of gates, reservoir operation, or other
laterals. The channel water levels greatly influence the sill height necessary to keep the
downstream water surface below the submergence limit, thus obtaining modular flow for the
needed discharge range.

The amount of downstream flow resistance and, hence, the water surface elevation, is
likely to vary with sediment deposits, debris, canal checking operations, vegetative growth,
and aging. For a new design, careful assessment of friction, including the effects of relative
roughness, is required. A thorough appraisal of needed canal operations is required to
determine the frequency of measuring different discharges, including the normal design flow
and the maximum design flow.

To select or design an appropriate flume for installation in an existing channel, full

advantage should be taken of making field measurements at different discharges to obtain
thorough knowledge of channel performance at the site. After tentatively selecting the flume
location, information should be obtained on the maximum and minimum flows to be
measured, the corresponding flow depths, the maximum velocity, and the dimensions of the
channel at the site. These measurements should include channel widths, side slopes, depths,
and the height of the upstream banks with special attention to their ability to contain the
increased depth caused by the flume installation.
33

q Erosion And Scour:

Ideally, the selected channel reach should have a stable bottom elevation. In some
channel reaches, sedimentation occurs in dry seasons or periods. These sediments may be
eroded again during the wet season. Sedimentation may change approach velocity or may
even bury the structure, and the erosion may undercut the foundation of the structure.

Based on the channel water levels and the required sill height, in combination with the
discharge versus head relationship of the structure, ponding at the upstream structure should
be assessed. Excessive ponding commonly causes sedimentation difficulties because of the
subsequent reduction in the approach flow velocities. To avoid upstream sedimentation, the
sill height plus the measuring head should be about the same as for the approach channel
over as much of the discharge measuring range as practical. This arrangement will help
reduce sedimentation upstream from the structure.

The required drop may exceed the capacity of soil-lined channels to resist scour, and
foundations may scour. Thus, rock armoring may be needed to prevent undermining.
34

Workmanship
In reference to Water Resources Research Laboratory, “Water Measurement Manual – 3rd Edition”

Flumes require accurate workmanship for satisfactory performance. Short flumes will
provide reasonably accurate flow measurements if the standard dimensions are attained during
construction. For accurate flow measurement, the flow surfaces must be correctly set or placed at
the proper elevation, the crest must be properly leveled, and the walls must be properly plumbed.
Although long-throated flumes can be computer recalibrated using as-built dimensions to correct
for moderate form slipping or errors of construction, correcting for throat-section slope in the
direction of flow is not always satisfactory. In any case, adequate care during construction is
preferable. The modified broad-crested weir flume has only one critical flow surface, and it is
level.

Flumes should be set on a solid, watertight foundation to prevent leakage around and
beneath the flume and prevent settlement or heaving. Collars or antiseep walls should be
attached to either or both the upstream and downstream flanges of the flume and should extend
well out into the channel banks and bottom to prevent bypass flow and foundation settlement
caused by erosion. A stable foundation without significant settling or leakage must be secured at
reasonable costs.

The flumes can be built of wood, concrete, galvanized sheet metal, or other materials.
Large flumes are usually constructed on the site, but smaller flumes may be purchased as
complete flumes and placed in one piece. Others are provided in bolt-together pieces, which are
assembled onsite. Some of these flumes are made of lightweight materials, which are then made
rigid and immobile by careful earth backfill or by placing concrete outside of the walls and
beneath the bottom.

When making a number of relatively small concrete flumes of the same size, use of portable and
knockdown reusable forms is economical and practical. These forms require high quality design
and workmanship. Good construction practice should be used in placing footings, setting the
forms, and pouring and tamping wall concrete to provide smooth surface finishes. Accuracy of
the short flumes depends on correct flume dimensions, proper setting, and proper use. As flume
size decreases, the influence of a small dimensional error becomes more prominent, and the
importance of this care increases.
35

Acknowledgements

I am most appreciative of Mr. Tiago Silva’s assistance and undue patience while helping
me on the “Flume in Biosolids Project” section of this paper.
36

References

§ J. A. Robertson & C. T. Crowe, “Engineering Fluid Mechanics – 6th Edition”, 1997.

§ D. M. Grant & B. D. Dawson, “ISCO Open Channel Flow Measurement Handbook – 5th
Edition”, 1998.

§ T. J. McGhee, “Water Supply and Sewerage – 6th Edition”, 1991.

§ Water Resources Research Laboratory, “Water Measurement Manual – 3rd Edition”:

http://www.usbr.gov/wrrl/fmt

§ LMNO Engineering, Research, and Software, Ltd, “Flume Graphic Calculator

Software”: http://www.lmnoeng.com/Flumes/flumes.htm

§ T. F. Silva: “Flume Building”, : http://www.intellitemps.net/silva/

§ O. Neumann: http://www.intellitemps.net/oneumann/prototyping%20frames.htm