Design of an irrigation gate flow meter

Item Type text; Thesis-Reproduction (electronic)

Authors Hiller, William Clark, 1933-

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this materialis made possible by the University Libraries, University of Arizona.Further transmission, reproduction or presentation (such aspublic display or performance) of protected items is prohibitedexcept with permission of the author.

Download date 01/06/2018 08:14:52

Link to Item http://hdl.handle.net/10150/348083

DESIGN OF AN IRRIGATION GATE FLOW METER

by

William Clark Hiller

A Thesis Submitted to the Faculty of the

:DEPARTMENT OF CIVIL ENGINEERING AND ENGINEERING MECHANICS

In Partial Fulfillment of the Requirements For the Degree of

MASTER OF SCIENCE WITH A MAJOR IN CIVIL ENGINEERING

In the Graduate College

THE UNIVERSITY OF ARIZONA

1 9 7 6

STATEMENT BY AUTHOR

This thesis has been submitted in partial fulfillment of re­quirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules, of the Library.

Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his judg­ment the proposed use of the material is in the interests of scholar­ship. In all other instances, however, permission must be obtained from the author.

SIGNED: LO CSLo^Sb

APPROVAL BY THESIS DIRECTOR

This thesis has been approved on the date shown below:

DateTHOMAS CARMODY Professor of Civil Engineering

ACKNOWLEDGMENTS

These experiments were conducted using the facilities of the

Civil Engineering Department, University of Arizona. Professor Q.M .

Mees, Head, Department of Civil Engineering and Engineering Mechanics,

was very cooperative in providing a pleasant and unhurried research

atmosphere. Dr. Thomas Carmody and Dr. Emmett Laursen, Professors,

Department of Civil Engineering, were of invaluable assistance in

providing advice, recommendation and support. Grateful acknowledgment

is given to Darell Zimbleman, Engineer, Salt River Valley Water User's

Association, Phoenix, Arizona, for cooperation in providing the irriga­

tion gates and gate transitions without which the project would not

have been possible. Very special thanks to Mr. Lou Gemson and Mr. Bill

Lichtenwalter, Department of Civil Engineering, for their help in

building and assembling the test apparatus. I am most thankful for my

wife's help and encouragement.

TABLE OF CONTENTS

Page

LIST OF ILLUSTRATIONS . . . .................. v

ABSTRACT ............. ........................................ . vi

1. INTRODUCTION .................. 1

Probe Development............. 4

2. EXPERIMENT . . . . . . .................... 6

R e s u l t s ........... 12Discussion of Results ................................ 13

3. CONCLUSIONS................ 22

APPENDIX A - PRELIMINARY M E T E R ......... 23

APPENDIX B - METER P O S I T I O N I N G ....... 25

APPENDIX C - DATA .......................... 27

LIST OF SYMBOLS .......................... 31

LIST OF REFERENCES . . . . . . . . . . . . . . . 32

iv

LIST OF ILLUSTRATIONS

Figure Page

1. Schematic Diagram of Gate Parameters ................ 3

2. Flow M e t e r ................ 5

3. Plan and Side View of Experimental Apparatus . . . . . . . . 7

4. General View of Experimental Apparatus . . . . . . . . . . . 8

5. View of Distribution Box . . . . . . . . . . . . . . . . . . 8

6 . View of Straight T r a n s i t i o n .................... 9

7. View of Ninety Degree Transition . . . . . . . 9

8 . Front and Section View of Distribution Gate ......... 11

9. ^ v s • AH 5 Gate 1 . . . . . . . . . . . . . . . . . . . . . . 14

10. 5. vs. AH, Gate 2 .................. 15K11. 7 7 vs. b, Gate 1 ........... 16K°12. 7 7 vs. b. Gate 2 .- . . 17K°13. £ vs. H, Gate 1 . . .............. ' ................... 18K°14. 7 7 vs. H, Gate 2 . . . ...................... 19o

15. Q Measured vs. Q Calculated ............ 20

16. Preliminary Flow Meter . . . . . . . . . . . . . . . . . . . 24

17. K vs. 0 and AH vs. 0, Preliminary M e t e r ................ 26

v

ABSTRACT

A study was performed to determine the feasibility of measuring

the discharge rates from a standard Salt River Project twelve-inch

rectangular irrigation delivery gate by insertion of a simple measuring

device in the vicinity of the gate lip. Tests indicated that a three-

fourths inch flow meter with stagnation-separation-zone taps placed

approximately one inch below and one inch forward of the gate lip

produced the desired results. Although the discharge coefficient varied

for different gate openings, it was approximately constant for drowned

and free flow configurations at given openings. Results are given in

the form of head measurements and discharge coefficients for the con­

figurations studied. Accuracy of the flow meter was within two percent

of the total flow.

CHAPTER 1

INTRODUCTION

An important item in the operation of a large scale irrigation

system is the measurement of the water delivered to the user. The

device and the technique for this measurement needs to be simple, accu­

rate, reliable, and easy to manipulate.

The Salt River Project, operating in Phoenix, Arizona, and its

environs since the first decade of this century is currently using

approximately 6 , 0 0 0 sluice gates to control flow from a multitude of

rectangular head boxes installed on some 250,000 acres of residential

and agricultural land. Volume rates are estimated through the use of

tabulated charts and direct measurement of gate opening and upstream

and downstream heads. Because of inherent uncertainties in the measure­

ments, there is perhaps an unconscious tendency for the Project to over­

supply its customers to preclude the possibility of not fulfilling its

contractual obligations. Such a practice may be reasonable when excess

is available, but it cannot be justified when demand approaches the

limit of supply.

The need for better control, both for the Salt River Project in

particular and for others in general, has led to the present undertaking,

a study of the feasibility of developing a portable measuring device by

which an operator can readily ascertain the flow under a sluice gate,

hopefully, to an accuracy of one or two percent.

Two-dimensional free flow under a sharp-edged sluice gate in

a plane-bottomed channel is well behaved and easily predicted, whereas

flow under a typical Salt River Project gate is decidely three-

dimensional and not fully understood.

Present practice in the Project is to measure the water surfaces

at I and II, as shown in Figure 1, and the opening b, and then to relate

these measurements to the discharge Q by past calibration of similar

gates. For free flow this technique is reasonable, although the measure­

ment at II is made at a location where the water surface is sloping and

where it may be uneven as it leaves the edge of the sluice gate. For

drowned flow, the measurement at II is extremely inaccurate, because the

water surface is fluctuating wildly.

No changes in the head box, sluice gate, or outlet pipe are

contemplated as a result of this study. The improvements that could be

made in the accuracy of the discharge measurement would be small and the

total cost of revisions would be quite large. Therefore, the flow

through the gate will be unchanged. The change comes in the measure­

ments made, or, in a sense, the places the measurements are made.

Figure 1 indicates a typical streamline pattern in a vertical

plane through the center of a gate for free outflow. Several significant

parameters used in this study are identified thereon. For drowned out­

flow with the same available head, the form of the streamline pattern

approaching the gate is unchanged: only the quantity of flow between the

lines is diminished as the extent of drowning increases. Thus, it is

reasoned that a substantial/pressure differential, measured in the

approach flow in a region of close streamline spacing, will indicate a

g a t e

Figure 1. Schematic Diagram of Gate Parameters

z

4characteristic velocity, and this in turn multiplied by a characteristic

area will indicate the desired flow rate. In equation form, then,

Q = KAv'SH • (1)

where K is independent of the free or drowned outflow condition, but is

a function of opening, head in the box, and a location of the gate in

the box.

Probe Development

A desirable probe is one which is easily transported, simply

connected and disconnected, small enough to cause minimal disturbance

in the stream, and of such a shape that it causes significant head

differential in the flow immediately surrounding it. A disk placed

normal to the flow produces this differential. The selected disk is

held in place by means of a bracket which is easily clamped to the lip

of the gate. From two piezometer taps, one on either side of the disk,

tubes can be run to a location convenient for measuring the heads.

A preliminary probe, described in Appendices A and B was studied

to select a favorable location for the disk. Test data indicated that

a reasonably optimal geometry was created by placing the center of the

upstream face of the disk at a point one inch below the lip of the gate

and one inch into the approaching flow, with the face tipped up such

that it lay in a plane five degrees from the vertical. Figure 2 shows

the probe in its currently developed form.

A T T A C H M E N TBLOCN

s t a g m k t i o u

0 . 12.511

F R L O N T V I E W

GATE L IP -

V8" TUBES TO UANOUETEES

S E P A AT \ OJO - 'E.OVOE T A P0.042" ( * 58 DIZtLL)

V A U E

S ID E V I E Wf u l l S c a l e

LIP

TAP

Figure 2. Flow Meter

CHAPTER 2

EXPERIMENT

The apparatus used in the study consisted of two Salt River

Project twelve-inch rectangular steel irrigation delivery gates, one

ninety-degree fiberglas transition, one straight fiberglas transition,

a wooden distribution-box, a flume, a valve, flow restrictors, and

necessary pipes and measuring devices (Figures 3 through 7).

Water was pumped from the Civil Engineering patio storage sump

to the overhead constant head tank. From the tank the water flowed down

a twelve-inch delivery pipe through a throttling valve into the distribu­

tion box. From the distribution box the water discharged by one of two

pathways, either through the discharge gate on the end or through the

gate at the side, then through a Salt River Project transition and, if

drowning was desired, through a restrictor. Water discharged from the

side-gate ninety-degree transition dumped directly into the patio

channel, where it then flowed over a V-notch weir and back into the

underground storage. Water from the end gate discharged into a wooden

flume, where it was stilled by a system of screens. Beyond the screens,

the water flowed over a straight weir and from there it was deflected

forty-five degrees into the same patio channel which served the side

gate.

The inlet of the delivery pipe was fourteen feet above the

bottom of the distribution box. The patio channel bottom was three feet

P A T tO C H A U U E LS.K.P. 30° Tr2AVJS\Tt0O A

FLOW (ZEST 121CTO(ZS

V- k)OTCW LOBtTZ-S.(2.P. G A T E S

STTZAtGHT S T lL L tU G 6C12EEUSJ E T DEFLECTOR.

FROM COUSTAOT H EAD TAUlZ -

SEE SECTtoiO A-A FtG. 4 FOIZ FLOW UETE1Z. PLACEUEOT

DEFLECTOR. DlSTtZtBUTtOlO BOX

P A TIO LEU EL

T H fZoTTLlDG V/AL\jECHAkJUEL BOTTOM

SCALE *4 " *• I'

Figure 3. Plan and Side View of Experimental Apparatus

Figure 4. General View of Experimental Apparatus

Figure 5. View of Distribution Box

Figure 6. View of Straight Transition

Figure 7. View of Ninety Degree Transition

10below the box with a channel run of ninety feet to the v-notch weir.

Water-air manometers were connected to both the gage and the distribu­

tion box to indicate: (1 ) head in distribution box, (2 ) stagnation

pressure from the flow meter forward tap, and (3) static pressure from

the flow meter rear. tap.

Problems were encountered in determining both the horizontal

placement of the flow meter from the gate lip and the stiffener angle

on the gate. It became apparent that the stiffener angle placed above

the Salt River Project gate lip was disturbing the flow and therefore

making meter placement critical. See Figure 8 for the gate stiffener

placement.

A preliminary meter was made and tests were run at constant

gate openings to determine the proper meter angle. A five-degree off­

set from the vertical was found to be an optimum angle. See Appendix A

for details on the preliminary meter. The information thus gained from

the preliminary meter indicated need for a second meter. The second

meter was made and positioned relative to the gate lip as shown in

Figure 4. With the meter position and geometry determined, tests were

run for both the end gate and the side gate. The test procedure was to

hold the gate openings constant at two inches, four inches, and six

inches, while varying the head from just above the lip to about two and

three-fourths feet above the lip. Each test was first measured with

free flow, then with flow drowned. The various parameters of distribu­

tion box head, meter AH, and flow were measured with the water-air

manometer point gages and tabulated. Head for the distribution box was

read directly from the water-air manometer. The difference between the

G k T H

11

-^-1

FLOW MLEk

FR.OUT VlBUO

ST tFFEDB.2.

F L OV3FLOW m e t e e :

SECTIOVJ k-ku o t t o S c a l e

Figure 8. Front and Section View of Distribution Gate

12stagnation-separation-zone pressures gave the AH for the meter. Total

flow was measured with the patio-channel V-notch weir, which is a

British standard ninety-degree notch. The discharge equation"*" used for

the weir was

Q = 2.48h2 '48 (2)

where h is the water level above the weir notch measured upstream.

A major source of error in flow measurement by the water-air

manometer was attributed to the fluctuation of the water surface in the

manometer tubes. The fluctuation was greatest in the high ranges of

velocity even though an entrance jet dissipator was placed at the water

delivery end of the distribution box. Another source of measurement

error was generated by the turbulence in the distribution box.

Results

In the present experimental work, the gate width w is constant,

so that the area A under the gate varies as the gate opening b only.

Thus equation (1) is rewritten in the form

5- = KvS T . (3). : - ' " <o

where K, which absorbs the width w , is a function of the gate opening

and upstream head.

1. F. M. Henderson, Open Channel Flow, New York, 1966, p. 178.

13Measured values of Q/b vs. AH are presented in Figures 9 and 10.

In view of the fact that, with the parameter K having a constant value.

Equation (3) takes the form of a parabola, an arbitrary "base" parabola,

with Kg given the value of 0 . 6 (which coincides with the mid-range of

the data) is plotted thereon. The proper value of K required to predict

a particular flow rate is then to be taken from Figures 11 or 12 where

K/K is plotted as a function of either an end or side gate, gate opening ob and upstream head H. The two families of curves, one family for Gate 1

and one family for Gate 2, are plotted to be most effective in reading

the required K/K^ value. They are cross plotted from the equivalent

families of curves in Figures 13 and 14 which are better for looking at

the raw data.

Figure 15 shows Q measured vs. Q calculated, with limits +_ 2%

from the ideal relationship also indicated. One sees at a glance that

almost all the data fall well within this range.

Discussion of Results

The response pattern of the probe is indicated quite clearly in

Figures 11 and 12, where it is shown that K/Kq decreases with both in­

crease in gate opening and increase in upstream water-surface elevation.

The influence of gate opening derives from the manner in which the probe

is supported in the flow. Its absolute location is fixed with respect

to the lip of the gate, so that as the gate opening increases, its

position relative to the entire flow pattern shifts toward the lip,

where it intercepts fluid having velocity magnitudes more nearly equal

to the maximum value found at the free streamline. Since the deflection

Q~b

C PS I w

BASE. PkRLABQLAQ o = o G - f A r r

F E E S FLOW DtZoUaMED FLouO

C . \ 0.2 0.3 0.4- 0.5 0.6 0.̂ 7 0.9 0.9 1.0 LlA H FT.

1.2. 1.3 1.4 1.5

Figure 9. g- vs. AH, Gate 1

BASE PATZ.ABOLA

Qo ~ ° ' G' f A H 1 ■’

-*

0.55

0.35

0.3o

0.15

Fe.EE. FLOW

D rzo o o io fD F l o w S o l ido .o5

3

A H FT.Figure 10. ^ vs. AH, Gate 2

H - 0 , 5 FT.

Z.5 FT.

4 5 *732

b ID,V

Figure 11. vs. b, Gate 1

H = 0.5 FT.

4 73 5Zb lO,

vFigure 12. -r- vs. b. Gate 2

3

1.0

0.9

F R L E E PLOU> o p e o

D E O U JIO B D FLOU) S O U D

0.5 l.o 1.5H FT.

vFigure 13. ̂ vs. H, Gate 1

Z.o Z .5

FIZeE PLOUi OPEO p e o u i u E D F L o w Solid

1,4

1.3 b = 2 to

.2

1,0 4 to

0.9A — —

-h

0 . 5 1.0 1-5H FT.

Figure 14 . ^ vs. H, Gate 2

Z.o Z.5

io

4.o

2o

0.0

0.6

Q c ) Q CKUCULA.TE.DQ m ) Q UENeUT2.BD

4.00.6 0.8 VO 2.0

Q c C F S

Figure 15. Q Measured vs. Q Calculated

21of particles of greater speed provides a AH of greater magnitude, it is

necessary that K decreases as the gate opening increases.

The effect of upstream water-surface elevation lies in its

governing the shape of the approaching streamline pattern. At lower

values, streamline orientation in the vicinity of the probe varies con­

siderable with water-surface elevation, and so the value of K varies .

with it, since the angle of attack combines with the speed of the water

(Reynolds effect) to influence the form of the eddy pocket and the

measured head differential. At higher values, however, both the stream­

line orientation and the effects of velocity increases are less variant,

thus the magnitude of K itself approaches a fixed value.

CHAPTER 4

CONCLUSIONS

1. Discharges for free and submerged flow from twelve-inch

Salt River Project rectangular steel gates may be measured using the

proposed meter. However, the allowable meter positions are restricted,

due to the support angle attached to the gates.

2. It is possible to measure discharge for both free and sub­

merged flow without differentiation between the two types of flow.

3. Separate calibration curves must be used for the side and

end gates, because the flow patterns are slightly, different as would be

expected .

4. The major source of error in measuring the levels of the

water-air manometer tubes was due to fluctuation of the fluid in the

manometer tubes. Large-sized tap holes in the meter caused large

fluctuation and short equalibrium time. Small-sized tap holes caused

small fluctuation and excessively long equalibrium time. A compromise

such as the one made in this study was therefore necessary.

5. The separation-zone inlet tap size needs to be restricted

for proper flow dampening to minimize inaccuracy in water-air manometer

reading.

6. Accuracy of the meter in measuring the flow ranges was good,

,• as the error seldom exceeded two percent of the actual flow.

22

APPENDIX A

PRELIMINARY METER

Figure 16 shows the configuration of the first meter constructed

for the experiment. Initially the static port was one-eighth inch in

diameter; however, it was found that better results were obtained when

the port was sized to a number sixty-eight drill (0.031 inches). This

reduction in port size dampened.the static water-air manometer fluctu­

ation, allowing more reliable pressure readings.

Using L, Figure 16, as the characteristic length, a Reynolds- 4 5number. Re, for the meter was calculated. The range was 3x10 to 1x10 .

Though the velocity used for the calculation was an order of magnitude

lower than the minimum velocity expected during the experiment, the Re

was well into the turbulent range for flow past a bluff object‘d. As ’ >"

this was a reasonable approximation for the rectangular shape used for

the meter, it was decided that the meter would therefore operate in the

•turbulent range for all desired flows. For the final meter, a hori­

zontal vane was placed behind the disk to stabilize the eddy pocket,

thus eliminating possible degradation of the static-port readings.

2. H. Rouse, Engineering Hydraulics, New York, 1950, p. 122,

23

24

B T A G U X T I O U

O. 1 2 5 "

f i z o or

7SEPATZ-ATtoO - 20K)E —

TAP 0 .0 3 1 "

FULL SCALEB O T T O M

T A P

Figure 16. Preliminary Flow Meter

APPENDIX B

METER POSITIONING

For determining the angle of the meter with respect to the flow,

a midrange gate opening was selected and the opening held constant.

Runs were performed varying the angle of the meter. As Figure 17 indi­

cates , both K and AH were slowly varying functions of angle between zero

and ten degrees from the horizontal. From this, a five degree angle

from vertical was selected as the optimum meter angle. The distance

forward and the distance down from the gate lip were subsequently set

at one inch each, as this placed the meter well within the flow field.

These also were acceptable dimensions for attaching the meter.

25

26

6 .00

1.-15

1.50

1.25KFT. 700

6.15

6.50

6.25

6 .00t 30o + 1 o + Zoo

©

150

1.25

1 . 0 0

AHFT. 0.15

0.50

0.25

0 . 0 0

+ 3o-20 O- i o

Figure 17. K vs. 0 and AH vs . 0, Preliminary Meter

APPENDIX C

Table C-l.

DATA

Data for Gate 1

H, Ft. AH, Ft. b=2 In. Q, CFS

Run 1 Free 0.646 0.161 0.673

Run 1 Drowned 1.137 0.158 0.670

Run 2 Free 1.176 0.354 0.953

Run 2 Drowned 1.706 0.339 0.939

Run 3 Free 1.811 0.553 1.211

Run 3 Drowned 2.219 0.536 1.195

Run 4 Free 2.524 0.769 1.435

Run 4 Drowned 2.807 0.770

b=4 In.

1.422

Run 1 Free 0.741 0.227 1.239

Run 1 Drowned 1.198 0.252 1.223

Run 2 Free 0.962 0.345 1.471

Run 2 Drowned 0.497 0.377 1.462

Run 3 Free 1.199 .0.486 1.696

Run 3 Drowned 1.757 0.499 1.691

Run 4 Free 1.505 0.662 1.952

Run 4 Drowned 2.116 0.635 1.947

Run 5 Free 1,824 0.883 2.190

Run 5 Drowned 2.342

27

0.820 2.178

28TABLE C-l— Coii'tiTiii6d.

H, Ft. AH, Ft. b=4 In. Q, CFS

Run 6 Free 2.185 1.056 2.431

Run 6 Drowned 2.795 0.969 2.431

b=6 In.

Run 1 Free 0.723 0.180 1.507

Run 1 Drowned - 1.267 0.210 1.493

Run 2 Free 0.983 0.316 2.000

Run 2 Drowned 1.446 0.357 1.963

Run 3 Free 1.329 0.557 2.499

Run 3 Drowned 1.717 0.563 2.468

Run 4 Free 1.744 0.848 3.002

Run 4 Drowned 2.316 0.836 2.988

29Table C-2. Data for Gate 2

", Ft. Q, CFS

Run 1 Free 0.752 0.180 0.659,

Run 1 Drowned 1.156 0.172 0.659

Run 2 Free 1.300 0.352 0.932

Run 2 Drowned 1.653 0.355 0.925

Run 3 Free 2.038 0.591 1.183

Run 3 Drowned 2.442 0.594 1.167

Run 4 Free 2.609 0.762 1.331

Run 4 Drowned - - -

b=4 In.

Run 1 Free 0.867 0.268 1.219

Run 1 Drowned , 1.490 0.266 1.211

Run 2 Free 1.055 0.391 1.466

Run 2 Drowned 1.951 0.396 1.457

Run 3 Free 1.300 0.514 1.706

Run 3 Drowned 2.285 0.513 1.662

Run 4 Free 1.675 0.668 1.936

Run 4 Drowned 2.454 0.694 1.889

Run 5 Free 2.076 0.873 2.138

Run 5 Drowned 2.719 0.886 2.127

Run 6 Free 2.522 1.082 2.407

Run 6 Drowned - - -

30

TABLE C—2— Continued

H, Ft. AH, Ft. b-6 In. Q, CFS

Run 1 Free 0.833 0.225 1.484

Run 1 Drowned 1.262 0.210 1.475

Run 2 Free 1.077 0.377 1.990

Run 2 Drowned 1.630 0.369 1.958

Run 3 Free 1.467 0.589 2.505

Run 3 Drowned 1.929 0.601 2.450

Run 4 Free 1.995 0.909 3.015

Run 4 Drowned 2.647 0.863 2.920

LIST OF SYMBOLS

Gate opening area

Gate opening, vertical dimension

Gravitational constant

Distribution box head

V-notch weir head

Meter tap pressure differential

Discharge equation coefficient

Discharge equation coefficient, reference value

Total flow

Total flow, reference value

Average upstream velocity

Average downstream velocity

Gate width

Upstream depth

Downstream depth

Subscript 1 refers to approach flow

Subscript 2 refers to downstream flow

Meter angle from horizontal

LIST OF REFERENCES

Henderson, F. M., Open Channel Flow, The MacMillan Company, New York, 1966.

Rouse, H., Engineering Hydraulics, edited by H. Rouse, John Wiley and Sons, Inc., New York, N. Y., 1950.

32