Sprinkle & Trickle Irrigation Lecture Notes PDF

Sprinkle & Trickle Irrigation
Lecture Notes
BIE 5110/6110
Fall Semester 2004
Biological and Irrigation Engineering Department

Utah State University, Logan, Utah
Merkley & Allen Page 2 Sprinkle & Trickle Irrigation Lectures
Preface
These lecture notes were prepared by Gary P. Merkley of the Biological and
Irrigation Engineering (BIE) Department at USU, and Richard G. Allen of the
University of Idaho, for use in the BIE 5110/6110 courses. The notes are intended
to supplement and build upon the material contained in the textbook Sprinkle and
Trickle Irrigation by Jack Keller and Ron D. Bliesner (Chapman-Hall Publishers
1990). Due to the close relationship between the lecture notes and the textbook,
some equations and other material presented herein is taken directly from Keller
and Bliesner (1990) – in these instances the material is the intellectual property of
the textbook authors and should be attributed to them. In all other cases, the
material contained in these lecture notes is the intellectual property right of G.P.
Merkley and R.G. Allen.
Copyright Notice
This material has been duplicated by special permission of the copyright

holders. It is not to be duplicated or used for purposes other than learning resource
support for Utah State University. Any violation of this agreement is punishable
under existing copyright laws and may include severe fines and or imprisonment.
© Copyright 1990 to 2004
These lecture notes are formatting for printing on both sides of the page, with
odd-numbered pages on the front. Each lecture begins on an odd-numbered
page, so some even-numbered pages are blank.
Sprinkle & Trickle Irrigation Lectures Page 3 Merkley & Allen

Contents
No. Lecture Title Page
Sprinkle Irrigation
1 Course Introduction.................................................................................... 9
2 Types of Sprinkler Systems; Soil-Water-Plant Relationships;
Planning Factors ...................................................................................... 11
3 Sprinkler Characteristics; Application Rates ............................................ 25
4 Set Sprinkler Uniformity & Efficiency........................................................ 39
5 Layout of Laterals & Mainline for Set Sprinklers; Pipe Hydraulics............ 51
6 Economic Pipe Selection Method............................................................. 59
7 Set Sprinkler Lateral Design .................................................................... 71
8 Set Sprinkler Lateral Design & Analysis................................................... 85
9 Mainline Pipe Design ............................................................................. 101
10 Minor Losses, Pressure Requirements & Pumps................................... 111
11 Pumps & System Curves; Affinity Laws & Cavitation ............................. 123
12 Center Pivot Design & Operation ........................................................... 145
13 Center Pivot Nozzling & Hydraulic Analysis ........................................... 155
14 Center Pivot Uniformity Evaluation; Linear Move Systems .................... 169
15 Maximizing Linear Move Field Length; Design Example ........................ 179
Trickle Irrigation
16 Components & Layout; Pressure Control & Flow Regulation ................. 189
17 Filtration for Trickle Irrigation Systems ................................................... 197
18 Trickle Irrigation Planning Factors; Salinity in Trickle Irrigation .............. 207
19 Water Requirements; Coefficient of Variation & System Capacity ......... 215
20 Emitter Selection & Design; Design Approach & Example ..................... 225
21 Pipe Specifications & Lateral Design; Manifold Location ....................... 231
22 Numerical Solution for Manifold Location; Derivations ........................... 241
23 Manifold Hydraulic Design ..................................................................... 253
24 Hydraulic Design of Mainline & Supply Line........................................... 275
Note: Equations are numbered consecutively in these lecture

notes as (xxx). Equations with the (xxx.xx) format refer to those
found in the textbook by Keller & Bliesner.

Units, Constants and Conversions
28.35 g/oz
15.85 gpm/lps (= 60/3.785)
7.481 gallons/ft3
448.86 gpm/cfs (= 7.481*60)
3.7854 litre/gallon
6.89 kPa/psi
1 cb = 1 kPa
10 mb/kPa, or 100 kPa/bar
2.308 ft/psi, or 9.81 kPa/m (head of water)
14.7 psi = 101.3 kPa = 10.34 m (head of water) = 1,013 mbar = 1 atm
62.4 lbs/ft3, or 1000 kg/m3 (max density of pure water at 4°C)
0.1333 kPa/mmHg
1 ppm ≈ 1 mg/liter (usually)

1 mmho/cm = 1 dS/m = 550 to 800 mg/liter
0.7457 kW/HP
1 langley = 1 cal/cm2
0.0419 MJ/m2 per cal/cm2
0.3048 m/ft
1.609 km/mile
2.471 acre/ha
43,560 ft2/acre
1,233 m3/acre-ft
57.2958 degrees/radian
π ≈ 3.14159265358979323846
e ≈ 2.71828182845904523536
ºC = (ºF – 32)/1.8
ºF = 1.8(ºC) + 32
Ratio of weight to mass at sea level and 45° latitude: g = 9.80665 m/s2
PVC = Polyvinyl chloride

PE = Polyethylene
ABS = Acrylonitrile-Butadiene-Styrene

Equation Chapter 1 Section 1Lecture 1
Course Introduction
I. Course Overview
• Design of sprinkle and trickle systems – perhaps the most comprehensive

course on the subject anywhere
• Previously, this was two separate courses at USU
• Everyone must be registered at least for audit
• Prerequisites: BIE 5010/6010; computer programming; hydraulics
• There will be two laboratory/field exercises
• Review of lecture schedules for sprinkle and trickle
II. Textbook and Other Materials
• Textbook by Keller and Bliesner

• Two textbooks are on reserve in the Merrill Library
• Lecture notes by Merkley and Allen are required
• We will also use other reference materials during the semester
III. Homework and Design Project
• Work must be organized and neat

• Working in groups is all right, but turn in your own work
• Computer programming and spreadsheet exercises
• Submitting work late (10% per day, starting after class)
• Sprinkle or trickle system design project
IV. Tests, Quizzes, and Grading Policy
• Maybe some quizzes (these will not be announced)

• Two mid-term exams
• Final exam is comprehensive
V. Units
• It is often necessary to convert units in design calculations

• Make it a habit to perform dimensional analysis when using equations; only
in some of the empirical equations will the units not work out correctly
VI. Irrigation Systems
• On-farm level (field)

• Project level (storage, conveyance, tertiary)

VII. General Types of On-Farm Irrigation Systems
Type U.S. Area World Area

Surface 65% 95%
Sprinkler 30% 3%
Micro Irrigation 3% 1%
Sub-Irrigation 2% 1%
These are approximate percent areas
VIII. Sprinkler Systems
Important Advantages
1. effective use of small continuous streams of water

2. greater application uniformity on non-homogeneous soils (provided there is
no appreciable surface runoff)
3. ability to adequately irrigate steep or undulating topographies w/o erosion
4. good for light and frequent irrigation where surface irrigation may be used
later in the growing season
5. labor is only needed for a short time each day (unless there are many fields)
6. labor can be relatively unskilled (pipe moving)
7. automation is readily available for many sprinkler systems
8. can be effective for weather (micro-climate) modification
Important Disadvantages
1. initial cost can be high (compared to surface irrigation systems) at $500 to

$3500 per ha
2. operating cost (energy) can be high compared to non-pressurized systems,
unless sufficient head is available from a gravity-fed supply
3. water quality can be a problem with overhead sprinklers if water is saline,
and in terms of clogging and nozzle wear. Also, some types of water are
corrosive to sprinkler pipes and other hardware
4. some fruit crops cannot tolerate wet conditions during maturation (unless
fungicides, etc., are used)
5. fluctuating flow rates at the water source can be very problematic
6. irregular field shapes can be difficult to accommodate
7. very windy and very dry conditions can cause high losses
8. low intake rate soils (< 3 mm/hr) cannot be irrigated by sprinkler w/o runoff
IX. Slides of Sprinkler Systems
[these will be shown in class]

Lecture 2
Types of Sprinkler Systems
I. Sprinkler System Categories
• Two broad categories: set and continuous-move

• Set systems can be further divided into: fixed and periodic-move
II. Set Systems:
Hand-Move
• very common type of sprinkler system

• costs about $30 - $90 per acre, or $75 - $220 per ha
• requires relatively large amount of labor
• laterals are usually aluminum: strong enough, yet light to carry
• usually each lateral has one sprinkler (on riser), at middle or end of pipe
• to move, pull end plug and begin draining of line, then pull apart
• lateral pipe is typically 3 or 4 inches in diameter
• usually for periodic move, but can be set up for a fixed system
• sprinklers are typically spaced at 40 ft along the pipe
• laterals are typically moved at 50- or 60-ft intervals along mainline
End-Tow
• similar to hand-move, but pipes are more rigidly connected

• tractor drags the lateral from position to position, straddling a mainline
• has automatically draining values (open when pressure drops)
• pipe is protected from wear during dragging by mounting it on skid plates or small
wheels
• least expensive of the mechanically-moved systems
• needs a 250-ft (75-m) “turning strip” about the mainline
Side-Roll
• very common in western U.S.

• costs about $150 - $300 per acre, or $360 - $750 per ha
• wheels are usually 65 or 76 inches in diameter

• lateral is the axle for the wheels; lateral pipe may have thicker walls adjacent to a
central “mover” to help prevent collapse of the pipe during moving
• uses “movers” or motorized units to roll the lateral; these may be mounted in middle
and or at ends, or may be portable unit that attaches to end of line
• self-draining when pressure drops
• must drain lines before moving, else pipe will break
• windy conditions can cause difficulties when moving the lateral, and can blow empty
lateral around the field if not anchored down
• can have trail tubes (drag lines) with one or two sprinklers each
• need to “dead-head” back to the starting point
• mainline is often portable
• has swivels at sprinkler and trail tube locations to keep sprinklers upright
• low growing crops only (lateral is about 3 ft above ground)
• can be automated, but this is not the typical case
Side-Move
• almost the same as side-roll, but lateral pipe is not axle: it is mounted on A frames
with two wheels each
• clearance is higher than for side-roll
• not as common as side-roll sprinklers
Gun
• 5/8-inch (16 mm) or larger nozzles

• rotate by rocker arm mechanism
• discharge is 100 to 600 gpm at 65 to 100 psi
• large water drops; commonly used on pastures, but also on other crops
Boom
• have big gun sprinklers mounted on rotating arms, on a trailer with wheels
• arms rotate due to jet action from nozzles
• arms supported by cables
Other Set Sprinklers
• Perforated Pipe
• Hose-Fed Sprinklers
• Orchard Sprinklers
Fixed (Solid-Set) Systems
• enough laterals to cover entire field at same time

• will not necessarily irrigate entire field at the same time, but if you do, a larger
system capacity is needed
• only fixed systems can be used for: frost protection, crop cooling, blossom delay
• easier to automate that periodic-move systems

• laterals and mainline can be portable and above ground (aluminum), or permanent
and buried (PVC or steel, or PE)
III. Continuous-Move Systems
Traveler
• could be big gun or boom on platform with wheels

• usually with a big gun (perhaps 500 gpm & 90 psi) sprinkler
• long flexible hose with high head loss
• may reel up the hose or be pulled by a cable
• some travelers pump from open ditch, like linear moves
• sprinkler often set to part circle so as not to wet the travel path
Center Pivot
• cost is typically $35,000 ($270/ac or $670/ha), plus $15,000 for corner system
• easily automated
• typical maximum (fastest) rotation is about 20 hrs
• don’t rotate in 24-hr increment because wind & evaporation effects will concentrate
• returns to starting point after each irrigation
• typical lateral length is 1320 ft (400 m), or ¼ mile (quarter “section” area)
• laterals are about 10 ft above the ground
• typically 120 ft per tower (range: 90 to 250 ft) with one horsepower electric motors
(geared down)
• IPS 6” lateral pipe is common (about 6-5/8 inches O.D.); generally 6 to 8 inches, but
can be up to 10 inches for 2640-ft laterals
• typical flow rates are 45 - 65 lps (700 to 1000 gpm)
• typical pressures are 140 - 500 kPa (20 to 70 psi)
• older center pivots can have water driven towers (spills water at towers)
• end tower sets rotation speed; micro switches & cables keep other towers aligned
• corner systems are expensive; can operate using buried cable; corner systems don’t
irrigate the whole corner
• w/o corner system, π/4 = 79% of the square area is irrigated
• for 1320 ft (not considering end gun), area irrigated is 125.66 acres
• with corner system, hydraulics can be complicated due to end booster pump
• center pivots are ideal for allowing for effective precipitation
• ignore soil water holding capacity (WHC)
• requires almost no labor; but must be maintained, or it will break down
• can operate on very undulating topography
• known to run over farmers’ pickups (when they leave it out there)!
• many variations: overhead & underneath sprinklers; constant sprinkler spacing;
varied sprinkler spacing; hoses in circular furrows, etc.
• sprinkler nearest the pivot point may discharge only a fine spray; constant radial
velocity but variable tangential speeds (fastest at periphery)
• some center pivots can be moved from field to field
Linear Move
• costs about $40,000 for 400 m of lateral

• field must be rectangular in shape
• typically gives high application uniformity
• usually guided by cable and trip switches (could be done by laser)
• usually fed by open ditch with moving pump, requiring very small (or zero slope) in
that direction
• can also be fed by dragging a flexible hose, or by automated arms that move
sequentially along risers in a mainline
• need to “dead-head” back to other side of field, unless half the requirement is
applied at each pass
• doesn’t have problem of variable tangential speeds as with center pivots
IV. LEPA Systems
• Low Energy Precision Application (LEPA) is a concept developed in the mid

to late 1970s in the state of Texas to conserve water and energy in
pressurized irrigation systems
• The principal objective of the technology was to make effective use of all
available water resources, including the use of rainfall and minimization of
evaporation losses, and by applying irrigation water near the soil surface
• Such applications of irrigation water led to sprinkler system designs
emphasizing lower nozzle positions and lower operating pressures, thereby
helping prevent drift and evaporative losses and decreasing pumping costs
• For example, many center pivot systems with above-lateral sprinklers have
been refitted to position sprinkler heads under the lateral, often with lower
pressure nozzle designs
• The commercialization of the LEPA technology has led to many modifications
and extensions to the original concept, and is a term often heard in
discussions about agricultural sprinkler systems
• The LEPA concept can be applied in general to all types of sprinkler
systems, and to many other types of irrigation systems

Soil-Water-Plant Relationships
I. Irrigation Depth
MAD
dx = Wa Z (1)
100
where dx is the maximum net depth of water to apply per irrigation; MAD is
management allowed deficit (usually 40% to 60%); Wa is the water holding
capacity, a function of soil texture and structure, equal to FC – WP (field
capacity minus wilting point); and Z is the root depth
• For most agricultural soils, field capacity (FC) is attained about 1 to 3 days
after a complete irrigation
• The dx value is the same as “allowable depletion.” Actual depth applied may
be less if irrigation frequency is higher than needed during peak use period.
• MAD can also serve as a safety factor because many values (soil data, crop
data, weather data, etc.) are not precisely known
• Assume that crop yield and crop ET begins to decrease below maximum
potential levels when actual soil water is below MAD (for more than one day)
• Water holding capacity for agricultural soils is usually between 10% and 20%
by volume
• Wa is sometimes called “TAW” (total available water), “WHC” (water holding
capacity), “AWHC” (available water holding capacity)
• Note that it may be more appropriate to base net irrigation depth calculations
on soil water tension rather than soil water content, also taking into account
the crop type – this is a common criteria for scheduling irrigations through the
use of tensiometers
II. Irrigation Interval
• The maximum irrigation frequency is:
dx
fx = (2)
Ud
where fx is the maximum interval (frequency) in days; and Ud is the average

daily crop water requirement during the peak-use period
• The range of fx values for agricultural crops is usually:
0.25 < fx < 80 days (3)

• Then nominal irrigation frequency, f’, is the value of fx rounded down to the
nearest whole number of days (
• But, it can be all right to round up if the values are conservative and if fx is
near the next highest integer value
• f’ could be fractional if the sprinkler system is automated
• f’ can be further reduced to account for nonirrigation days (e.g. Sundays),
whereby f ≤ f’
• The net application depth per irrigation during the peak use period is dn =
f’Ud, which will be less than or equal to dx. Thus, dn <= dx, and when dn = dx,
f’ becomes fx (the maximum allowable interval during the peak use period).
• Calculating dn in this way, it is assumed that Ud persists for f’ days, which
may result in an overestimation if f’ represents a period spanning many days
III. Peak Use Period
• Irrigation system design is usually for the most demanding conditions:
• The value of ET during the peak use period depends on the crop type and on
the weather. Thus, the ET can be different from year to year for the same
crop type.
• Some crops may have peak ET at the beginning of the season due to land
preparation requirements, but these crops are normally irrigated by surface
systems.
• When a system is to irrigate different crops (in the same or different
seasons), the crop with the highest peak ET should be used to determine
system capacity.
• Consider design probabilities for ET during the peak use period, because
peak ET for the same crop and location will vary from year-to-year due to
weather variations.
• Consider deficit irrigation, which may be feasible when water is very scarce
and or expensive (relative to the crop value). However, in many cases
farmers are not interested in practicing deficit irrigation.
IV. Leaching Requirement
• Leaching may be necessary if annual rains are not enough to flush the root
zone, or if deep percolation from irrigation is small (i.e. good application
uniformity and or efficiency).
• If ECw is low, it may not be necessary to consider leaching in the design
(system capacity).
• Design equation for leaching:
EC w
LR = (4)
5ECe − EC w
where LR is the leaching requirement; ECw is the EC of the irrigation water

(dS/m or mmho/cm); and ECe is the estimated saturation extract EC of the
soil root zone for a given yield reduction value
• Equation 4 is taken from FAO Irrigation and Drainage Paper 29

• When LR > 0.1, the leaching ratio increases the depth to apply by 1/(1-LR);
otherwise, LR does not need to be considered in calculating the gross depth
to apply per irrigation, nor in calculating system capacity:
dn
LR ≤ 0.1: d= (5)
Ea
0.9 dn
LR > 0.1 d= (6)
(1 − LR)Ea
• When LR < 0.0 (a negative value) the irrigation water is too salty, and the
crop would either die or suffer severely
• Standard salinity vs. crop yield relationships (e.g. FAO) are given for
electrical conductivity as saturation extract
• Obtain saturation extract by adding pure water in lab until the soil is
saturated, then measure the electrical conductivity
• Here are some useful conversions: 1 mmho/cm = 1 dS/m = 550 to 800 mg/l
(depending on chemical makeup, but typically taken as 640 to 690). And, it
can usually be assumed that 1 mg/l ≈ 1 ppm, where ppm is by weight (or
mass).

V. Leaching Requirement Example
Suppose ECw = 2.1 mmhos/cm (2.1 dS/m) and ECe for 10% reduction in crop
yield is 2.5 dS/m. Then,
EC w 2.1
LR = = = 0.20 (7)
5ECe − EC w 5(2.5) − 2.1
Thus, LR > 0.1. And, assuming no loss of water due to application nonuniformity,
the gross application depth is related to the net depth as follows:
dn
d = dn + (LR)d = (8)
1 − LR
and,
dn
d= = 1.25dn (9)
1 − 0.20
See Eq. 5.3 from the textbook regarding nonuniformity losses.
Sprinkle Irrigation Planning Factors
I. Farm Systems vs. Field Systems
• The authors of the textbook only devote a few paragraphs to this topic, but it
is one of great importance
• A complete understanding of the distinctions between farm and field systems
comes only through years of experience
• Variability in design, operation and management conditions is limitless
“A poorly designed system that is well managed can often perform

better than a well designed system that is poorly managed”
• Farm systems may have many field systems

• Planning considerations should include the possibility of future expansions
and extra capacity
• Permanent buried mainlines should generally be oversized to allow for future
needs -- it is much cheaper to put a larger pipe in at the beginning than to
install a secondary or larger line later
• Consider the possibility of future automation
• Consider the needs for land leveling before burying pipes
• How will the system be coordinated over many fields?

• What if the cropping patterns change? (tolerance to salinity, tolerance to
foliar wetting, peak ET rate, root depth, need for crop cooling or frost
protection, temporal shifting of peak ET period)
• What if energy costs change?
• What if labor availability and or cost change?
• What if the water supply is changed (e.g. from river to groundwater, or from
old well to new well)?
• What if new areas will be brought into production?
II. Outline of Sprinkler Design Procedure
1. Make an inventory of resources
• visit the field site personally if at all possible, and talk with the farmer
• get data on soil, topography, water supply, crops, farm schedules, climate, energy,
etc.
• be suspicious of parameter values and check whether they are within reasonable
ranges
2. Calculate a preliminary value for the maximum net irrigation depth, dx

3. Obtain values for peak ET rate, Ud, and cumulative seasonal ET, U (Table 3.3)
4. Calculate maximum irrigation frequency, fx, and nominal frequency, f’
• this step is unnecessary for automated fixed systems and center pivots
5. Calculate the required system capacity, Qs
• first, calculate gross application depth, d

• for center pivots use d/f = Ud, and T ≈ 90% of 24 hrs/day = 21.6
6. Determine the “optimum” (or maximum) water application rate
• a function of soil type and ground slope (Table 5.4)
7. Consider different types of feasible sprinkle systems

8. For periodic-move and fixed (solid-set) systems:
(a) determine Se, qa, nozzle size, and P for optimum application rate
(Tables 6.4 to 6.7)
(b) determine number of sprinklers to operate simultaneously to meet Qs
(Nn = Qs/qa) (Chapter 7)
(c) decide upon the best layout of laterals and mainline (Chapter 7)
(d) Adjust f, d, and/or Qs to meet layout conditions
(e) Size the lateral pipes (Chapter 9)
(f) Calculate the maximum pressure required for individual laterals
9. Calculate the mainline pipe size(s), then select from available sizes
10. Adjust mainline pipe sizes according to the “economic pipe selection method”
(Chapter 10)
11. Determine extreme operating pressure and discharge conditions (Chapter 11)
12. Select the pump and power unit (Chapter 12)
13. Draw up system plans and make a list of items with suggestions for operation
III. Summary
• Note that MAD is not a precise value; actual precision is less than two
significant digits; this justifies some imprecision in other values (don’t try to
obtain very precise values for some parameters when others are only rough
estimates)
• Why use f to determine Qs but f’ to determine net application depth?
(because Qs must be based on gross requirements; not irrigating 24 hrs/day
and 7 days/week does not mean that the crop will not transpire water 7
days/week)
• When determining the seasonal water requirements we subtract Pe from U.
However, to be safe, the value of Pe must be reliable and consistent from
year to year, otherwise a smaller (or zero) value should be used.
• Note that lateral and sprinkler spacings are not infinitely adjustable: they
come in standard dimensions from which designers must choose. The same
goes for pipe diameters and lengths.
• Note that design for peak Ud may not be appropriate if sprinklers are used
only to germinate seeds (when later irrigation is by a surface method).
IV. Example Calculations for a Periodic-Move System
Given:
Crop is alfalfa. Top soil is 1.0 m of silt loam, and subsoil is 1.8 m of clay loam.
Field area is 35 ha. MAD is 50% and ECw is 2.0 dS/m. Application efficiency is
estimated at 80%, and the soil intake rate is 15 mm/hr. Lateral spacing is 15 m and
lateral length is 400 m. Assume it takes ½ hour to change sets. Seasonal effective
rainfall is 190 mm; climate is hot. Assume one day off per week (irrigate only 6
days/week).
From tables in the textbook:
Hot climate, table 3.3 gives............Ud = 7.6 mm/day, and U = 914 mm/season
Top soil, table 3.1 gives ...........................................................Wa = 167 mm/m
Sub soil, table 3.1 gives ...........................................................Wa = 183 mm/m
Root depth, table 3.2 gives .........................................Z = (1.2 + 1.8)/2 = 1.5 m
Salinity for 10% yield reduction, table 3.5 gives ........................ECe = 3.4 dS/m
1. Average water holding capacity in root zone:
top soil is 1.0 m deep; root zone is 1.5 m deep...

1.0 (167 ) + (1.5 − 1.0 )(183 )
Wa = = 172.3 mm/m (10)
1.5
2. Max net application depth (Eq. 3.1):
MAD ⎛ 50 ⎞
dx = Wa Z = ⎜ ⎟ (172.3 )(1.5 ) = 129.2 mm (11)
100 ⎝ 100 ⎠
3. Maximum irrigation interval (Eq. 3.2):
dx 129.2 mm
fx = = = 17.0 days (12)
Ud 7.6 mm/day
4. Nominal irrigation interval (round down, or truncate):
f ' = trunc ( fx ) = 17 days (13)
5. Net application depth:
dn = f 'Ud = (17 days ) ( 7.6 mm/day ) = 129.2 mm (14)
6. Operating time for an irrigation:
17 days is just over two weeks, and depending on which day is off, there
could be 3 off days in this period. So, with one day off per week, we will
design the system capacity to finish in 17 - 3 = 14 days. Thus, f = 14 days.
But, remember that we still have to apply 17 days worth of water in these 14
days (we irrigate 6 days/week but crop transpires 7 days/week)
7. Leaching requirement (Eq. 3.3):
ECw 2.0
LR = = = 0.13 (15)
5ECe − ECw 5 ( 3.4 ) − 2.0
LR > 0.1; therefore, use Eq. 5.3 b...
8. Gross application depth (Eq. 5.3b):
0.9dn 0.9 (129.2 )

d= = = 167.1 mm (16)
(1 − LR )(Ea /100 ) (1 − 0.13 )(0.8 )
9. Nominal set operating time:
With 167.1 mm to apply and a soil intake rate of 15 mm/hr, this gives 11.14
hrs minimum set time (so as not to exceed soil intake rate). Then, we can
make the nominal set time equal to 11.5 hours for convenience. With 0.5 hrs
to move each set, there are a total of 12.0 hrs/set, and the farmer can
change at 0600 and 1800 (for example).
At this point we could take the lateral spacing, Sl, sprinkler spacing, Se, and
actual application rate to determine the flow rate required per sprinkler.
10. Sets per day:
From the above, we can see that there would be two sets/day.
11. Number of sets per irrigation:
(14 days/irrigation)(2 sets/day) = 28 sets
12. Area per lateral per irrigation:
Lateral spacing on mainline is Sl = 15 m. Lateral length is 400 m. Then, the

area per lateral is:
(15 m/set)(28 sets)(400 m/lateral) = 16.8 ha/lateral
13. Number of laterals needed:
35 ha
= 2.08 laterals (17)
16.8 ha/lateral
Normally we would round up to the nearest integer, but because this is so
close to 2.0 we will use two laterals in this design.
14. Number of irrigations per season:
U − Pe 914 mm - 190 mm
= = 5.6 irrigations (18)
dn 129.2 mm/irrig
Thus, it seems there would be approximately six irrigations in a season. But,

the initial Rz value is less than 1.5 m, so there may actually be more than six
irrigations.

15. System flow capacity (Eq. 5.4):
with 11.5 hours operating time per set and two sets per day, the system runs
23 hrs/day...
Qs = 2.78
Ad
= 2.78
( 35 ha )(167.1 mm ) = 50.5 lps (800 gpm)
fT (14 days )( 23 hrs/day ) (19)
This is assuming no effective precipitation during the peak ET period.

Lecture 3
Sprinkler Characteristics
I. Hardware Design Process
1. Sprinkler selection
2. Design of the system layout
3. Design of the laterals
4. Design of the mainline
5. Pump and power unit selection
II. Classification of Sprinklers and Applicability
(see Table 5.1 from the textbook)
• Agricultural sprinklers typically have flow rates from 4 to 45 lpm (1 to 12

gpm), at nozzle pressures of 135 to 700 kPa (20 to 100 psi)
• “Gun” sprinklers may have flow rates up to 2,000 lpm (500 gpm; 33 lps) or
more, at pressures up to 750 kPa (110 psi) or more
• Sprinklers with higher manufacturer design pressures tend to have larger
wetted diameters
• But, deviations from manufacturer’s recommended pressure may have the
opposite effect (increase in pressure, decrease in diameter), and uniformity
will probably be compromised
• Sprinklers are usually made of plastic, brass, and or steel

• Low pressure nozzles save pumping costs, but tend to have large drop sizes
and high application rates
• Medium pressure sprinklers (210 - 410 kPa, or 30 to 60 psi) tend to have the
best application uniformity
• Medium pressure sprinklers also tend to have the lowest minimum
application rates
• High pressure sprinklers have high pumping costs, but when used in
periodic-move systems can cover a large area at each set
• High pressure sprinklers have high application rates
• Rotating sprinklers have lower application rates because the water is only
wetting a “sector” (not a full circle) at any given instance...
• For the same pressure and discharge, rotating sprinklers have larger wetted
diameters
• Impact sprinklers always rotate; the “impact” action on the stream of water is
designed to provide acceptable uniformity, given that much of the water
would otherwise fall far from the sprinkler (the arm breaks up part of the
stream)
• Check out Web sites such as www.rainbird.com
III. Precipitation Profiles
• Typical examples of low, correct, and high sprinkler pressures (see Fig 5.5).
Pressure is too low
Pressure is OK
Pressure is too high
• The precipitation profile (and uniformity) is a function of many factors:
1. nozzle pressure
2. nozzle shape & size
3. sprinkler head design
4. presence of straightening vanes
5. sprinkler rotation speed
6. trajectory angle
7. riser height
8. wind
• Straightening vanes can be used to compensate for consistently windy

conditions
• Overlapping sprinkler profiles (see Fig. 5.7)
uniform! uniform!
lateral
lateral
lateral

• Simulate different lateral spacings by “overlapping” catch-can data in the
direction of lateral movement (overlapping along the lateral is automatically
included in the catch-can data, unless it’s just one sprinkler)
IV. Field Evaluation of Sprinklers
• Catch-can tests are typically conducted to evaluate the uniformities of

installed sprinkler systems and manufacturers’ products
• Catch-can data is often overlapped for various sprinkler and lateral spacings
to evaluate uniformities for design and management purposes
• A computer program developed at USU does the overlapping: CATCH3D;
you can also use a spreadsheet program to simulate overlapping (e.g. Ctrl-C,
Edit | Paste Special, Operation: Add)
• Note that catch-can tests represent a specific wind and pressure situation
and must be repeated to obtain information for other pressures or wind
conditions
• Typical catch-can spacings are 2 or 3 m on a square grid, or 1 to 2 m
spacings along one or more “radial legs”, with the sprinkler in the center
• Set up cans with half spacing from sprinklers (in both axes) to facilitate
overlap calculations
• See Merriam & Keller (1978); also see ASAE S398.1 and ASAE S436
V. Choosing a Sprinkler
• the system designer doesn’t “design” a sprinkler, but “selects” a sprinkler

• there are hundreds of sprinkler designs and variations from several
manufacturers, and new sprinklers appear on the market often
• the system designer often must choose between different nozzle sizes and
nozzle designs for a given sprinkler head design
• the objective is to combine sprinkler selection with Se and Sl to provide
acceptable application uniformity, acceptable pumping costs, and acceptable
hardware costs
• manufacturers provide recommended spacings and pressures
• there are special sprinklers designed for use in frost control
VI. Windy Conditions
• When winds are consistently recurring at some specific hour, the system can
be shut down during this period (T in Eq. 5.4 is reduced)
• For center pivots, rotation should not be a multiple of 24 hours, even if there
is no appreciable wind (evaporation during day, much less at night)
• If winds consistently occur, special straightening vanes can be used
upstream of the sprinkler nozzles to reduce turbulence; wind is responsible
for breaking up the stream, so under calm conditions the uniformity could
decrease
• For periodic-move systems, laterals should be moved in same direction as
prevailing winds to achieve greater uniformity (because Se < Sl)
• Laterals should also move in the direction of wind to mitigate problems of salt
accumulating on plant leaves
• Wind can be a major factor on the application uniformity on soils with low
infiltration rates (i.e. low application rates and small drop sizes)
• In windy areas with periodic-move sprinkler systems, the use of offset
laterals (½Sl) may significantly increase application uniformity
• Alternating the time of day of lateral operation in each place in the field may
also improve uniformity under windy conditions
• Occasionally, wind can help increase uniformity, as the randomness of wind
turbulence and gusts helps to smooth out the precipitation profile
Wind effects on the diameter of throw:
0-3 mph wind: reduce manufacturer’s listed diameter of throw by 10% for an
effective value (i.e. the diameter where the application of
water is significant)
over 3 mph wind: reduce manufacturer’s listed diameter of throw by an

additional 2.5% for every 1 mph above 3 mph (5.6% for every
1 m/s over 1.34 m/s)
In equation form:
For 0-3 mph (0-1.34 m/s):
diam = 0.9 diammanuf (20)
For > 3 mph (> 1.34 m/s):
⎣ (
diam = diammanuf ⎡0.9 − 0.025 windmph − 3 ⎤
⎦ ) (21)
or,
diam = diammanuf ⎡⎣0.9 − 0.056 ( windm / s − 1.34 ) ⎤⎦ (22)
Example: a manufacturer gives an 80-ft diameter of throw for a certain sprinkler

and operating pressure. For a 5 mph wind, what is the effective diameter?
80 ft - (0.10)(0.80) = 72 ft (23)
72 ft - (5 mph - 3 mph)(0.025)(72 ft) = 68 ft (24)

or,
diam = 80(0.9-0.025(5-3))=68 ft (25)
VII. General Spacing Recommendations
• Sprinkler spacing is usually

rectangular or triangular
• Triangular spacing is more common
under fixed-system sprinklers
• Sprinkler spacings based on average
(moderate) wind speeds:
1. Rectangular spacing is 40% (Se)

by 67% (Sl) of the effective
diameter
2. Square spacing is 50% of the
effective diameter
3. Equilateral triangle spacing is
62% of the effective diameter
[lateral spacing is 0.62 cos (60°/2)
= 0.54, or 54% of the effective
diameter, Deffec]
• See Fig. 5.8 about profiles and spacings

VIII. Pressure-Discharge Relationship
• Equation 5.1:
q = Kd P (26)
where q is the sprinkler flow rate; Kd is an empirical coefficient; and P is the

nozzle pressure
• The above equation is for a simple round orifice nozzle

• Eq. 5.1 can be derived from Bernoulli’s equation like this:
P V2 q2
= = (27)
γ 2g 2gA 2
2gA 2P
= Kd P = q (28)
ρg
where the elevations are the same (z1 = z2) and the conversion through the
nozzle is assumed to be all pressure to all velocity
• P can be replaced by H (head), but the value of Kd will be different

• Eq. 5.1 is accurate within a certain range of pressures
• See Table 5.2 for P, q, and Kd relationships
• Kd can be separated into an orifice coefficient, Ko, and nozzle bore area, A:
q = KoA P (29)
whereby,
Ko = 2 / ρ (30)
where the value of Ko is fairly consistent across nozzle sizes for a specific
model and manufacturer

• From Table 5.2 in the textbook, the values of Ko are as follows:
Flow Rate Head or Pressure Nozzle Bore Ko

q H or P d
lps m mm 0.00443
lps kPa mm 0.00137
lpm m mm 0.258
lpm kPa mm 0.0824
gpm ft inch 24.2
gpm psi inch 36.8
• Similar values can be determined from manufacturer’s technical information

• Note also that nozzle diameter (bore) can be determined by rearranging the
above equation as follows:
4q
d= (31)
πK o P
• The value of d can then be rounded up to the nearest available diameter

(64ths of an inch, or mm)
• Then, either P or q are adjusted as necessary in the irrigation system design
• Below is a sample pressure versus discharge table for a RainBird© sprinkler

Application Rates
I. Flow Control Nozzles
• More expensive than regular nozzles

(compare $0.60 for a brass nozzle to
about $2.70 for a flow control nozzle)
• May require more frequent
maintenance
• The orifice has a flexible ring that
decreases the opening with higher
pressures, whereby the value of
A P in the equation remains
approximately constant
• It can be less expensive to design
laterals and mainline so that these types of nozzles are not required, but this
is not always the case
• FCNs are specified for nominal discharges (4, 4.5, 4.8, 5.0 gpm, etc.)
• The manufacturer’s coefficient of variation is about ±5% of q; don’t use FCNs
unless pressure variation is greater than about 10% (along lateral and for
different lateral positions)
1.10P ≈ 1.05 P (32)
II. Low-Pressure Sprinklers
1. Pressure alone is not sufficient to break up the stream in a standard nozzle

design for acceptable application uniformity
2. Need some mechanical method to reduce drop sizes from the sprinkler:
• pins that partially obstruct the stream of water

• sharp-edged orifices
• triangular, rectangular, oval nozzle shapes
3. Some sprinkler companies have invested much into the design of such
devices for low-pressure sprinkler nozzles
4. Low-pressure nozzles can be more expensive, possibly with reduced
uniformity and increased application rate, but the trade-off is in operating cost

III. Gross Application Depth
dn
d= , for LR ≤ 0.1 (33)
Epa
where Epa is the “designer” application efficiency (decimal; Eq. 6.9). And,
0.9 dn
d= , for LR > 0.1 (34)
(1 − LR)Epa
• The gross application depth is the total equivalent depth of water which must
be delivered to the field to replace (all or part of) the soil moisture deficit in
the root zone of the soil, plus any seepage, evaporation, spray drift, runoff
and deep percolation losses
• The above equations for d presume that the first 10% of the leaching
requirement will be satisfied by the Epa (deep percolation losses due to
application variability). This presumes that areas which are under-irrigated
during one irrigation will also be over-irrigated in the following irrigation, or
that sufficient leaching will occur during non-growing season (winter) months.
• When the LR value is small (ECw ≤ ½ECe), leaching may be accomplished
both before and after the peak ET period, and the first equation (for LR ≤ 0.1)
can be used for design and sizing of system components. This will reduce
the required pipe and pump sizes because the “extra” system capacity during
the non-peak ET periods is used to provide water for leaching.
IV. System Capacity
• Application volume can be expressed as either Qt or Ad, where Q is flow

rate, t is time, A is irrigated area and d is gross application depth
• Both terms are in units of volume
• Thus, the system capacity is defined as (Eq. 5.4):
Ad
Qs = K (35)
fT
where,
Qs = system capacity;
T = hours of system operation per day (obviously, T≤ 24; also, t = fT)
K = coefficient for conversion of units (see below)
d = gross application depth (equals Ud/Eff during f’ period)
f = time to complete one irrigation (days); equal to f’ minus the days
off
A = net irrigated area supplied by the discharge Qs

Value of K:
• Metric: for d in mm, A in ha, and Qs in lps: K = 2.78

• English: for d in inches, A in acres, and Qs in gpm: K = 453
Notes about system capacity:
• Eq. 5.4 is normally used for periodic-move and linear-move sprinkler

systems
• The equation can also be used for center pivots if f is decimal days to
complete one revolution and d is the gross application depth per
revolution
• For center pivot and solid-set systems, irrigations can be light and
frequent (dapplied < d): soil water is maintained somewhat below field
capacity at all times (assuming no leaching requirement), and there is
very little deep percolation loss
• Also, there is a margin of safety in the event that the pump fails (or the
system is temporarily out of operation for whatever reason) just when
MAD is reached (time to irrigate), because the soil water deficit is never
allowed to reach MAD
• However, light and frequent irrigations are associated with higher
evaporative losses, and probably higher ET too (due to more optimal soil
moisture conditions). This corresponds to a higher basal crop coefficient
(Kcb + Ks), where Ks is increased, and possibly Kcb too.
• When a solid-set (fixed) system is used for frost control, all sprinklers
must operate simultaneously and the value of Qs is equal to the number
of sprinklers multiplied by qa. This tends to give a higher Qs than that
calculated from Eq. 5.4.
V. Set Sprinkler Application Rate
• The average application rate is calculated as (after Eq. 5.5):
3600qR e
I= (36)
SeSl
where I is the application rate (mm/hr); q is the flow rate (lps); Se is the
sprinkler spacing (m); Sl is the lateral spacing (m); and Re is the fraction of
water emitted by the nozzle that reaches the soil (takes into account the
evaporative/wind loss)
• Re is defined in Chapter 6 of the textbook

• The instantaneous application rate for a rotating sprinkler (after Eq. 5.6):

3600 qRe
Ii = (37)
⎛ S ⎞
πR 2j ⎜ a ⎟
⎝ 360 ⎠
where Ii is the application rate (mm/hr); Rj is the radius of throw, or wetted

radius (m); and Sa is the segment wetted by the sprinkler when the sprinkler
is not allowed to rotate (degrees)
• Note that due to sprinkler overlap, the

instantaneous application rate may actually be
higher than that given by Ii above
• For a non-rotating sprinkler, the instantaneous
application rate is equal to the average
application rate
• For a rotating sprinkler, the instantaneous
p ray
application rate may be allowed to exceed the s
basic intake rate of the soil because excess
(ponded) water has a chance to infiltrate while the
sprinkler completes each rotation
• See sample calculation 5.3 in the textbook
• Higher pressures can give lower instantaneous application rates, but if the
wetted radius does not increase significantly with an increase in pressure,
the instantaneous rate may increase
• The minimum tangential rotation speed at the periphery of the wetted area
should normally be about 1.5 m/s. For example, for 1 rpm:
(1.5 m / s)(60 s / min)

= 14.3m (radius) (38)
(1rev / min)(2π rad / rev )
• Thus, a sprinkler with a wetted radius of 14.3 m should rotate at least 1

rpm
• “Big gun” sprinklers can rotate slower than 1 rpm and still meet this
criterion
VI. Intake & Optimum Application Rates
• Factors influencing the rate at which water should be applied:
1. Soil intake characteristics, field slope, and crop cover

2. Minimum application rate that will give acceptable uniformity
3. Practicalities regarding lateral movement in periodic-move systems

• Impact of water drops on bare soil can cause “surface sealing” effects,
especially on heavy-textured (clayey) soils
• The result is a reduction in infiltration rate due to the formation of a semi-
impermeable soil layer
• Sprinklers typically produce drops from ½ to 5 mm

• Terminal velocity of falling drops is from 2 to 22 m/s
• Water drops from sprinklers typically reach their terminal velocity before
arriving at the soil surface (especially sprinklers with high risers)
• See Tables 5.3 and 5.4 in the textbook
V. Approximate Sprinkler Trajectory
• The trajectory of water from a sprinkler can be estimated according to

physics equations
• The following analysis does not consider aerodynamic resistance nor wind
effects, and is applicable to the largest drops issuing from a sprinkler
operating under a recommended pressure
• Of course, smaller water drops tend to fall nearer to the sprinkler
• In the figure below, Rj refers to the approximate wetted radius of the sprinkler
• If the velocity in the vertical direction at the nozzle, Vy, is taken as zero at
time t1, then,
( Vy )t
1
= V0 sin α − g t1 = 0 (39)
where V0 is the velocity of the stream leaving the nozzle (m/s); α is the angle
of the nozzle; t1 is the time for a drop to travel from the nozzle to the highest
point in the trajectory (s); and g is the ratio of weight to mass (9.81 m/s2)
• Note that the term Vosin α in Eq. 37 is the initial velocity component in the
vertical direction, and the term gt1 is the downward acceleration over time t1
• The above equation can be solved for t1
• The initial velocity, V0, can be calculated based on the sprinkler discharge
and the nozzle diameter
• Values of α can be found from manufacturers’ information
• Now, what is the highest point in the trajectory?
• First, solve for t1 in the previous equation:
Vo sin α
t1 = (40)
g
then,
g t12 V02 sin2 α

h1 = V0 sin α t1 − =
2 2g (41)
• Assuming no acceleration in the horizontal direction,
x1 = V0 cos α t1 (42)
solving for h2,
gt 22
h2 = hr + h1 = Vy t 2 +
2 (43)
where hr is the riser height (m); t2 is the time for a drop of water to travel from
the highest point in the trajectory to impact on the ground; and Vy = 0
• Then, x2 is defined as:
2 (hr + h1 )
x 2 = V0 cos α t 2 = V0 cos α (44)
g
And, the approximate wetted radius of the sprinkler is:
R j = x1 + x 2
(45)
• In summary, if air resistance is ignored and the sprinkler riser is truly vertical,
the theoretical value of Rj is a function of:
1. Angle, α
2. Nozzle velocity (qa/A)
3. Riser height, hr
• And, qa is a function of P

Lecture 4
Set Sprinkler Uniformity & Efficiency
I. Sprinkler Irrigation Efficiency
1. Application uniformity
2. Losses (deep percolation, evaporation, runoff, wind drift, etc.)
• It is not enough to have uniform application if the average depth is not

enough to refill the root zone to field capacity
• Similarly, it is not enough to have a correct average application depth if the
uniformity is poor
• Consider the following examples:
• We can design a sprinkler system that is capable of providing good

application uniformity, but depth of application is a function of the set time (in
periodic-move systems) or “on time” (in fixed systems)
• Thus, uniformity is mainly a function of design and subsequent system
maintenance, but application depth is a function of management

II. Quantitative Measures of Uniformity
Distribution uniformity, DU (Eq. 6.1):
⎛ avg depth of low quarter ⎞

DU = 100 ⎜ ⎟ (46)
⎝ avg depth ⎠
• The average of the low quarter is obtained by measuring application from a
catch-can test, mathematically overlapping the data (if necessary), ranking
the values by magnitude, and taking the average of the values from the low
¼ of all values
• For example, if there are 60 values, the low quarter would consist of the 15
values with the lowest “catches”
Christiansen Coefficient of Uniformity, CU (Eq. 6.2):
⎛
⎜ ∑
n
j=1 (
abs z j − m ) ⎟⎞
CU = 100 1.0 − (47)
⎜ ⎟
∑
n
⎜ z ⎟
⎝ j=1 j ⎠
where z are the individual catch-can values (volumes or depths); n is the
number of observations; and m is the average of all catch volumes.
• Note that CU can be negative if the distribution is very poor

• There are other, equivalent ways to write the equation
• These two measures of uniformity (CU & DU) date back to the time of slide
rules (more than 50 years ago; no electronic calculators), and are designed
with computational ease in mind
• More complex statistical analyses can be performed, but these values have
remained useful in design and evaluation of sprinkler systems
• For CU > 70% the data usually conform to a normal distribution, symmetrical
about the mean value. Then,
⎛ avg depth of low half ⎞

CU ≈ 100 ⎜ ⎟ (48)
⎝ avg depth ⎠
another way to define CU is through the standard deviation of the values,
⎛ σ 2⎞
CU = 100 ⎜ 1.0 − ⎟⎟ (49)
⎜ m π
⎝ ⎠

where σ is the standard deviation of all values, and a normal distribution is
assumed (as previously)
• Note that CU = 100% for σ = 0

• The above equation assumes a normal distribution of the depth values,
whereby:
∑ z − m = nσ 2/ π (50)
• By the way, the ratio σ/m is known in statistics as the coefficient of variation
• Following is the approximate relationship between CU and DU:
CU ≈ 100 − 0.63(100 − DU) (51)

or,
DU ≈ 100 − 1.59(100 − CU) (52)
• These equations are used in evaluations of sprinkler systems for both design
and operation
• Typically, 85 to 90% is the practical upper limit on DU for set systems
• DU > 65% and CU > 78% is considered to be the minimum acceptable
performance level for an economic system design; so, you would not
normally design a system for a CU < 78%, unless the objective is simply to
“get rid of water or effluent” (which is sometimes the case)
• For shallow-rooted, high value crops, you may want to use DU > 76% and
CU > 85%
III. Alternate Sets (Periodic-Move Systems)
• The effective uniformity (over multiple irrigations) increases if “alternate sets”

are used for periodic-move systems (½Sl)
• This is usually practiced by placing laterals halfway between the positions
from the previous irrigation, alternating each time
• The relationship is:
CUa ≈ 10 CU
(53)
DUa ≈ 10 DU
• The above are also valid for “double” alternate sets (Sl/3)
• Use of alternate sets is a good management practice for periodic-move
systems
• The use of alternate sets approaches an Sl of zero, which simulates a
continuous-move system

IV. Uniformity Problems
• Of the various causes of non-uniform sprinkler application, some tend to

cancel out with time (multiple irrigations) and others tend to concentrate (get
worse)
• In other words, the “composite” CU for two or more irrigations may be (but
not necessarily) greater than the CU for a single irrigation
1. Factors that tend to Cancel Out
• Variations in sprinkler rotation speed

• Variations in sprinkler discharge due to wear
• Variations in riser angle (especially with hand-move systems)
• Variations in lateral set time
2. Factors that may both Cancel Out and Concentrate
• Non-uniform aerial distribution of water between sprinklers
3. Factors that tend to Concentrate
• Variations in sprinkler discharge due to elevation and head loss

• Surface ponding and runoff
• Edge effects at field boundaries
V. System Uniformity
• The uniformity is usually less when the entire sprinkler system is considered,
because there tends to be greater pressure variation in the system than at
any given lateral position.
⎡1
( ⎤
system CU ≈ CU ⎢ 1 + Pn / Pa ⎥
⎣2 ⎦
) (54)
⎡1
( ⎤
system DU ≈ DU ⎢ 1 + 3 Pn / Pa ⎥
⎣4 ⎦
) (55)
where Pn is the minimum sprinkler pressure in the whole field; and Pa is the
average sprinkler pressure in the entire system, over the field area.
• These equations can be used in design and evaluation

• Note that when Pn = Pa (no pressure variation) the system CU equals the CU
• If pressure regulators are used at each sprinkler, the system CU is
approximately equal to 0.95CU (same for DU)
• If flexible orifice nozzles are used, calculate system CU as 0.90CU (same for
DU)
• The Pa for a system can often be estimated as a weighted average of Pn &
Px:
2Pn + Px
Pa = (56)
3
where Px is the maximum nozzle pressure in the system
VI. Computer Software and Standards
• There is a computer program called “Catch-3D” that performs uniformity

calculations on sprinkler catch-can data and can show the results graphically
• Jack Keller and John Merriam (1978) published a handbook on the
evaluation of irrigation systems, and this includes simple procedures for
evaluating the performance of sprinkler systems
• The ASAE S436 (Sep 92) is a detailed standard for determining the
application uniformity under center pivots (not a set sprinkler system, but a
continuous move system)
• ASAE S398.1 provides a description of various types of information that can
be collected during an evaluation of a set sprinkler system

VII. General Sprinkle Application Efficiency
The following material leads up to the development of a general sprinkle

application efficiency term (Eq. 6.9) as follows:
Design Efficiency:
Epa = DEpaReOe
(57)
where DEpa is the distribution efficiency (%); Re is the fraction of applied

water that reaches the soil surface; and Oe is the fraction of water that does
not leak from the system pipes.
• The design efficiency, Epa, is used to determine gross application depth (for
design purposes), given the net application depth
• In most designs, it is not possible to do a catch-can test and data analysis –
you have to install the system in the field first; thus, use the “design
efficiency”
• The subscript “pa” represents the “percent area” of the field that is
adequately irrigated (to dn, or greater) – for example, E80 and DE80 are the
application and distribution efficiencies when 80% of the field is adequately
irrigated
• Question: can “pa” be less than 50%?

20% of area underirrigated
U
80% of area overirrigated
C
w
Lo
Relative Applied Depth
h CU
Hig
Desired Net Application Depth

1.0
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%
Area Receiving at Least the Desired Application
VIII. Distribution Efficiency
• This is used to define the uniformity and adequacy of irrigation

• DE is based on statistical distributions and application uniformity
• For a given uniformity (CU) and a given percent of land adequately irrigated
(equal to or greater than required application depth), Table 6.2 gives values
of DE that determine how much water must be applied in excess of the
required depth so that the given percent of land really does receive at least
the required depth
Percent area adequately irrigated (pa)

CU
95 90 85 80 75 70 65 60 55 50
94 87.6 90.4 92.2 93.7 94.9 96.1 97.1 98.1 99.1 100.0
92 83.5 87.1 89.6 91.6 93.2 94.7 96.1 97.5 98.7 100.0
90 79.4 83.9 87.0 89.4 91.5 93.4 95.2 96.8 98.4 100.0
88 75.3 80.7 84.4 87.3 89.8 92.1 94.2 96.2 98.1 100.0
86 71.1 77.5 81.8 85.2 88.2 90.8 93.2 95.6 97.8 100.0
84 67.0 74.3 79.2 83.1 86.5 89.5 92.3 94.9 97.5 100.0
82 62.9 71.1 76.6 81.0 84.8 88.2 91.3 94.3 97.2 100.0
80 58.8 67.9 74.0 78.9 83.1 86.8 90.3 93.6 96.8 100.0
78 54.6 64.7 71.4 76.8 81.4 85.5 89.4 93.0 96.5 100.0
76 50.5 61.4 68.8 74.7 79.7 84.2 88.4 92.4 96.2 100.0
74 46.4 58.2 66.2 72.6 78.0 82.9 87.4 91.7 95.9 100.0
72 42.3 55.0 63.6 70.4 76.3 81.6 86.5 91.1 95.6 100.0
70 38.1 51.8 61.0 68.3 74.6 80.3 85.5 90.5 95.3 100.0
• See Fig. 6.7

IX. Wind Drift and Evaporation Losses
• These losses are typically from 5% to 10%, but can be higher when the air is
dry, there is a lot of wind, and the water droplets are small
• Effective portion of the applied water, Re. This is defined as the percentage
of applied water that actually arrives at the soil surface of the irrigated field.
• This is based on:
• climatic conditions
• wind speed
• spray coarseness
• Figure 6.8 gives the value of Re for these different factors

• The Coarseness Index, CI, is defined as (Eq. 6.7):
⎛ P1.3 ⎞
CI = 0.032 ⎜
⎜ B ⎟⎟
(58)
⎝ ⎠
where P is the nozzle pressure (kPa) and B is the nozzle diameter (mm)
CI > 17 17 ≥ CI ≥ 7 CI < 7
fine spray between fine and coarse coarse spray
• When the spray is between fine and coarse, Re is computed as a weighted

average of (Re)fine and (Re)coarse (Eq. 6.8):
(CI − 7) (17 − CI)

Re = (Re )fine + (Re )coarse (59)
10 10
• Allen and Fisher (1988) developed a regression equation to fit the curves in
Fig. 6.8:
Re = 0.976 + 0.005ETo − 0.00017ETo2 + 0.0012 W

−0.00043(CI)(ETo ) − 0.00018(CI)(W) (60)
−0.000016(CI)(ETo )(W)
where ETo is the reference ET in mm/day (grass-based); CI is the

coarseness index (7 ≤ CI ≤ 17); and W is the wind speed in km/hr
• For the above equation, if CI < 7 then set it equal to 7; if CI > 17 then set it
equal to 17

X. Leaks and Drainage Losses
1. Losses due to drainage of the system after shut-down
• upon shut-down, most sprinkler systems will partially drain

• water runs down to the low elevations and or leaves through
automatic drain valves that open when pressure drops
• fixed (solid-set) systems can have anti-drain valves at sprinklers that
close when pressure drops (instead of opening, like on wheel lines)
2. Losses due to leaky fittings, valves, and pipes
• pipes and valves become damaged with handling, especially with

hand-move and side-roll systems, but also with orchard sprinklers and
end-tow sprinklers
• gaskets and seals become inflexible and fail
• These losses are quantified in the Oe term

• For systems in good condition these losses may be only 1% or 2%, giving an
Oe value of 99% or 98%, respectively
• For system in poor condition these losses can be 10% or higher, giving an Oe
value of 90% or less
XI. General Sprinkle Application Efficiency
• As given above, Eq. 6.9 from the textbook, it is:
Epa = DEpaReOe
(61)
where DEpa is in percent; and Re and Oe are in fraction (0 to 1.0). Thus, Epa
is in percent.
XII. Using CU or DU instead of DEpa
1. Application Efficiency of the Low Quarter, Eq
• Given by Eq. 6.9 when DU replaces DEpa

• Useful for design purposes for medium to high-value crops
• Only about 10% of the area will be under-irrigated
• Recall that DU is the average of low quarter divided by average
2. Application Efficiency of the Low Half, Eh
• Given by Eq. 6.9 when CU replaces DEpa

• Useful for design purposes for low-value and forage crops
• Only about 20% of the area will be under-irrigated
• Recall that CU is the average of low half divided by average
XIII. Procedure to Determine CU, Required Pressure, Se and Sl for a Set

System
1. Specify the minimum acceptable Epa and target pa

2. Estimate Re and Oe (these are often approximately 0.95 and 0.99,
respectively)
3. Compute DEpa from Epa, Re and Oe
4. Using DEpa and pa, determine the CU (Table 6.2) that is required to achieve
Epa
5. Compute the set operating time, tso, then adjust f’ and dn so that tso is an
appropriate number of hours
6. Compute qa based on I, Se and Sl (Eq. 5.5)
7. Search for nozzle size, application rate, Se and Sl to obtain the CU
8. Repeat steps 5, 6 and 7 as necessary until a workable solution is found
XIV. How to Measure Re
• The textbook suggests a procedure for estimating Re

• You can also measure Re from sprinkler catch-can data:
1. Compute the average catch depth over the wetted area (if a single
sprinkler), or in the area between four adjacent sprinklers (if in a
rectangular grid)
2. Multiply the sprinkler flow rate by the total irrigation time to get the volume
applied, then divide by the wetted area to obtain the gross average
application depth
3. Divide the two values to determine the effective portion of the applied
water
XV. Line- and Point-Source Sprinklers
• Line-source sprinklers are sometimes used by researches to determine the

effects of varying water application on crop growth and yield
• A line-source sprinkler system consists of sprinklers spaced evenly along a
straight lateral pipe in which the application rate varies linearly with distance
away from the lateral pipe, orthogonally
• Thus, a line-source sprinkler system applies the most water at the lateral
pipe, decreasing linearly to zero to either side of the lateral pipe
• A point-source sprinkler is a single sprinkler that gives linearly-varying
application rate with radial distance from the sprinkler

• With a point-source sprinkler, the contours of equal application rate are
concentric circles, centered at the sprinkler location (assuming the riser is
vertical and there is no wind)

Lecture 5
Layout of Laterals for Set Sprinklers
I. Selecting Sprinkler Discharge, Spacing, and Pressure
• In Chapter 6 of the textbook there are several tables that provide guidelines
for nozzle sizes for different:
• Wind conditions
• Application rates
• Sprinkler spacings
• For selected values of wind, application rate, and spacing, the tables provide
recommended nozzle sizes for single and double-nozzle sprinklers,
recommended sprinkler pressure, and approximate uniformity (CU)
• Table values are for standard (non-flexible) nozzles

• Table values are for standard sprinkler and lateral spacings
• More specific information can be obtained from manufacturers’ data
• Recall that the maximum application rate is a function of soil texture, soil
structure, and topography (Table 5.4)
• For a given spacing and application rate, the sprinkler discharge, qa, can be
determined from Eq. 5.5
I ( SeSl ) d S SO
qa = = n e l e (62)
3600 3600EpaSto
where qa is in lps; I is in mm/hr; dn is in mm; Sto is the operating time for each
set, in hours; and Sl and Se are in m
• Why is the Oe term included in the above equation? (because Epa includes
Oe, as previously defined, and must be cancelled out when considering an
individual sprinkler)
II. Number of Operating Sprinklers
• After calculating the system capacity and the design flow rate for sprinklers,
the number of sprinklers that will operate at the same time is:
Qs
Nn = (63)
qa

where Nn is the minimum number of sprinklers operating, and Qs and qa have
the same units
• It is recommendable to always operate the same number of sprinklers when

the system is running. This practice can help avoid the need for pressure
regulation, and can avoid uniformity problems. It can also help avoid wasting
energy at the pump.
• For odd-shaped fields, and sometimes for rectangular fields, it is not possible
to operate the same number of sprinklers for all sets. In this case, pressure
regulation may be necessary, or other steps can be taken (multiple pumps,
variable-speed motor, variable application rates).
III. Lateral Design Criteria
• Lateral pressure varies from inlet to extreme end due to:
1. friction loss
2. elevation change
• The fundamental basis upon which sprinkler laterals are designed is:
“pressure head variation in the lateral should not exceed

20% of the average design pressure for the sprinklers”
• This is a design assumption that has been used for many years, and is
based on a great deal of experience
• The 20% for pressure variation is not an “exact” value; rather, it is based on
judgment and some cost comparisons
• A designer could change this value, but it would affect system performance
(uniformity), initial system cost, operating cost, and possibly other factors
• Computer programs could be written to search for an “optimal” percent
pressure variation according to initial and operating costs, and according to
crop value -- such an “optimal” value would vary from system to system
IV. Sprinkler Lateral Orientation
• It is usually preferable to run laterals on contours (zero slope) so that

pressure variation in the lateral pipes is due to friction loss only
• It is advantageous to run laterals downhill, if possible, because the gain in
energy due to elevation change will allow longer laterals without violating the
20% rule. But, if the slope is too steep, pressure regulators or flow control
nozzles may be desirable.
• If the ground slope is equal to the friction loss gradient, the pressure in the
lateral will be constant. However, the friction loss gradient is nonlinear
because the flow rate is decreasing with distance along the lateral.
• It is usually not recommendable to run laterals in an uphill direction. In this
case:
1. both friction loss and elevation are working to reduce pressure toward
the end of the lateral, and length is more restricted if the 20% rule is
still used
2. However, for small slopes, running laterals uphill may be required to
reduce the total length of the mainline pipe
• Note that V2/2g in the lateral pipe is normally converted into total head as the
water flows through the nozzle body. Therefore, the velocity head (and EL)
should normally be considered in lateral design. However, since a portion of
the velocity head is lost during deceleration of the water at the entrance into
risers and as turbulence inside the sprinkler head, and since V2/2g in a
lateral pipe is typically small (< 1 ft of head, or 0.2 psi, or 0.3 m head, or 3
kPa), it is normally neglected during design, and the HGL is used.
• Aside from limits on pressure variation, laterals should be oriented so that
they move in the direction of the prevailing winds -- this is because of salinity
problems and application uniformity
• Figure 7.1 gives examples of layouts on different topographies
V. Lateral Sizing Limitations
• lateral pipes can be designed with multiple diameters to accommodate

desirable pressure distributions, but...
• hand-move laterals should have only one or two different pipe sizes to
simplify handling during set changes
• in practice, hand-move systems and wheel lines usually have only one size
of lateral pipe
• some wheel lines, greater than 400 m in length, may have 5-inch pipe near
the inlet and then 4-inch pipe at the end
Layout of Mainline for Set Sprinklers
I. Mainline Layout and Sizing
• if possible, run the mainline up or down slope so the laterals can be on

contours (lateral pressure variation due to friction loss only)
• can also run the mainline along a ridge so the laterals run downhill on both
sides (lateral friction loss partially offset by elevation change)
• should consider possible future expansions when sizing the mainline

“Split-Line” Lateral Operation:
• laterals operate on both sides of the mainline

• the mainline can be sized for only half capacity halfway down the mainline if
laterals are run in different directions
• sometimes interferes with cultural practices
• it is convenient to have the water supply in the center of one side of the field,
but this is seldom a design variable (the well is already there, or the canal is
already there)
• may not need pumping if the water supply is at a higher elevation than the
field elevation (e.g. 50 psi = 115 ft or 35 m of head) -- when pumping is not
required, this changes the mainline layout and pipe sizing strategy
• in some cases it will be justifiable to include one or more booster pumps in
the design -- even when the water source is a well (the well pump may not
provide enough pressure for any of the lateral settings)
• we will discuss mainline economics in the next few lectures, then we will look
at mainline design in more detail later
II. Design Variables to Accommodate Layout
• Number of sprinklers operating

• Average application rate
• Gross application depth
• Average sprinkler discharge
• Sprinkler spacing
• Operating hours per day
• Irrigation frequency
• Total operating time (fT)
• System capacity
• Percent probability of rain during peak-use period
• MAD
• It may be necessary to adjust the layout if a suitable combination of the

above variables cannot be found
• Can also use flow control nozzles or pressure regulators to accommodate a
given layout
III. Sample Calculation
• Consider a periodic-move system with Sl = 50 ft, Se = 40 ft, f = 8 days, T =

11.5 hrs @ 2 sets/day, d = 2.7”, and qa = 4.78 gpm
• The field size is 80 acres (½ of a “quarter section”), 2,640 ft on one side and
1,320 ft on the other, rectangular
• The laterals will have to be 1,320-ft long

• System capacity:
453(80 ac)(2.7 inch)

Qs = = 532gpm
(8 days)(2 sets / day)(11.5 hrs / set) (64)
• Number of sprinklers operating:
Qs 532
Ns = = = 111 sprinklers
qa 4.78 (65)
• Number of laterals,
1320 ft / lateral
= 33 sprinklers / lateral
40 ft / sprinkler (66)
111 sprinklers
= 3.36 laterals
33 sprinklers / lateral (67)
...so, round up to 4 laterals
• Thus, two laterals on each side of the mainline (symmetry)
1320 ft per lateral pair

= 26.4 (68)
50 ft / position
• Round this up from 26.4 to 27 positions per lateral pair

• This gives 2 x 27 = 54 total lateral positions, and 54/4 = 13.5 sets/lateral
• Use 13 sets for two laterals and 14 sets for the other two laterals
• Then, there will be 14 sets per irrigation, even though the last set will only
have two laterals operating
• Adjusted irrigation frequency:
14 sets
f= = 7 days (69)
2 sets / day
• Note that the value of f was for an 8-day interval

• Thus, we need to increase Qs to complete the irrigation in less time
• Adjusted system capacity:
Qs = (4 laterals)(33 sprinklers / lateral)(4.78 gpm / sprinkler)

(70)
= 631 gpm
• Another way to adjust the system capacity:
⎛ 8 days ⎞
Qs = ⎜ ⎟ (532 gpm) = 608 gpm (71)
⎝ 7 days ⎠
• You might say that we are “effectively” finishing in somewhat less than 7
days, because the last set has only two laterals in operation, giving a system
capacity of 608 instead of 631
• Consider this calculation: there are 2 x 13 + 2 x 14 = 54 sets, but the last 2
sets have only 2 laterals. So, (52/54) x 631 = 608 gpm, as calculated above.
• Which is correct?
• There are (52/54)*(4 laterals) = 3.85 laterals operating on average during

each irrigation of the field
• However, you cannot always base the system capacity on the average
number of laterals operating
• The system capacity should be based on the “worst case”, which is when all
four laterals operate simultaneously
• This means that the required capacity is 631 gpm, not 608 gpm
• Note that many farmers will accept some increase in system capital cost to
provide more operational flexibility and safety
• In summary, we have essentially lowered f to accommodate the system
configuration (layout), but:
• same gross depth

• same number of hours per set
• same sprinkler flow rate
• same sprinkler spacing
• increased system capacity

Pipeline Hydraulics
I. Review
• Read Chapter 8 of the textbook to review the hydraulics of pipelines

• For pipe friction loss we will be using the Hazen-Williams and Darcy-
Weisbach equations
• Be familiar with the Moody diagram, for use with the Darcy-Weisbach
equation
• You can use the Swamee-Jain equation instead of the Moody diagram:
0.25
f= 2
(72)
⎡ ⎛ ε 5.74 ⎞ ⎤
⎢log10 ⎜ + 0.9 ⎟ ⎥
⎣ ⎝ 3.75D NR ⎠ ⎦
which is valid for turbulent flow in the range: 4,000 ≤ NR ≤ 1.0(10)8. The ratio
ε/D is called “relative roughness.” The roughness height, ε, varies widely
• We will also use the Blasius equation (Eq. 8.6) to determine the value of “f,”
in some cases, for “smooth pipes”

Lecture 6
Economic Pipe Selection Method
I. Introduction
• The economic pipe selection method (Chapter 8 of the textbook) is used to

balance fixed (initial) costs for pipe with annual energy costs for pumping
• With larger pipe sizes the average flow velocity for a given discharge
decreases, causing a corresponding decrease in friction loss
• This reduces the head on the pump, and energy can be saved
• However, larger pipes cost more to purchase
total
minimum
total
Cost
Energy costs =
annualized fixed costs
fixed energy
Pipe Size (diameter)
• To balance these costs and find the minimum cost we will annualize the fixed
costs, compare with annual energy (pumping) costs
• We can also graph the results so that pipe diameters can be selected
according to their maximum flow rate
• We will take into account interest rates and inflation rates to make the
comparison
• This is basically an “engineering economics” problem, specially adapted to
the selection of pipe sizes
• This method involves the following principal steps:
1. Determine the equivalent annual cost for purchasing each available pipe
size
2. Determine the annual energy cost of pumping
3. Balance the annual costs for adjacent pipe sizes
4. Construct a graph of system flow rate versus section flow rate on a log-
log scale for adjacent pipe sizes
• We will use the method to calculate “cut-off” points between adjacent pipe
sizes so that we know which size is more economical for a particular flow
rate
• We will use HP and kW units for power, where about ¾ of a kW equals a HP
• Recall that a Watt (W) is defined as a joule/second, or a N-m per second
• Multiply W by elapsed time to obtain Newton-meters (“work”, or “energy”)
II. Economic Pipe Selection Method Calculations
1. Select a period of time over which comparisons will be made between fixed
and annual costs. This will be called the useful life of the system, n, in years.
• The “useful life” is a subjective value, subject to opinion and financial

amortization conditions
• This value could alternatively be specified in months, or other time period,
but the following calculations would have to be consistent with the choice
2. For several different pipe sizes, calculate the uniform annual cost of pipe per
unit length of pipe.
• A unit length of 100 (m or ft) is convenient because J is in m/100 m or

ft/100 ft, and you want a fair comparison (the actual pipe lengths from the
supplier are irrelevant for these calculations)
• You must use consistent units ($/100 ft or $/100 m) throughout the
calculations, otherwise the ∆J values will be incorrect (see Step 11 below)
• So, you need to know the cost per unit length for different pipe sizes
• PVC pipe is sometimes priced by weight of the plastic material (weight
per unit length depends on diameter and wall thickness)
• You also need to know the annual interest rate upon which to base the
calculations; this value will take into account the time value of money,
whereby you can make a fair comparison of the cost of a loan versus the
cost of financing it “up front” yourself
• In any case, we want an equivalent uniform annual cost of the pipe over
the life of the pipeline
• Convert fixed costs to equivalent uniform annual costs, UAC, by using the
“capital recovery factor”, CRF
UAC = P ( CRF ) (73)

i (1 + i )
n
CRF = (74)
(1 + i) − 1
n
where P is the cost per unit length of pipe; i is the annual interest rate
(fraction); and n is the number of years (useful life)
• Of course, i could also be the monthly interest rate with n in months, etc.
• Make a table of UAC values for different pipe sizes, per unit length of pipe
• The CRF value is the same for all pipe sizes, but P will change depending
on the pipe size
• Now you have the equivalent annual cost for each of the different pipe
sizes
3. Determine the number of operating (pumping) hours per year, Ot:
(irrigated area)(gross annual depth)

Ot = = hrs / year (75)
(system capacity)
• Note that the maximum possible value of Ot is 8,760 hrs/year (for 365
days)
• Note also that the “gross depth” is annual, so if there is more than one
growing season per calendar year, you need to include the sum of the
gross depths for each season (or fraction thereof)
4. Determine the pumping plant efficiency:
• The total plant efficiency is the product of pump efficiency, Epump, and
motor efficiency, Emotor
Ep = EpumpEmotor
(76)
• This is equal to the ratio of “water horsepower”, WHP, to “brake

horsepower”, BHP (Epump ≡ WHP/BHP)

• Think of BHP as the power going into the pump through a spinning shaft,
and WHP is what you get out of the pump – since the pump is not 100%
efficient in energy conversion, WHP < BHP
• WHP and BHP are archaic and confusing terms, but are still in wide use
• Emotor will usually be 92% or higher (about 98% with newer motors and
larger capacity motors)
• Epump depends on the pump design and on the operating point (Q vs.
TDH)
• WHP is defined as:
QH
WHP = (77)
102
where Q is in lps; H is in m of head; and WHP is in kilowatts (kW)
• If you use m in the above equation, UAC must be in $/100 m

• If you use ft in the above equation, UAC must be in $/100 ft
• Note that for fluid flow, “power” can be expressed as ρgQH = γQH
• Observe that 1,000/g = 1,000/9.81 ≈ 102, for the above units (other
conversion values cancel each other and only the 102 remains)
• The denominator changes from 102 to 3,960 for Q in gpm, H in ft, and
WHP in HP
5. Determine the present annual energy cost:
OtCf
E=
Ep
(78)
where Cf is the cost of “fuel”
• For electricity, the value of Cf is usually in dollars per kWh, and the value
used in the above equation may need to be an “average” based on
potentially complex billing schedules from the power company
• For example, in addition to the energy you actually consume in an electric
motor, you may have to pay a monthly fee for the installed capacity to
delivery a certain number of kW, plus an annual fee, plus different time-
of-day rates, and others
• Fuels such as diesel can also be factored into these equations, but the
power output per liter of fuel must be estimated, and this depends partly
on the engine and on the maintenance of the engine
• The units of E are dollars per WHP per year, or dollars per kW per year;
so it is a marginal cost that depends on the number of kW actually
required
6. Determine the marginal equipment cost:
• Note that Cf can include the “marginal” cost for the pump and power unit
(usually an electric motor)
• In other words, if a larger pump & motor costs more than a smaller pump,
then Cf should reflect that, so the full cost of friction loss is considered
• If you have higher friction loss, you may have to pay more for energy to
pump, but you may also have to buy a larger pump and/or power unit
(motor or engine)
• It sort of analogous to the Utah Power & Light monthly power charge,
based solely on the capacity to deliver a certain amount of power
Cf ($/kWh) = energy cost + marginal cost for a larger pump & motor
where “marginal” is the incremental unit cost of making a change in the

size of a component
• This is not really an “energy” cost per se, but it is something that can be
taken into account when balancing the fixed costs of the pipe (it falls
under the operating costs category, increasing for decreasing pipe costs)
• That is, maybe you can pay a little more for a larger pipe size and avoid
the need to buy a bigger pump, power unit and other equipment
• To calculate the marginal annual cost of a pump & motor:
MAC =
(
CRF $big − $small )
(
Ot kWbig − kWsmall ) (79)
where MAC has the same units as Cf; and $big -$small is the difference in
pump+motor+equipment costs for two different capacities
• The difference in fixed purchase price is annualized over the life of the
system by multiplying by the CRF, as previously calculated
• The difference in pump size is expressed as ∆BHP, where ∆BHP is the
difference in brake horsepower, expressed as kW
• To determine the appropriate pump size, base the smaller pump size on a
low friction system (or low pressure system)
• For BHP in kW:
QsHpump
BHP =
102Ep
(80)

• Round the BHP up to the next larger available pump+motor+equipment
size to determine the size of the larger pump
• Then, the larger pump size is computed as the next larger available pump
size as compared to the smaller pump
• Then, compute the MAC as shown above
• The total pump cost should include the total present cost for the pump,
motor, electrical switching equipment (if appropriate) and installation
• Cf is then computed by adding the cost per kWh for energy
• Note that this procedure to determine MAC is approximate because the

marginal costs for a larger pump+motor+equipment will depend on the
magnitude of the required power change
• Using $big -$small to determine MAC only takes into account two (possibly
adjacent) capacities; going beyond these will likely change the marginal
rate
• However, at least we have a simple procedure to attempt to account for
this potentially real cost
7. Determine the equivalent annualized cost factor:
• This factor takes inflation into account:
⎡ (1 + e )n − (1 + i )n ⎤ ⎡ i ⎤
EAE = ⎢ ⎥⎢ ⎥ (81)
⎢ e −i ⎥ ⎢ (1 + i )n − 1⎥
⎣ ⎦⎣ ⎦
where e is the annual inflation rate (fraction), i is the annual interest rate
(fraction), and n is in years
• Notice that for e = 0, EAE = unity (this makes sense)

• Notice also that the above equation has a mathematical singularity for e =
i (but i is usually greater than e)
8. Determine the equivalent annual energy cost:
E ' = (EAE)(E) (82)
• This is an adjustment on E for the expected inflation rate

• No one really knows how the inflation rate might change in the future
• How do you know when to change to a larger pipe size (based on a
certain sectional flow rate)?
Beginning with a smaller pipe size (e.g. selected based

on maximum velocity limits), you would change to a
larger pipe size along a section of pipeline if the
difference in cost for the next larger pipe size is less than
the difference in energy (pumping) savings
• Recall that the velocity limit is usually taken to be about 5 fps, or 1.5 m/s
9. Determine the difference in WHP between adjacent pipe sizes by equating the
annual plus annualized fixed costs for two adjacent pipe sizes:
E' (HPs1 ) + UACs1 = E' (HPs2 ) + UACs2

(83)
or,
∆WHPs1−s2 =
(UACs2 − UACs1)
E' (84)
• The subscript s1 is for the smaller of the two pipe sizes

• The units of the numerator might be $/100 m per year; the units of the
denominator might be $/kW per year
• This is the WHP (energy) savings needed to offset the annualized fixed
cost difference for purchasing two adjacent pipe sizes; it is the economic
balance point
10. Determine the difference in friction loss gradient between adjacent pipe sizes:
⎛ ∆WHPs1− s2 ⎞
∆Js1− s2 = 102 ⎜ ⎟
⎝ Qs ⎠ (85)
• This is the head loss difference needed to balance fixed and annual costs
for the two adjacent pipe sizes
• The coefficient 102 is for Qs in lps, and ∆WHP in kW
• You can also put Qs in gpm, and ∆WHP in HP, then substitute 3,960 for
102, and you will get exactly the same value for ∆J
• As before, ∆J is a head loss gradient, in head per 100 units of length (m
or ft, or any other unit)
• Thus, ∆J is a dimensionless “percentage”: head, H, can be in m, and
when you define a unit length (e.g. 100 m), the H per unit meter becomes
dimensionless
• This is why you can calculate ∆J using any consistent units and you will
get the same result
11. Calculate the flow rate corresponding to this head loss difference:
• Using the Hazen-Williams equation:

1.852
∆J = Js1 − Js2
q⎞
= 16.42(10) ⎜ ⎟
⎝C⎠
6⎛
(Ds1−4.87 − Ds2−4.87 ) (86)
where q is in lps, and D is the inside diameter of the pipe in cm
• Or, using the Darcy-Weisbach equation:
∆J =
800 f q2
gπ 2 (Ds1−5 − Ds2−5 ) (87)
• Solve for the flow rate, q (with q in lps; D in cm):
0.54
⎡ ⎤
∆J
q = C⎢ ⎥
( )
(88)
⎢ 16.42(10)6 D−4.87 − D−4.87 ⎥
⎣ s1 s2 ⎦
• This is the flow rate for which either size (Ds1 or Ds2) will be the most
economical (it is the balancing point between the two adjacent pipe sizes)
• For a larger flow rate you would choose size Ds2, and vice versa
12. Repeat steps 8 through 11 for all other adjacent pipe sizes.
13. You can optionally create a graph with a log-log scale with the system flow
rate, Qs, on the ordinate and the section flow rate, q, on the abscissa:
• Plot a point at Qs and q for each of the adjacent pipe sizes

• Draw a straight diagonal line from lower left to upper right corner
• Draw a straight line at a slope of -1.852 (or -2.0 for Darcy-Weisbach)
through each of the points
• The slope will be different if the log scale on the axes are not the same
distance (e.g. if you do the plot on a spreadsheet, the ordinate and
abscissa may be different lengths, even if the same number of log cycles).
• In constructing the graph, you can get additional points by changing the
system flow rate, but in doing so you should also increase the area, A, so
that Ot is approximately the same as before. It doesn’t make sense to
change the system flow rate arbitrarily.
• Your graph should look similar to the one shown below

14. Applying the graph.
• Find the needed flow rate in a given section of the pipe, q, make an
intersection with the maximum system capacity (Qs, on the ordinate), then
see which pipe size it is
• You can use the graph for different system capacities, assuming you are
considering different total irrigated areas, or different crop and or climate
values
• Otherwise, you can just skip step 13 and just do the calculations on a
spreadsheet for the particular Qs value that you are interested in
• The graph is perhaps didactic, but doesn’t need to be constructed to
apply this economic pipe selection method
III. Notes on the Use of this Method
1. If any of the economic factors (interest rate, inflation rate, useful life of the
system) change, the lines on the graph will shift up or down, but the slope
remains the same (equal to the inverse of the velocity exponent for the head
loss equation: 1.852 for Hazen-Williams and 2.0 for Darcy-Weisbach).
2. Computer programs have been developed to use this and other economic
pipe selection methods, without the need for constructing a graphical solution
on log-log paper. You could write such a program yourself.
3. The economic pipe selection method presented above is not necessarily

valid for:
• looping pipe networks
• very steep downhill slopes
• non-“worst case” pipeline branches
4. For loops, the flow might go in one direction some of the time, and in the
opposite direction at other times. For steep downhill slopes it is not
necessary to balance annual operation costs with initial costs because there
is essentially no cost associated with the development of pressure – there is
no need for pumping. Non-“worst case” pipeline branches may not have the
same pumping requirements (see below).
5. Note that the equivalent annual pipe cost considers the annual interest rate,
but not inflation. This is because financing the purchase of the pipe would be
done at the time of purchase, and we are assuming a fixed interest rate. The
uncertainty in this type of financing is assumed by the lending agency.
6. This method is not normally used for designing pipe sizes in laterals. For
one thing, it might recommend too many different sizes (inconvenient for
operation of periodic-move systems). Another reason is that we usually use
different criteria to design laterals (the “20%” rule on pressure variation).
7. Other factors could be included in the analysis. For example, there may be
certain taxes or tax credits that enter into the decision making process. In
general, the analysis procedure in determining pipe sizes can get as
complicated as you want it to – but higher complexity is better justified for
larger, more expensive irrigation systems.

IV. Other Pipe Sizing Methods
• Other methods used to size pipes include the following:
1. Unit head loss method: the designer specifies a limit on the

allowable head loss per unit length of pipe
2. Maximum velocity method: the designer specifies a maximum
average velocity of flow in the pipe (about 5 to 7 ft/s, or 1.5 to 2.0 m/s)
3. Percent head loss method: the designer sets the maximum pressure
variation in a section of the pipe, similar to the 20%Pa rule for lateral
pipe sizing
• It is often a good idea to apply more than one pipe selection method and
compare the results
• For example, don’t accept a recommendation from the economic selection
method if it will give you a flow velocity of more than about 10 ft/s (3 m/s),
otherwise you may have water hammer problems during operation
• However, it is usually advisable to at least apply the economic selection
method unless the energy costs are very low
• In many cases, the same pipe sizes will be selected, even when applying
different methods
• For a given average velocity, V, in a circular pipe, and discharge, Q, the

required inside pipe diameter is:
4Q
D= (89)
πV
• The following tables show maximum flow rates for specified average velocity
limits and different pipe inside diameters

Gallons per Minute Litres per Second
Velocity Limit Velocity Limit
D (inch) A (ft2) 5 fps 7 fps D (mm) A (m2) 1.5 m/s 2 m/s
0.5 0.00136 3.1 4.3 10 0.00008 0.1 0.2
0.75 0.00307 6.9 9.6 20 0.00031 0.5 0.6
1 0.00545 12.2 17.1 25 0.00049 0.7 1.0
1.25 0.00852 19.1 26.8 30 0.00071 1.1 1.4
1.5 0.01227 27.5 38.6 40 0.00126 1.9 2.5
2 0.02182 49.0 68.5 50 0.00196 2.9 3.9
3 0.04909 110 154 75 0.00442 6.6 8.8
4 0.08727 196 274 100 0.00785 11.8 15.7
5 0.13635 306 428 120 0.01131 17.0 22.6
6 0.19635 441 617 150 0.01767 26.5 35.3
8 0.34907 783 1,097 200 0.03142 47.1 62.8
10 0.54542 1,224 1,714 250 0.04909 73.6 98.2
12 0.78540 1,763 2,468 300 0.07069 106 141
15 1.22718 2,754 3,856 400 0.12566 188 251
18 1.76715 3,966 5,552 500 0.19635 295 393
20 2.18166 4,896 6,855 600 0.28274 424 565
25 3.40885 7,650 10,711 700 0.38485 577 770
30 4.90874 11,017 15,423 800 0.50265 754 1,005
40 8.72665 19,585 27,419 900 0.63617 954 1,272
50 13.63538 30,602 42,843 1000 0.78540 1,178 1,571
1100 0.95033 1,425 1,901
Cubic Feet per Second 1200 1.13097 1,696 2,262
Velocity Limit 1300 1.32732 1,991 2,655
D (ft) A (ft2) 5 fps 7 fps 1400 1.53938 2,309 3,079
1 0.785 3.93 5.50 1500 1.76715 2,651 3,534
2 3.142 15.71 21.99 1600 2.01062 3,016 4,021
3 7.069 35.34 49.48 1700 2.26980 3,405 4,540
4 12.566 62.83 87.96 1800 2.54469 3,817 5,089
5 19.635 98.17 137.44 1900 2.83529 4,253 5,671
6 28.274 141.37 197.92 2000 3.14159 4,712 6,283
7 38.485 192.42 269.39 2100 3.46361 5,195 6,927
8 50.265 251.33 351.86 2200 3.80133 5,702 7,603
9 63.617 318.09 445.32 2300 4.15476 6,232 8,310
10 78.540 392.70 549.78 2400 4.52389 6,786 9,048
11 95.033 475.17 665.23 2500 4.90874 7,363 9,817
12 113.097 565.49 791.68 2600 5.30929 7,964 10,619
13 132.732 663.66 929.13 2700 5.72555 8,588 11,451
14 153.938 769.69 1,077.57 2800 6.15752 9,236 12,315
15 176.715 883.57 1,237.00 2900 6.60520 9,908 13,210
16 201.062 1,005.31 1,407.43 3000 7.06858 10,603 14,137
17 226.980 1,134.90 1,588.86 3100 7.54768 11,322 15,095
18 254.469 1,272.35 1,781.28 3200 8.04248 12,064 16,085
19 283.529 1,417.64 1,984.70 3300 8.55299 12,829 17,106
20 314.159 1,570.80 2,199.11 3400 9.07920 13,619 18,158

Lecture 7
Set Sprinkler Lateral Design
I. Basic Design Criterion
• The basic design criterion is to size lateral pipes so that pressure variation
along the length of the lateral does not exceed 20% of the nominal design
pressure for the sprinklers
• This criterion is a compromise between cost of the lateral pipe and
application uniformity in the direction of the lateral
• Note that the locations of maximum and minimum pressure along a lateral
pipe can vary according to ground slope and friction loss gradient
II. Location of Average Pressure in the Lateral
• We are interested in the location of average pressure along a lateral pipe

because it is related to the design of the lateral
• Recall that friction head loss along a multiple-outlet pipe is nonlinear
• The figure below is for a lateral laid on level ground – pressure variation is
due to friction loss only...
• For equally-spaced outlets (sprinklers) and approximately thirty outlets (or

more), three-quarters of the pressure loss due to friction will occur between
the inlet and the location of average pressure
• The location of average pressure in the lateral is approximately 40% of the
lateral length, measured from the lateral inlet
• If there were only one outlet at the end of the lateral pipe, then one-half the
pressure loss due to friction would take place between the lateral inlet and
the location of average pressure, as shown below
• A computer program can be written to solve for the head loss in the lateral
pipe between each sprinkler
• Consider the following equations:
Total friction head loss:
(hf )total = ∑i=1(hf )i

n
(90)
Friction head loss to location of ha:
∑i=1∑ j=1(hf ) j
n i
(hf )a =
n +1 (91)
where n is the number of sprinklers; (hf)total is the total friction head loss from
0 to L; (hf)i is the friction head loss in the lateral pipe between sprinklers i-1
and i; and (hf)a is the friction loss from the lateral inlet to the location of ha
• As indicated above, (hf)a occurs over approximately the first 40% of the
lateral
• Note that between sprinklers, the friction head loss gradient is linear in the
lateral pipe
• Note also that (hf)0 = 0, but it is used in calculating (ha)f, so the denominator
is (n+1), not n

• In applying these equations with sample data, the following result can be
found:
( h f )a
≈ 0.73 (92)
(hf )total
• This supports the above claim that approximately ¾ of the friction head loss
occurs between the lateral inlet and the location of ha
• Also, from these calculations it can be seen that the location of ha is
approximately 38% of the lateral length, measured from the inlet, for laterals
with approximately 30 or more sprinklers
• But, this analysis assumes a constant qa, which is not quite correct unless
flow control nozzles and or pressure regulators are used at each sprinkler
• We could eliminate this assumption of constant qa, but it involves the solution
of a system of nonlinear equations
III. Location of Minimum Pressure in Laterals Running Downhill
• The location of minimum pressure in a lateral running downhill is where the

slope of the friction loss curve, J, equals the ground slope

• The above assertion is analogous to a pre-calculus “max-min problem”,
where you take the derivative of a function and set it equal to zero (zero
slope)
• Here we are doing the same thing, but the slope is not necessarily zero
Hazen-Williams Equation:
1.852
12 ⎛ Q ⎞
J = 1.21(10) ⎜ ⎟ D−4.87 (93)
⎝C⎠
for J in meters of friction head loss per 100 m (or ft/100 ft); Q in lps; and D in
mm
• In this equation we will let:
⎛q ⎞
Q = Ql − ⎜ a ⎟ x (94)
⎝ Se ⎠
for multiple, equally-spaced sprinkler outlets spaced at Se (m) from each
other, with constant discharge of qa (lps). Ql is the flow rate at the lateral
inlet (entrance).
• To find the location of minimum pressure, let J = S, where S is the ground
slope (in %, because J is per 100 m), which is negative for downhill-sloping
laterals
• Combining the two above equations and solving for x,
x=
Se ⎡
qa ⎣
(
Ql − 3(10)−7 C ( −S)0.54 D2.63 ⎤
⎦ ) (95)
where x is the distance, in m, from the lateral inlet to the minimum pressure
• S is in percent; Se and x are in m; D is in mm; and Ql and qa are in lps
• Note that the valid range of x is: 0 ≤ x ≤ L, and that you won’t necessarily get
J = S over this range of x values:
• If you get x < 0 then the minimum pressure is at the inlet

• If you get x > L then the minimum pressure is at the end
• This means that the above equation for x is valid for all ground slopes: S = 0,
S > 0 and S < 0
IV. Required Lateral Inlet Pressure Head
• Except for the most unusual circumstances (e.g. non-uniform downhill slope
that exactly matches the shape of the hf curve), the pressure will vary with
distance in a lateral pipe
• According to Keller & Bliesner’s design criterion, the required inlet pressure
head to a sprinkler lateral is that which makes the average pressure in the
lateral pipe equal to the required sprinkler pressure head, ha
• We can force the average pressure to be equal to the desired sprinkler

operating pressure by defining the lateral inlet pressure head as:
3 1
hl = ha + hf + ∆he (96)
4 2
• hl is the required pressure head at the lateral inlet
• Strictly speaking, we should take approximately 0.4∆he in the above
equation, but we are taking separate averages for the friction loss and
elevation gradients – and, this is a design equation
• Of course, instead of head, h, in the above equation, pressure, P, could be
used if desired

The value of ∆he is negative for laterals running
downhill
• For steep downhill slopes, where the minimum pressure would be at the
lateral inlet, it is best to let
hf = −∆he (97)
• Thus, we would want to consume, or “burn up”, excess pressure through

friction loss by using smaller pipes
• To achieve this equality for steep downhill slopes, it may be desirable to have
more than one pipe diameter in the lateral
• A downhill slope can be considered “steep” when (approximately)…
−∆he > 0.3ha (98)
• We now have an equation to calculate lateral inlet pressure based on ha, hf,
and he

• However, for large values of hf there will be correspondingly large values of hl
• Thus, for zero ground slope, to impose a limit on hf we will accept:
hf = 0.20ha (for S = 0 only) (99)
• This is the same as saying that we will not allow pipes that are too small, that
is, pipes that would produce a large hf value
• An additional head term must be added to the equation for hl to account for
the change in elevation from the lateral pipe to the sprinkler (riser height):
3 1
hl = ha + hf + ∆he + hr (100)
4 2
or, in terms of pressure…
3 1
Pl = Pa + Pf + ∆Pe + Pr (101)
4 2
V. Friction Losses in Pipes with Multiple Outlets
• Pipes with multiple outlets have decreasing flow rate with distance (in the
direction of flow), and this causes the friction loss to decrease by
approximately the square of the flow rate (for a constant pipe diameter)
• Sprinkler and trickle irrigation laterals fall into this hydraulic category
• Multiply the head loss for a constant discharge pipe by a factor “F” to reduce
the total head loss for a lateral pipe with multiple, equally spaced outlets:
JFL
hf = (102)
100
where F is from Eq. 8.9a
1 1 b −1
F= + + (103)
b + 1 2N 6N2
for equally spaced outlets, each with the same discharge, and going all the
way to the end of the pipe.
• All of the flow is assumed to leave through the outlets, with no “excess”
spilled out the downstream end of the pipe
• N is the total number of equally spaced outlets
• The value of b is the exponent on Q in the friction loss equation

• Darcy-Weisbach: b = 2.0
• Hazen-Williams: b = 1.852
• The first sprinkler is assumed to be located a distance of Se from the lateral

inlet
• Eq. 8.9b (see below) gives F(α), which is the F factor for initial outlet
spacings less than or equal to Se
NF − (1 − α)
F(α) = (104)
N − (1 − α)
where 0 < α ≤ 1
• Note that when α = 1, F(α) = F

• Many sprinkler systems have the first sprinkler at a distance of ½Se from the
lateral inlet (α = 0.5), when laterals run in both orthogonal directions from the
mainline
VI. Lateral Pipe Sizing for a Single Pipe Size
• If the minimum pressure is at the end of the lateral, which is the case for no
ground slope, uphill, and slight downhill slopes, then the change in pressure
head over the length of the lateral is:
∆h = h f + ∆h e (105)
If we allow ∆h = 0.20 ha, then
0.20ha = hf + ∆he (106)
JaFL
0.20ha − ∆he = (107)
100
and,
⎛ 0.20ha − ∆he ⎞
Ja = 100 ⎜ ⎟ (108)
⎝ FL ⎠
where Ja is the allowable friction loss gradient.
• Lateral pipe diameter can be selected such that J ≤ Ja

• The above is part of a standard lateral design criteria and will give a system
CU of approximately 0.97CU if lateral inlet pressures are the same for each
lateral position, for set sprinkler systems
• If the lateral is sloping downhill and the minimum pressure does not occur at
the end of the lateral, then we will attempt to consume the elevation gain in
friction loss as follows:
hf = −∆he (109)
⎛ −∆he ⎞
Ja = 100 ⎜ ⎟ (110)
⎝ FL ⎠
• Note that in this case ∆h ≠ hf + ∆he. Rather, ∆h = hmax - hmin, where:
1. hmax is either at the lateral inlet or at the end of the lateral, and
2. hmin is somewhere between the lateral inlet and the end
• Given a value of Ja, the inside diameter of the lateral pipe can be calculated
from the Hazen-Williams equation:
0.205
⎡ K ⎛ Q ⎞1.852 ⎤
D=⎢ ⎜ l⎟ ⎥ (111)
⎢⎣ Ja ⎝ C ⎠ ⎥⎦
where Ql is the flow rate at the lateral inlet (Nqa) and K is the units coefficient
in the Hazen-Williams equation
• The calculated value of D would normally be rounded up to the next available

internal pipe diameter
VII. Lateral Design Example
VI.1. Given information:
L= 396 m (lateral length)

qa = 0.315 lps (nominal sprinkler discharge)
Se = 12 m (sprinkler spacing)
hr = 1.0 m (riser height)
slope = -2.53% (going downhill)
Pa = 320 kPa (design nozzle pressure)
pipe material = aluminum

VI.2. Calculations leading to allowable pressure head loss in the lateral:
Nn = 396/12 = 33 sprinklers
F = 0.36
Ql = (0.315)(33) = 10.4 lps
∆he = SL = (-0.0253)(396) = -10.0 m
(Pf)a = 0.20Pa - ∆he = 0.20(320 kPa) – 9.81(-10.0 m) = 162 kPa

(hf)a = 162/9.81 = 16.5 m
VI.3. Calculations leading to required lateral pipe inside diameter:
0.3Pa = 0.3(320 kPa) = 96.0 kPa

0.3ha = 96.0/9.81 = 9.79 m
Now, 0.3ha < -∆he (steep downhill). Therefore, may want to use hf = -∆he.
Then, Ja is:
⎛ −∆he ⎞ ⎛ −( −10.0m) ⎞
Ja = 100 ⎜ ⎟ = 100 ⎜ ⎟ = 7.01m /100m (112)
⎝ FL ⎠ ⎝ (0.36)(396) ⎠
However, if 0.3ha > -∆he, Ja would be calculated as:
⎛ 0.20ha − ∆he ) ⎞ ⎛ 16.5 ⎞

Ja = 100 ⎜ ⎟ = 100 ⎜ ⎟ = 11.6m /100m (113)
⎝ FL ⎠ ⎝ (0.36)(396) ⎠
For now, let’s use Ja = 7.01 m/100 m. Then, the minimum pipe inside
diameter is (C ≈ 130 for aluminum):
0.205
⎡ 1.21E12 ⎛ 10.4 ⎞1.852 ⎤
D=⎢ ⎜ ⎟ ⎥ = 77.7mm (114)
⎢⎣ 7.01 ⎝ 130 ⎠ ⎥⎦
which is equal to 3.06 inches.
In the USA, 3” aluminum sprinkler pipe has an ID of 2.9” (73.7 mm), so

for this design it would be necessary to round up to a 4” nominal pipe size
(ID = 3.9”, or 99.1 mm).
However, it would be a good idea to also try the 3” size and see how the
lateral hydraulics turn out (this is done below; note also that for Ja = 11.6,
D = 70.0 mm).

VI.4. Check the design with the choices made thus far
The real friction loss will be:
1.852
⎛ 10.4 ⎞
J = 1.21E12 ⎜ ⎟ ( 99.1 mm)−4.87 = 2.14m /100m (115)
⎝ 130 ⎠
JFL (2.14)(0.36)(396)
hf = = = 3.06 m (116)
100 100
The required lateral inlet pressure head is:
hl = ha + 0.75hf + 0.5∆he + hr
(117)
hl = 320 / 9.81 + 0.75(3.06) + 0.5( −10.0) + 1.0 = 30.9 m
Thus, Pl is (30.9)(9.81) = 303 kPa, which is less than the specified Pa of

320 kPa, and this is because the lateral is running downhill
VI.5. Calculate the pressure and head at the end of the lateral pipe
hend = hl − hf − ∆he = 30.9 − 3.06 − (−10.0) = 37.8 m (118)
which is equal to 371 kPa. Thus, the pressure at the end of the lateral
pipe is greater than the pressure at the inlet.
To determine the pressure at the last sprinkler head, subtract the riser
height to get 37.8 m – 1.0 m = 36.8 m (361 kPa)
VI.6. Calculate the location of minimum pressure in the lateral pipe
x=
Se ⎡
qa ⎣
(
Ql − 3(10)−7 C( −S)0.54 D2.63 ⎤
⎦ )
(119)
x=
12 ⎡
0.315 ⎣
(
10.4 − 3(10)−7 130(2.53)0.54 (99.1)2.63 ⎤ = −39.6 m
⎦ )
The result is negative, indicating that that minimum pressure is really at
the entrance (inlet) to the lateral pipe. The minimum sprinkler head
pressure is equal to hl – hr = 30.9 – 1.0 = 29.9 m, or 293 kPa.

VI.7. Calculate the percent pressure variation along the lateral pipe
The maximum pressure is at the last sprinkler (end of the lateral), and the
minimum pressure is at the first sprinkler (lateral inlet). The percent
pressure variation is:
Pmax − Pmin 361 − 293

∆P = = = 0.21% (120)
Pa 320
That is, 21% pressure variation at the sprinklers, along the lateral
This is larger than the design value of 0.20, or 20% variation. But it is
very close to that design value, which is somewhat arbitrary anyway.
VI.8. Redo the calculations using a 3” lateral pipe instead of the 4” size
In this case, the location of the minimum pressure in the lateral pipe is:
x=
12 ⎡
0.315 ⎣
(
10.4 − 3(10)−7 130(2.53)0.54 (73.7)2.63 ⎤ = 196 m
⎦ ) (121)
which is the distance from the upstream end of the lateral.
There are about 196/12 = 16 sprinklers from the lateral inlet to the
location of minimum pressure, and about 17 sprinklers from x to the end
of the lateral.
Friction loss from x to the end of the lateral is:
1.852
⎛ (17)(0.315) ⎞
Jx −end = 1.21E12 ⎜ ⎟ ( 73.7 )−4.87 = 2.65 m /100 m
⎝ 130 ⎠ (122)
(2.65)(0.38)(396 − 196)
(hf )x −end = = 2.01 m (123)
100
Friction loss from the inlet to the end is:
1.852
⎛ 10.4 ⎞
Jinlet −end = 1.21E12 ⎜ ⎟ ( 73.7 )−4.87 = 9.05 m /100 m (124)
⎝ 130 ⎠

(9.05)(0.36)(396)
(hf )inlet −end = = 12.9 m (125)
100
Then, friction loss from inlet to x is:
(hf )inlet − x = 12.9 − 2.01 = 10.9 m (126)
The required lateral pipe inlet head is:
hl = ha + 0.75hf + 0.5∆he + hr
(127)
hl = 320 / 9.81 + 0.75(12.9) + 0.5(−10.0) + 1.0 = 38.3 m
giving a Pl of (38.3)(9.81) = 376 kPa, which is higher than Pl for the 4”

pipe
The minimum pressure head (at distance x = 196 m) is:
hx = hl − (hf )inlet − x − ( ∆he )inlet − x

(128)
hx = 38.3 − 10.9 − (−0.0253)(196) = 32.4 m
giving a Px of (32.4)(9.81) = 318 kPa, which is very near Pa.
The pressure head at the end of the lateral pipe is:
hend = hl − hf − ∆he = 38.3 − 12.9 + 10.0 = 35.4 m (129)
giving Pend of (35.4)(9.81) = 347 kPa, which is less than Pl. So, the
maximum lateral pipe pressure is at the inlet.
The percent variation in pressure at the sprinklers is based on Pmax = 376

-(1.0)(9.81) = 366 kPa, and Pmin = 318 - (1.0)(9.81) = 308 kPa:
Pmax − Pmin 366 − 308

% ∆P = = = 0.18 (130)
Pa 320
which turns out to be slightly less than the design value of 20%

VI.9. What if the lateral ran uphill at 2.53% slope?
In this case, the maximum allowable head loss gradient is:
⎛ 0.20ha − ∆he ) ⎞
Ja = 100 ⎜ ⎟
⎝ FL ⎠
(131)
⎛ 0.2(320 / 9.81) − 10.0 ⎞
= 100 ⎜ ⎟ = −2.44m /100m
⎝ (0.36)(396) ⎠
which is negative because ∆he > 0.2ha, meaning that it is not possible to
have only a 20% variation in pressure along the lateral, that is, unless
flow control nozzles and or other design changes are made.
VI.10. Some observations about this design example
Either the 3” or 4” aluminum pipe size could be used for this lateral
design. The 4” pipe will cost more than the 3” pipe, but the required
lateral inlet pressure is less with the 4” pipe, giving lower pumping costs,
assuming pumping is necessary.
Note that it was assumed that each sprinkler discharged 0.315 lps, when
in reality the discharge depends on the pressure at each sprinkler. To
take into account the variations in sprinkler discharge would require an
iterative approach to the mathematical solution (use a computer).
Most sprinkler laterals are laid on slopes less than 2.5%, in fact, most are
on fields with less than 1% slope.

Lecture 8
Set Sprinkler Lateral Design
I. Dual Pipe Size Laterals
• Sometimes it is useful to design a lateral pipe with two different diameters to

accomplish either of the following:
1. a reduction in hf
2. an increase in hf
• In either case, the basic objective is to reduce pressure variations along the
lateral pipe by arranging the friction loss curve so that it more closely
parallels the ground slope
• It is not normally desirable to have more than one pipe size in portable
laterals (hand-move, wheel lines), because it usually makes set changes
more troublesome
• For fixed systems with buried laterals, it may be all right to have more than
two pipe diameters along the laterals
• For dual pipe size laterals, approximately 5/8 of the pressure loss due to
friction occurs between the lateral inlet and the location of average pressure
• Case 1: a lateral on level ground where one pipe size is too small, but the
next larger size is too big...
• d1 is the larger diameter, and d2 is the smaller diameter

• note that (hf)single is much larger than (hf)dual

• Case 2: a lateral running downhill where one pipe size is too big, but the next
smaller size is too small...
• The composite friction loss curve for d1 and d2 more closely parallels the
ground slope than the curve with only d1, which means that the pressure
variation along the lateral is less with the dual pipe size design
II. Location of Average Pressure in Dual Size Laterals
• Do you believe that 5/8(hf)dual occurs between hl and ha?

• Consider the analysis shown graphically below (5/8 = 0.625)
• The plot is for a dual pipe size lateral with D1 = 15 cm, D2 = 12 cm, 100
equally-spaced outlets, 900 m total lateral length, Hazen-Williams C factor of
130, uniform sprinkler flow rate of 0.4 lps, and zero ground slope

• Notice where the ¾ value is on the left-hand ordinate
• Notice that the head loss from hl to ha is approximately 74% when the pipe size
is all D1 (1.0 on the abscissa) and when the pipe size is all D2 (0.0 on the
abscissa)
• The hl values (inlet pressure head) would be different for each point on the curve
if it were desired to maintain the same ha for different lateral designs
• Notice that the distance from the lateral inlet to the location of average pressure
head is roughly 40% of the total lateral length, but varies somewhat depending
on the ratio of lengths of D1 to D2 (in this example)
• These calculations can be set up on a spreadsheet to analyze any particular

combination of pipe sizes and other hydraulic conditions. Below is an example:
Section Flow Distance Diameter hf Sum (hf) d(he) head diff from hf/(hf)total
(lps) (m) (cm) (m) (m) (m) (m) ha (%)
1 40.00 9.00 15.00 0.31 0.31 0 49.69 12.08 0.016
2 39.60 18.00 15.00 0.31 0.62 0 49.38 11.77 0.032
3 39.20 27.00 15.00 0.30 0.92 0 49.08 11.47 0.048
4 38.80 36.00 15.00 0.29 1.21 0 48.79 11.18 0.064
5 38.40 45.00 15.00 0.29 1.50 0 48.50 10.89 0.079
6 38.00 54.00 15.00 0.28 1.79 0 48.21 10.61 0.094

III. Determining X1 and X2 in Dual Pipe Size Laterals
• The friction loss is:
⎛ J FL J F x ⎞ ⎛ J F x ⎞
hf = ⎜ 1 1 − 2 2 2 ⎟ + ⎜ 3 2 2 ⎟ (132)
⎝ 100 100 ⎠ ⎝ 100 ⎠
where,
hf = total lateral friction head loss for dual pipe sizes
J1 = friction loss gradient for D1 and Qinlet

J2 = friction loss gradient for D1 and Qinlet - (qa)(x1)/Se
J3 = friction loss gradient for D2 and Qinlet - (qa)(x1)/Se
F1 = multiple outlet reduction coefficient for L/Se outlets
F2 = multiple outlet reduction coefficient for x2/Se outlets
x1 = length of D1 pipe (larger size)
x2 = length of D2 pipe (smaller size)
x1 + x2 = L
• As in previous examples, we assume constant qa

• As for single pipe size laterals, we will fix hf by
∆h = hf + ∆he = 20%ha (133)

and,
hf = 20%ha − ∆he (134)
• Find d1 and d2 in tables (or by calculation) using Qinlet and...
( J)d1 ≤ Ja ≤ ( J)d2 (135)
for,
⎛ 20%ha − ∆he ⎞
Ja = 100 ⎜ ⎟
⎝ FL ⎠ (136)
• Now there are two adjacent pipe sizes: d1 and d2

• Solve for x1 and x2 by trial-and-error, or write a computer program, and make
hf = 0.20ha - ∆he (you already have an equation for hf above)

IV. Setting up a Computer Program to Determine X1 and X2
• If the Hazen-Williams equation is used, the two F values will be:
1 ⎛ 4 ⎞
F1 ≈ 0.351 + ⎜ 1 + ⎟ (137)
2N1 ⎝ 13N1 ⎠
1 ⎛ 4 ⎞
F2 ≈ 0.351 + ⎜1 + ⎟ (138)
2N2 ⎝ 13N2 ⎠
where
L
N1 = (139)
Se
L − x1
N2 = (140)
Se
• The three friction loss gradients are:
1.852
⎛Q ⎞
J1 = K ⎜ 1 ⎟ D1−4.87 (141)
⎝C⎠
1.852
⎛Q ⎞
J2 = K ⎜ 2 ⎟ D1−4.87 (142)
⎝ C ⎠
1.852
⎛Q ⎞
J3 = K ⎜ 2 ⎟ D2−4.87 (143)
⎝ C ⎠
where
⎛ L ⎞
Q1 = ⎜ ⎟ qa (144)
⎝ e⎠
S
⎛ L − x1 ⎞
Q2 = ⎜ ⎟ qa (145)
⎝ e ⎠
S
• The coefficient K in Eqs. 141-143 is 1,050 for gpm & inches; 16.42(10)6 for
lps and cm; or 1.217(10)12 for lps and mm
• Combine the above equations and set it equal to zero:
f ( x1 ) = α1 ⎡α 2 − α3 (L − x1 ) F2 ⎤ − 0.2ha + ∆he = 0
2.852
(146)
⎢⎣ ⎥⎦
where
K
α1 = (147)
100 C1.852
1.852
⎛q L⎞
α2 = ⎜ a ⎟ D1−4.87F1 L (148)
⎝ Se ⎠
1.852
α3 = ( D1−4.87 − D2−4.87 ) ⎛ qa ⎞
⎜ ⎟
⎝ Se ⎠
(149)
• The three alpha values are constants

• Eq. 146 can be solved for the unknown, x1, by the Newton-Raphson method
• To accomplish this, we need the derivative of Eq. 146 with respect to x1
df(x1)
=
dx1
(150)
⎡ S (L − x1)0.852 ⎛ 8Se ⎞ ⎤
= α1α3 ⎢ 2.852F2 (L − x1)1.852 − e ⎜ 1 + ⎟⎥
⎢⎣ 2 ⎝ 13(L − x1) ⎠ ⎥⎦
• Note that the solution may fail if the sizes D1 & D2 are inappropriate
• To make things more interesting, give the computer program a list of inside
pipe diameters so that it can find the most appropriate available values of D1
& D2
• Note that the Darcy-Weisbach equation could be used instead of Hazen-
Williams
• In Eq. 146 you could adjust the value of the 0.2 coefficient on ha to determine
its sensitivity to the pipe diameters and lengths
• The following screenshot is of a small computer program for calculating
diameters and lengths of dual pipe size sprinkler laterals

V. Inlet Pressure for Dual Pipe Size Laterals
5 1
hl = ha + hf + ∆he + hr
8 2 (151)
• This is the same as the lateral inlet pressure head equation for single pipe
size, except that the coefficient on hf is 5/8 instead of 3/4
• Remember that for a downhill slope, the respective pressure changes due to
friction loss and due to elevation change are opposing
VI. Laterals with Flow Control Devices
• Pressure regulating valves can be located at the base of each sprinkler:

These have approximately 2 to 5 psi (14 to 34 kPa) head loss
• Also, flow control nozzles (FCNs) can be installed in the sprinkler heads
• FCNs typically have negligible head loss
• For a lateral on level ground, the minimum pressure is at the end:

• The lateral inlet pressure head, hl, is determined such that the minimum
pressure in the lateral is enough to provide ha at each sprinkler...
hl = ha + hf + ∆he + hr + hcv (152)
where hcv is the pressure head loss through the flow control device
• For a lateral with flow control devices, the average pressure is not equal to
the nominal sprinkler pressure
havg ≠ ha
(153)
• If the pressure in the lateral is enough everywhere, then
2
⎛q ⎞
ha = ⎜ a ⎟
⎝ Kd ⎠ (154)
where ha is the pressure head at the sprinklers

• Below is a sketch of the hydraulics for a downhill lateral with flow control
devices

VII. Anti-Drain Valves
• Valves are available for preventing flow through sprinklers until a certain
minimum pressure is reached
• These valves are installed at the base of each sprinkler and are useful where
sprinkler irrigation is used to germinate seeds on medium or high value crops
• The valves help prevent seed bed damage due to low pressure streams of
water during startup and shutdown
• But, for periodic-move, the lines still must be drained before moving

Gravity-Fed Lateral Hydraulic Analysis
I. Description of the Problem
• A gravity-fed sprinkler lateral with evenly spaced outlets (sprinklers),

beginning at a distance Se from the inlet:
• The question is, for known inlet head, H0, pipe diameter, D, sprinkler spacing,
Se, ground slope, So, sprinkler discharge coefficient, Kd, riser height, hr, and
pipe material (C factor), what is the flow rate through each sprinkler?
• Knowing the answer will lead to predictions of application uniformity
• In this case, we won’t assume a constant qa at each sprinkler
II. Friction Loss in the Lateral
Hazen-Williams equation:
JL
hf = (155)
100
1.852
6⎛Q⎞
J = 16.42 (10 ) ⎜ ⎟ D−4.87
⎝C⎠ (156)
for Q in lps; D in cm; J in m/100 m; L in m; and hf in m.

• Between two sprinklers,
1.852
JSe ⎛Q⎞
= 16.42 (10 ) Se ⎜ ⎟ D−4.87
4
hf =
100 ⎝C⎠ (157)
or,
hf = hw Q1.852 (158)
where Q is the flow rate in the lateral pipe between two sprinklers, and
hw = 16.42 (10 ) SeC−1.852D−4.87

4
(159)
III. Sprinkler Discharge
typically,
q = Kd h (160)
where q is the sprinkler flow rate in lps; h is the pressure head at the
sprinkler in m; and Kd is an empirical coefficient: Kd = KoA, where A is the
cross sectional area of the inside of the pipe
IV. Develop the System of Equations
• Suppose there are only four sprinklers, evenly spaced (see the above figure)
• Suppose that we know H0, Kd, C, D, hr, So, and Se
(Q − Q )
2 2
⎛q ⎞
q1 = K d H1 − hr → H1 = hr + ⎜ 1 ⎟ = hr + 1 2 2 (161)
⎝ Kd ⎠ Kd
(Q − Q )
2 2
⎛q ⎞
q2 = K d H2 − hr → H2 = hr + ⎜ 2 ⎟ = hr + 2 2 3 (162)
⎝ Kd ⎠ Kd
(Q − Q )
2 2
⎛q ⎞
q3 = K d H3 − hr → H3 = hr + ⎜ 3 ⎟ = hr + 3 2 4 (163)
⎝ Kd ⎠ Kd
2
⎛ q4 ⎞ Q42
q4 = K d H4 − hr → H4 = hr + ⎜ ⎟ = hr + 2 (164)
⎝ Kd ⎠ Kd

• Pressure heads can also be defined independently in terms of friction loss
along the lateral pipe
H1 = H0 − hw Q11.852 − ∆he (165)
H2 = H1 − hw Q21.852 − ∆he (166)
H3 = H2 − hw Q31.852 − ∆he (167)
H4 = H3 − hw Q41.852 − ∆he (168)
where,
Se
∆he = (169)
So−2 +1
and So is the ground slope (m/m)
• The above presumes a uniform, constant ground slope

• Note that in the above equation, ∆he is always positive. So it is necessary to
multiply the result by –1 (change the sign) whenever So < 0.
• Note also that So < 0 means the lateral runs in the downhill direction
• Combining respective H equations:
( Q1 − Q2 ) 2 = H − hw Q11.852 − ∆he − hr (170)

0
K d2
( Q 2 − Q3 ) 2 = H 1.852
1 − h w Q2 − ∆he − hr (171)
K d2
( Q3 − Q 4 ) 2 = H − hw Q31.852 − ∆he − hr (172)

2
K d2
Q4 2
= H3 − hw Q41.852 − ∆he − hr (173)
K d2

• Setting the equations equal to zero:
f1 = H0 −
( Q1 − Q2 )
2
−h 1.852
− ∆he − hr = 0
w Q1 (174)
K 2d
f2 =
( Q1 − Q2 )
2
−
( Q2 − Q3 )
2
−h 1.852
− ∆he − hr = 0
w Q2 (175)
K 2d K 2d
f3 =
( Q2 − Q3 )
2
−
( Q3 − Q4 )
2
−h 1.852
− ∆he − hr = 0
w Q3 (176)
K 2d K 2d
f4 =
( Q3 − Q4 )
2
−
Q 42
−h 1.852
− ∆he − hr = 0
w Q4 (177)
K 2d K 2d
The system of equations can be put into matrix form as follows:
⎡ ∂f1 ∂f1 ⎤ ⎡ ⎤ ⎡ ⎤
⎢ ∂Q ∂Q2 ⎥ ⎢ δQ1 ⎥ ⎢ f1 ⎥
⎢ 1 ⎥ ⎢ ⎥ ⎢ ⎥
⎢ ∂f2 ∂f2 ∂f2 ⎥ ⎢ ⎥ ⎢ ⎥
⎢ ∂Q ∂Q2 ∂Q3 ⎥ ⎢ δQ2 ⎥ ⎢ f2 ⎥
⎢ 1 ⎥ ⎢ ⎥ =⎢ ⎥ (178)
⎢ ∂f3 ∂f3 ∂f3 ⎥ ⎢ ⎥ ⎢ ⎥
⎢ ⎥ ⎢ δQ3 ⎥ ⎢ f3 ⎥
⎢ ∂Q2 ∂Q3 ∂Q4 ⎥
⎢ ⎥ ⎢ ⎥
⎢ ∂f4 ∂f4 ⎥ ⎢ δQ ⎥ ⎢ f4 ⎥
⎢ ⎥ ⎢ 4
⎥ ⎢ ⎥
⎣ ∂Q3 ∂Q4 ⎦ ⎣ ⎦ ⎣ ⎦
where the two values in the first row of the square matrix are:
∂f1 −2 ( Q1 − Q2 )
= − 1.852hw Q10.852 (179)
∂Q1 2
Kd
∂f1 2 ( Q1 − Q2 )
= (180)
∂Q2 K 2d

The two values in the last row of the square matrix, for n sprinklers, are:
∂fn 2 ( Qn−1 − Qn )
= (181)
∂Qn−1 K 2d
∂fn −2Qn−1
= − 1.852hw Qn0.852 (182)
∂Qn 2
Kd
and the three values in each intermediate row of the matrix are:
∂fi 2 ( Qi−1 − Qi )
= (183)
∂Qi−1 K 2d
∂fi 2 ( Qi+1 − Qi−1 )

= − 1.852hw Qi0.852 (184)
∂Qi Kd2
∂fi 2 ( Qi − Qi+1 )
= (185)
∂Qi+1 K 2d
where i is the row number
• This is a system of nonlinear algebraic equations

• The square matrix is a Jacobian matrix; all blank values are zero
• Solve for Q1, Q2, Q3, and Q4 (or up to Qn, in general) using the Newton-
Raphson method, Gauss elimination, and backward substitution (or other
solution method for a linear set of equations)
• Knowing the flow rates, you can go back and directly calculate the pressure
heads one by one
• The problem could be further generalized by allowing for different pipe sizes
in the lateral, by including minor losses, by allowing variable elevation
changes between sprinkler positions, etc.
• However, it is still a problem of solving for x unknowns and x equations
• For pumped systems (not gravity, as above), we could include a
mathematical representation of the pump characteristic curve to determine
the lateral hydraulic performance; that is, don’t assume a constant H0, but
replace it by some function

V. Brute-Force Approach
• There is a computer program that will do the above calculations for a gravity-
fed lateral with multiple sprinklers
• But, if you want to write your own program in a simpler way, you can do the
calculations by “brute-force” as follows:
1. Guess the pressure at the end of the lateral

2. Calculate q for the last sprinkler
3. Calculate hf over the distance Se to the next sprinkler upstream
4. Calculate ∆he over the same Se
5. Get the pressure at that next sprinkler and calculate the
sprinkler flow rate
6. Keep moving upstream to the lateral inlet
7. If the head is more than the available head, reduce the end
pressure and start over, else increase the pressure and start
over
• Below is a screenshot of a computer program that will do the above

calculations for a gravity-fed lateral with multiple sprinklers

Lecture 9
Mainline Pipe Design
I. Split-Line Laterals
Laterals are usually distributed evenly along a mainline because:
• More equal pump load at different lateral positions

• Reduced mainline cost
• Don’t need to “dead-head” back when finished (cross over to other side)
• But, split-line laterals may interfere with cultural operations (wet areas at both
ends of field)
Consider twin laterals operating in the same direction:
• In the above case, and for only a single lateral on the mainline, the design of
the mainline is relatively simple – it is easy to find the most extreme
operating position
• However, the friction loss along the mainline is about four times greater than
for the split-line configuration
• Note that for the above two configurations the first sprinkler on the laterals
would be at 0.5Se from the inlet, unless the mainline is laid upon a roadway
in the field
Twin split-line laterals with dual mainline...
• Same combined mainline length

• No valves on mainline -- elbow at each lateral inlet
• More labor required, but mainline costs less because no valves, and because
the mainline is sized for the flow rate of one lateral over the entire mainline
length (not half at twice the capacity)
• For different lateral positions, you remove pieces of the mainline from the
longer section and put on the shorter section
II. General Design Considerations
• Look at extreme operating conditions for the mainline by varying lateral

positions (this can be more complicated for irregular field shapes and non-
uniform field slopes)
• Can use economic pipe selection method, but don’t make a big sacrifice in
terms of pressure uniformity along the mainline to save pumping costs
• Buried mainlines do not obstruct traffic, nor do they remove any land from
production. But, they cannot be moved from field to field
• Buried mainlines tend to last longer, because they are not handled and
banged up after installation
Uphill Split-Line Mainline Design
I. Definition of the Example Problem
• See example 10.1 from the textbook, an uphill mainline design for two split-
line laterals
• For design, consider the two extreme lateral positions:
1. Both laterals at position B (mid-point of mainline)

2. One lateral at position A and the other at position C
• Divide the mainline into two logical lengths, at the mid-point, according to the
two extreme lateral positions
• Determine the total allowable head loss due to friction in each of these logical
lengths, then find two adjacent pipe sizes for each length
• Determine the lengths of each pipe size so that the total head loss is just
equal to the allowable head loss
• This is somewhat analogous to the procedure for designing dual pipe size
laterals
• This is the system layout (shown with both laterals at position B):
• Pump provides at least 172 ft of head at P
• Lateral inlet pressure head is given as 125 ft of head (Eq. 9.2)
• Supply line and mainline are to be aluminum, in 30-ft lengths
• The figure below shows the hydraulic schematic for this mainline, with
separate friction loss profiles for the two extreme lateral positions
• The mainline is tentatively divided into sizes D1 and D2 for the first half (L1),
and D3 and D4 for the second half (L2). So, there are potentially four different
pipe sizes in the mainline from A to C.
• We do not yet know what these pipe diameters will be

• We do not yet know the required lengths of the different diameter pipes
• hf1 is for the case when both laterals are at B

• hf2, hf3, and hf4 are for the case when one lateral is at A and the other at C
• hf1 is for 500 gpm over L1

• hf2 is for 250 gpm over L1 + L2
II. Select the Size of the Supply Line
• We need to select the size of the supply line to know what the head loss is
from P to A (pressure at P is given as 172 ft of head)
• Assume no elevation change between P and A
• From continuity, Q = A V, then for an allowable velocity of 5 ft/s:
4Q 4 (1.11 cfs )
D= = = 0.53 ft (186)
πV π ( 5 fps )
• This is 6.4 inches. In Table 8.4, the 6-inch pipe has an inside diameter of
5.884 inches. With this size, the velocity at 500 gpm would be 5.9 ft/s, which
we will accept (could use 8-inch pipe, but 6-inch is probably OK)
• From Table 8.4, the head loss gradient in the 6-inch supply line at 500 gpm is
2.27 ft/100 ft. Then,
(hf )P− A = ( 2.27 ft /100 ft ) ⎛⎜

440 ft ⎞
⎟ = 10.0 ft (187)
⎝ 100 ⎠
• This means that the pressure head at A is 172 ft - 10.0 ft = 162 ft
III. Determine D1 and D2 for both Laterals at B
• We will tolerate (hf)1 head loss over section L1 of the mainline when both
laterals are operating at B. This will give the required hl at B.
• We can see that (hf)1 is defined as:
hf1 = 162 ft − hl + 0.5 ∆he (188)
hf1 = 162 ft − 125 ft − 7 ft = 30 ft (189)
• The allowable loss gradient in section L1 for both laterals operating at B is
⎛ 30 ft ⎞
( Ja )L1 = 100 ⎜ ⎟ = 5 ft per 100 ft (190)
⎝ 600 ft ⎠
• From Table 8.4, this is between the 5- and 6-inch pipe sizes, which have
respective loss gradients of 5.54 ft/100 ft and 2.27 ft/100 ft for the 500 gpm
flow rate. Therefore, choose D1 = 6 inch and D2 = 5 inch.
• Now we must find out how long D1 should be so that the friction loss is really
equal to 30 ft of head...
LD1 ( 2.27 ) + ( 600 − LD1 )( 5.54 ) = 100 ( 30 ft ) (191)
• Solving the above, LD1 = 99.1 ft

• Using 30-ft pipe lengths, we adjust the length to...
LD1 = 90 ft of 6" pipe ( 3 sections ) (192)

LD2 = 510 ft of 5" pipe (17 sections ) (193)
• With the adjusted lengths, we will get 30.3 ft of head loss over section L1 for
500 gpm (this is close enough to the allowable 30 ft)
IV. Determine D3 and D4 for Laterals at A and C
• We will tolerate (hf)3 + (hf)4 head loss over the whole length of the mainline
when one lateral is operating at A and the other at C
• We can calculate (hf)4 straight away because we already know the pipe sizes
and lengths in section L1...
hf 4 =
( 90 ft )( 0.63 ) + ( 510 ft )(1.53 ) = 8.37 ft (194)
100
where the friction loss gradients for 250 gpm are 0.63 ft/100 ft (6” size) and
1.53 ft/100 ft (5” size). These values were taken from Table 8.4.
• Now we need to know the allowable loss for (hf)3, such that the pressure in
the mainline at C will be equal to hl (we know that the pressure at A is 162 ft -
- it is more than enough)...
hf 2 = hf1 + 0.5 ∆he = 23.0 ft (195)

hf 3 = hf 2 − hf 4 = 23.0 ft − 8.37 ft = 14.6 ft (196)
• The allowable loss gradient in section L2 for laterals at A and C is
⎛ 14.6 ft ⎞
( Ja )L2 = 100 ⎜ ⎟ = 2.43 ft per 100 ft (197)
⎝ 600 ft ⎠
flow rate. Therefore, choose D3 = 5 inch and D4 = 4 inch.
• Now we must find out how long D3 should be so that the friction loss is really
equal to 14.6 ft of head...
LD3 (1.53 ) + ( 600 − LD3 )( 4.66 ) = 100 (14.6 ft ) (198)
• Solving the above, LD3 = 427 ft

LD3 = 420 ft of 5 " pipe (14 sections) (199)

LD4 = 180 ft of 4 " pipe (6 sections) (200)
• With the adjusted lengths, we will get 14.8 ft of head loss over section L2 for
250 gpm (this is close enough to the allowable 14.6 ft)
V. Check this Mainline Design for an Intermediate Position
• Just to be sure, suppose that one lateral is operating halfway between A and
B, and the other halfway between B and C
• The allowable friction loss from point A to the furthest lateral is (hf)2 + ¼∆he,
or 23.0 ft + 3.5 ft = 26.5 ft. The actual friction loss would be:
hf = 0.01[(2.27)(90) + (5.54)(210) + (1.53)(600)] = 22.9 ft (201)
• OK, the head in the mainline at the furthest lateral is more than enough
• See the figure below for a graphical interpretation of the two laterals in
intermediate positions
VI. Comments About the Mainline Design
• Both D2 and D3 are the same size in this example

• If we were lucky, both D1 and D2 (or D3 and D4) could be the same size, but
that means the friction loss gradient would have to be just right
• The lateral inlet pressure will be just right when both laterals operate at B
• The lateral inlet pressure will be just right for a lateral operating at C
• The lateral inlet pressure will always be too high for a lateral operating
between A and B (the inlet pressure to the mainline, at A, is always 162 ft)
• We designed D1 and D2 for the condition when both laterals are at B. This is
a more demanding condition for L1 than when one lateral is at A and the
other at C (in this case, only half the system flow rate is in L1). So, we don’t
need to “check” D1 and D2 again for the case when the laterals are at A and
C.
• We didn’t consider the hydrant loss from the mainline into the sprinkler
lateral, but this could be added to the requirements (say, effective hl)
• This design could be also done using the economic pipe selection method (or
another pipe selection method. It would be a good idea to check to see if the
172 ft at the pump (point A) could be reduced by using larger supply and
mainline pipes, thus reducing the annual energy costs. However, if the 172 ft
were due to gravity supply, the design would still be all right.
• However, the velocity in the 5-inch pipe at 500 gpm is too high, at 8.5 fps
(always check velocity limits when sizing pipes!)
VII. Both Laterals Operating at Point C
• How would the mainline design change if it were not split line operation, and
both laterals were operating at location C?
• In this case, intuition and past experience tells us location C is the critical
lateral position – if you don’t agree, then you should test other lateral
positions to convince yourself
• We will tolerate (hf)2 head loss over the entire 1,200-ft length of the mainline
when both laterals are operating at C. This will give the required hl at C.
• We can see that (hf)2 is defined as:
hf 2 = 162 ft − hl + ∆he (202)
hf 2 = 162 ft − 125 ft − 14 ft = 23 ft (203)
• The allowable loss gradient over the length of the mainline for both laterals
operating at C is
⎛ 23 ft ⎞
Ja = 100 ⎜ ⎟ = 1.92 ft per 100 ft (204)
⎝ 1,200 ft ⎠
flow rate
• Determine the respective pipe lengths so that the friction loss is really equal
to 23 ft of head...
L8 " ( 0.56 ) + (1,200 − L8 " )( 2.27 ) = 100 ( 23 ft ) (205)
• Solving the above, L8” = 248 ft

L8 " = 270 ft of 8" pipe ( 9 sections ) (206)

L6 " = 930 ft of 6" pipe ( 31 sections ) (207)
Lecture 10
Minor Losses & Pressure Requirements
I. Minor Losses
• Minor (or “fitting”, or “local”) hydraulic losses along pipes can often be
estimated as a function of the velocity head of the water within the particular
pipe section:
V2
hml = Kr (208)
2g
where hml is the minor loss (m or ft); V is the mean flow velocity, Q/A (m/s or
fps); g is the ratio of weight to mass (9.81 m/s2 or 32.2 ft/s2); and Kr is a
coefficient, dependent on the type of fitting (valve, bend, transition,
constriction, etc.)
• Minor losses include head losses through/past hydrants, couplers, valves,

pipe elbows, “tees” and other fittings (see Tables 11.1 and 11.2)
• For example, there is some loss when water flows through a hydrant, but
also some loss when water flows in a pipe past the location of a closed
hydrant
• Kr = 0.3 to 0.6 for flow in a pipeline going past a closed hydrant, whereby the
velocity in the pipeline is used to compute hml
• Kr = 0.4 to 0.8 for flow in a pipeline going past an open hydrant; again, the
velocity in the pipeline is used to compute hml
• Kr = 6.0 to 8.0 for flow from a pipeline through a completely open hydrant. In
this case, compute hml using the velocity of the flow through the lateral fitting
on the hydrant, not the flow in the source pipeline.
• For flow through a partially open hydrant, Kr increases beyond the 6.0 to 8.0
magnitude, and the flow rate decreases correspondingly (but not linearly)
• In using Tables 11.1 and 11.2 for hydrants, the nominal diameter (3, 4, 5,
and 6 inches) is the diameter of the hydrant and riser pipe, not the diameter
of the source pipeline
• Use the diameter of the hydrant for Kr and for computing Vr. However, for
line flow past a hydrant, use the velocity in the source pipeline, as indicated
above.
• Always use the largest velocity along the path which the water travels – this
may be either upstream or downstream of the fitting
• Do not consider velocities along paths through which the water does not flow
• In Table 11.2, for a sudden contraction, Kr should be defined as:
( )
2
Kr = 0.7 1 − Dr2 (209)
where Dr is the ratio of the small to large inside diameters (Dsmall/Dlarge)

• Allen (1991) proposed a regression equation for gradual contractions and
expansions using data from the Handbook of Hydraulics (Brater & King
1976):
( )
2
Kr = K f 1 − Dr2 (210)
where Kf is defined as:
K f = 0.7 − cos(f) ⎡⎣cos(f) ( 3.2cos(f) − 3.3 ) + 0.77 ⎤⎦ (211)
and f is the angle of the expansion or contraction in the pipe walls (degrees
or radians), where f ≥ 0
• For straight sides (no expansion or contraction), f = 0° (whereby Kf = 0.03)

• For an abrupt change in pipe diameter (no transition), f = 90° (whereby Kf =
0.7)
• The above regression equation for Kf gives approximate values for
approximate measured data, some of which has been disputed
• In any case, the true minor head loss depends on more than just the angle of
the transition
Expansion
Contraction
• For a sudden (abrupt) expansion, the head loss can also be approximated as
a function of the difference of the mean flow velocities upstream and
downstream:
hml =
( Vus − Vds )
2
(212)
2g
• An extreme (albeit unrealistic) case is for Vds = 0 and hml = Vus2/2g (total
conversion of velocity head)
• Various other equations (besides those given above) for estimating head loss
in pipe expansions and contractions have been proposed and used by
researchers and engineers
Minor Loss Example
• A mainline with an open lateral hydrant valve has a diameter of 200 mm ID

upstream of the hydrant, and 150 mm downstream of the hydrant
• The diameter of the hydrant opening to the lateral is 75 mm
• Qupstream = 70 lps and Qlateral = 16 lps
• The pressure in the mainline upstream of the hydrant is 300 kPa
• The mean flow velocities are:
0.070 m3 / s
V200 = = 2.23 m / s (213)
⎛ π(0.200 m)2 ⎞
⎜⎜ ⎟⎟
⎝ 4 ⎠
0.070 − 0.016 m3 / s
V150 = = 3.06 m / s (214)
⎛ π(0.150 m)2 ⎞
⎜⎜ ⎟⎟
⎝ 4 ⎠
0.016 m3 / s
Vhydrant = = 3.62 m / s (215)
⎛ π(0.075 m)2 ⎞
⎜⎜ ⎟⎟
⎝ 4 ⎠
• Note that V200 and V150 are both above the normal design limit of about 2 m/s
• The head loss past the open hydrant is based on the higher of the upstream
and downstream velocities, which in this example is 3.06 m/s
• From Table 11.1, the Kr for flow past the open hydrant (line flow; 6” mainline)
is 0.5; thus,
(3.06)2
(hml )past = 0.5 = 0.24m (216)
2(9.81)
• The head loss due to the contraction from 200 mm to 150 mm diameter (at
the hydrant) depends on the transition
• If it were an abrupt transition, then:
2
⎡ ⎛ 150 ⎞2 ⎤
Kr = 0.7 ⎢1 − ⎜ ⎟ ⎥ = 0.13 (217)
⎢⎣ ⎝ 200 ⎠ ⎥⎦
• And, if it were a 45° transition, Kf = 0.67, also giving a Kr of 0.13

• Then, the head loss is:
(3.06)2
(hml )contraction = 0.13 = 0.06m (218)
2(9.81)
• Thus, the total minor loss in the mainline in the vicinity of the open hydrant is
about 0.24 + 0.06 = 0.30 m (0.43 psi).
• The loss through the hydrant is determined by taking Kr = 8.0 (Table 11.1; 3”
hydrant):
(3.62)2
(hml )through = 8.0 = 5.3m (219)
2(9.81)
• This is a high loss through the hydrant (about 7.6 psi), so it may be advisable
to use a larger diameter hydrant.
• The pressure in the mainline downstream of the hydrant is (9.81 kPa/m):
⎛ V200
2 2 ⎞
− V150
P150 = P200 − γ (hml )past + γ⎜ ⎟
⎜ 2g ⎟
⎝ ⎠
(220)
⎛ (2.23) − (3.06)
2 2 ⎞
P150 = 300 − (9.81)(0.24) + 9.81⎜ = 295kPa
⎜ 2(9.81) ⎟⎟
⎝ ⎠
II. Total Dynamic Head
• The Total Dynamic Head (TDH) is the head that the pump “feels” or “sees”
while working, and is calculated to determine the pump requirements
• It includes the elevation that the water must be lifted from the source (not
necessarily from the pump elevation itself) to the outlet, the losses due to
“friction”, the pressure requirement at the outlet, and possibly the velocity
head in the pipeline
• For a sprinkler system, the value of TDH depends on the positions of the
laterals, so that it can change with each set. Pump selection is usually made
for the “critical” or extreme lateral positions, that is, for the “worst case
scenario”.
• Keller & Bliesner recommend the addition of a “miscellaneous” loss term,
equal to 20% of the sum of all “friction” losses. This accounts for:
1. Uncertainty in the Kr values (minor losses)

2. Uncertainty in the Hazen-Williams C values
3. Aging of pipes (increase in losses)
4. Wear of pump impellers and casings
• Losses in connectors or hoses from the mainline to laterals, if present, must

also be taken into account when determining the TDH
• See Example Calculation 11.2 in the textbook
• The next two lectures will provide more information about TDH and pumps
III. The System Curve
• The system curve determines the relationship between TDH and flow rate
• This curve is approximately parabolic, but can take more complex shapes
• Note that head losses in pipe systems are approximately proportional to the
square of the flow rate (Q2 or V2)
• For the Hazen-Williams equation, these losses are actually proportional to
Q1.852 or V1.852
• For standard, non-FCN, sprinkler nozzles, the head at the sprinkler is also
proportional to Q2
• Sprinkler systems can have a different system curve for each position of the
lateral(s)
• Defining the system curve, or the “critical” system curve, is important for
pump selection because it determines, in part, the operating point (TDH and
Q) for the system
IV. Valving a Pump
• A throttle valve may be required at a pump:
(a) Filling of the system’s pipes
• The head is low, and the flow rate is high

• Pump efficiency is low and power requirements may be higher
• Water hammer damage can result as the system fills
• Air vents and other appurtenances can be “blown off”
• For the above reasons, it is advisable to fill the system slowly
(b) To avoid cavitation, which damages the pump, pipes and

appurtenances
(c) To control the TDH as the sprinklers are moved to different sets
• Throttle valves can be automatic or manual
Pressure Requirements & Pumps
I. Types of Pumps
1. Positive Displacement
• Piston pumps
• Rotary (gear) pumps
• Extruding (flexible tube) pumps
2. Variable Displacement
• Centrifugal pumps
• Injector pumps
• Jet pumps
• The above lists of pump types are not exhaustive

• Positive displacement pumps have a discharge that is nearly independent of
the downstream (resistive) pressure. That is, they produce a flow rate that is
relatively independent of the total dynamic head, TDH
Positive Displacement Pumps
Axial-Flow Impeller
Closed Centrifugal Pump Impeller
Jet Pump
• But, with positive displacement pumps, the required pumping energy is a
linear function of the pressure
• Positive displacement pumps can be used with thick, viscous liquids. They
are not commonly used in irrigation and drainage, except for the injection of
chemicals into pipes and for sprayers
• Piston-type pumps can develop high heads at low flow rates
• Air injection, or jet pumps are typically used in some types of well drilling
operations. The air bubbles effectively reduce the liquid density and this
assists in bringing the drillings up out of the well. Needs a large capacity air
compressor.
• Homologous pumps are geometrically similar pumps, but of different sizes
II. Centrifugal Pumps
1. Volute Case This is the most common type of irrigation and drainage pump
(excluding deep well pumps). Produce relatively high flow rates at low
pressures.
2. Diffuser (Turbine) The most common type for deep wells. Designed to lift
water to high heads, typically using multiple identical “stages” in series,
stacked up on top of each other.
3. Mixed Flow Uses a combination of centrifugal and axial flow action. For high
capacity at low heads.
4. Axial Flow Water flows along the axis of impeller rotation, like a boat
propeller. Appropriate for high discharge under very low lift (head). An
example is the pumping plant on the west side of the Great Salt Lake.
5. Regenerative The characteristics of these pumps are those of a combination

of centrifugal and rotary, or gear, pumps. Shut-off head is well-defined, but
efficiency is relatively low. Not used in irrigation and drainage.
• In general, larger pumps have higher maximum efficiencies (they are more
expensive, and more effort is given toward making them more efficient)
• Impellers can be open, semi-open, or closed. Open impellers are usually
better at passing solids in the pumped liquid, but they are not as strong as
closed impellers
• Double suction inlet pumps take water in from both sides and can operate
without axial thrust
Closed Impeller Semi-Open Impeller Open Impeller
Characteristic Curve
• The pump “characteristic curve” defines the relationship between total

dynamic head, TDH, and discharge, Q
• The characteristic curve is unique for a given pump design, impeller
diameter, and pump speed
• The characteristic curve has nothing to do with the “system” in which the
pump operates
• The “shut-off” head is the TDH value when Q is zero (but the pump is still
operating)
• The shut-off head can exceed the recommended operating pressure, or even
the bursting pressure, especially with some thin-wall plastic pipes
Total Dynamic Head, TDH
Efficiency
Cha
ra cte
risi
cC
urv
Shut-Off Head
Power
0
0 Flow Rate, Q
III. Centrifugal Pumps in Parallel
• Pumps in PARALLEL

means that the total
flow is divided into two
or more pumps
• Typical installations are
On Two
for a single inlet pipe, e in P
Pu aral
branched into two lel
m
pumps, with the outlets
p
from the pumps
converging to a single
discharge pipe
• If only one of the
pumps operates, some 0
type of valve may be 0 Flow Rate, Q
required so that flow does not flow backwards through the idle pump
• Flow rate is additive in this case
pump 1
pump 2
Two Pumps in Parallel
IV. Centrifugal Pumps in Series
• Pumps in SERIES means that the total flow passes through each of two or
more pumps in line
• Typical installations are for increasing pressure, such as with a booster pump
• Head is additive in this case
Two
in Se
rie
s
One P
ump
0
0 Flow Rate, Q
• It is common for turbine (well) pumps to operate in series
• For centrifugal pumps, it is necessary to exercise caution when installing in
series because the efficiency can be adversely affected
• May need straightening vanes between pumps to reduce swirling
• Note that the downstream pump could cause negative pressure at the outlet
of the US pump, which can be a problem
pump 1 pump 2
Two Pumps in Series
Lecture 11
Pumps & System Curves
I. Pump Efficiency and Power
• Pump efficiency, Epump
water horsepower WHP

Epump = = (221)
brake horsepower BHP
where brake horsepower refers to the input power needed at the pump shaft
(not necessarily in “horsepower”; could be watts or some other unit)
• Pump efficiency is usually given by the pump manufacturer

• Typically use the above equation to calculate required BHP, knowing Epump
• Water horsepower is defined as:
QH
WHP = (222)
3956
where WHP is in horsepower; Q in gpm; and H in feet of head. The
denominator is derived from:
( 62.4 lbs/ft3 ) ( gal/min )( ft ) QH

γQH = ≈
( 33,000 ft-lbs/min-HP ) (7.481 gal/ft3 ) 3956
(223)
where γ = ρg, and ρ is water density. In metric units:
(1000 kg/m3 )(9.81 m/s2 ) (l/s )(m ) QH

WHP = ρgQH = =
( )
(224)
1000 l/m3
(1000 W/kW ) 102
where WHP is in kW; Q in lps; and H in meters of head
1 HP=0.746 kW (225)
• Total Dynamic Head, TDH, is defined as:
P V2
TDH = ∆Elev + hf + + (226)
γ 2g
where the pressure, P, and velocity, V, are measured at the pump outlet, and
hf is the total friction loss from the entrance to the exit, including minor losses
• At zero flow, with the pump running,
P
TDH = ∆Elev + (227)
γ
but recognizing that in some cases P/γ is zero for a zero flow rate
• The elevation change, ∆Elev, is positive for an increase in elevation (i.e.

lifting the water)
• Consider a turbine pump in a well:
Consider a centrifugal pump:
II. Example TDH & WHP Calculation
• Determine TDH and WHP for a centrifugal pump discharging into the air...
Head loss due to friction:
hf = hscreen + 3helbow + hpipe

(228)
for PVC, ε ≈ 1.5(10)-6 m, relative roughness is:
ε 1.5(10)−6
= = 0.0000051 (229)
D 0.295
Average velocity,
Q 4(0.102)
V= = = 1.49 m/s (230)
A π(0.295)2
Reynolds number, for 10°C water:
VD (1.49 m/s )( 0.295 m )

NR = = = 336,600 (231)
ν 1.306(10)−6 m2 / s
• From the Moody diagram, f = 0.0141

• From the Blasius equation, f = 0.0133
• From the Swamee-Jain equation, f = 0.0141 (same as Moody)
Using the value from Swamee-Jain,
⎛ 1,530 ⎞ (1.49 )
2
L V2
hpipe =f = 0.0141⎜ ⎟ = 8.27 m (232)
D 2g ⎝ 0.295 ⎠ 2 ( 9.81)
Kinematic Viscosity
Water Temperature (°C)
(m2/s)
0 0.000001785
5 0.000001519
10 0.000001306
15 0.000001139
20 0.000001003
25 0.000000893
30 0.000000800
40 0.000000658
50 0.000000553
60 0.000000474
The values in the above table can be closely approximated by:
( )
−1
ν = 83.9192 T 2 + 20,707.5 T + 551,173 (233)
where T is in ºC; and ν is in m2/s
From Table 11.2, for a 295-mm (12-inch) pipe and long radius 45-deg flanged
elbow, the Kr value is 0.15
V2 (1.49)2
helbow = Kr = (0.15) = (0.15)(0.11) = 0.017 m (234)
2g 2(9.81)
For the screen, assume a 0.2 m loss. Then, the total head loss is:
hf = 0.2 + 3(0.017) + 8.27 = 8.5 m (235)
With the velocity head of 0.11 m, the total dynamic head is:
TDH = 31 + 8.5 + 0.11 ≈ 40 m (236)
The water horsepower is:
QH (102 lps )( 40 m )
WHP = = = 40 kW (54 HP)
102 102 (237)
The required brake horsepower is:
WHP 40 kW
BHP = = ≈ 53 kW (71 HP)
Epump 0.76
(238)
• This BHP value would be used to select a motor for this application
• These calculations give us one point on the system curve (Q and TDH)
• In this simple case, there would be only one system curve:
System Curve
60
50
Total Dynamic Head (m)
40
30
20
10
0
0 20 40 60 80 100 120 140 160 180
Discharge (lps)
III. System Curves
• The “system curve” is a graphical representation of the relationship between

discharge and head loss in a system of pipes
• The system curve is completely independent of the pump characteristics
• The basic shape of the system curve is parabolic because the exponent on
the head loss equation (and on the velocity head term) is 2.0, or nearly 2.0
• The system curve will start at zero flow and zero head if there is no static lift,
otherwise the curve will be vertically offset from the zero head value
• Most sprinkle and trickle irrigation systems have more than one system curve
because either the sprinklers move between sets (periodic-move systems),
move continuously, or “stations” (blocks) of laterals are cycled on and off
• The intersection between the system and pump characteristic curves is the
operating point (Q and TDH)
• A few examples of system curves:
1. All Friction Loss and No Static Lift
2. Mostly Static Lift, Little Friction Loss
3. Negative Static Lift
4. Two Different Static Lifts in a Branching Pipe
5. Two Center Pivots in a Branching Pipe Layout
• The figure below shows two center pivots supplied by a single pump on a
river bank
• One of the pivots (#1) is at a higher elevation than the other, and is
further from the pump – it is the “critical” branch of the two-branch pipe
system
• Center pivot #2 will have excess pressure when the pressure is correct at
Center pivot #1, meaning it will need pressure regulation at the inlet to the
pivot lateral
• Use the critical branch (the path to Center pivot #1, in this case) when
calculating TDH for a given operating condition – Do Not Follow Both
Branches when calculating TDH
• if you cannot determine which is the critical branch by simple inspection,
you must test different branches by making calculations to determine
which is the critical one
• Note that the system curve will change with center pivot lateral position
when the topography is sloping and or uneven within the circle
• Of course, the system curve will also be different if only one of the center
pivots is operating
Center pivot #1
Center pivot #2
275 kPa
833 m
275 kPa
750 m
308 m river
pump
6. A Fixed Sprinkler System with Multiple Operating Laterals
• The next figure shows a group of laterals in parallel, attached to a

common mainline in a fixed sprinkler system
• All of the sprinklers operate at the same time (perhaps for frost control or
crop cooling purposes, among other possibilities)
• This is another example of a branching pipe system
• Since the mainline runs uphill, it is easy to determine by inspection that
the furthest lateral will be the critical branch in this system layout – use
this branch to determine the TDH for a given system flow rate
• Hydraulic calculations would be iterative because you must also
determine the flow rate to each of the laterals since the flow rate is
changing with distance along the mainline
• But in any case, Do Not Follow Multiple Branches when determining the
TDH for a given system flow rate
• Remember that TDH is the resistance “felt” by the pump for a given flow
rate and system configuration
lateral #5 (critical lateral)
lateral #4
lateral #3
mainline
uphill
lateral #2
lateral #1
pump
7. Two Flow Rates for Same Head on Pump Curve
• Consider the following graph

• “A” has a unique Q for each TDH value
• “B” has two flow rates for a given head, over a range of TDH values
• Pumps with a characteristic curve like “B” should usually be avoided
1 2
Unstable Stable
A
B
urv e
te mC
Sys
0
0 Flow Rate, Q
Affinity Laws and Cavitation
I. Affinity Laws
1. Pump operating speed:
2 3
Q1 N1 H1 ⎛ N1 ⎞ BHP1 ⎛ N1 ⎞
= =⎜ ⎟ =⎜ ⎟ (239)
Q2 N2 H2 ⎝ N2 ⎠ BHP2 ⎝ N2 ⎠
where Q is flow rate; N is pump speed (rpm); H is head; and BHP is “brake
horsepower”
• The first relationship involving Q is valid for most pumps

• The second and third relationships are valid for centrifugal, mixed-flow,
and axial-flow pumps
2. Impeller diameter:
2 3
Q1 D1 H1 ⎛ D1 ⎞ BHP1 ⎛ D1 ⎞
= =⎜ ⎟ =⎜ ⎟ (240)
Q2 D 2 H2 ⎝ D2 ⎠ BHP2 ⎝ D2 ⎠
• These three relationships are valid only for centrifugal pumps

• These relationships are not as accurate as those involving pump
operating speed, N (rpm)
Comments:
• The affinity laws are only valid within a certain range of speeds, impeller
diameters, flow rates, and heads
• The affinity laws are more accurate near the region of maximum pump
efficiency (which is where the pump should operate if it is selected correctly)
• It is more common to apply these laws to reduce the operating speed or to

reduce the impeller diameter (diameter is never increased)
• We typically use these affinity laws to fix the operating point by shifting the
pump characteristic curve so that it intersects the system curve at the
desired Q and TDH
II. Fixing the Operating Point
Combine the first two affinity law relationships to obtain:
2
H1 ⎛ Q1 ⎞
=⎜ ⎟ (241)
H2 ⎝ Q2 ⎠
• If this relationship is plotted with the pump characteristic curve and the
system curve, it is called the “equal efficiency curve”
• This is because there is typically only a small change in efficiency with a
small change in pump speed
• Note that the “equal efficiency curve” will pass through the origin (when Q is
zero, H is zero)
• Follow these steps to adjust the: (1) speed; or, (2) impeller diameter, such
that the actual operating point shifts up or down along the system curve:
1. Determine the head, H2, and discharge, Q2, at which the

system should operate (the desired operating point)
2. Solve the above equation for H1, and make a table of H1 versus
Q1 values (for fixed H2 and Q2):
2
⎛Q ⎞
H1 = H2 ⎜ 1 ⎟ (242)
⎝ Q2 ⎠
3. Plot the values from this table on the graph that already has the
pump characteristic curve
4. Locate the intersection between the pump characteristic curve
and the “equal efficiency curve”, and determine the Q3 and H3
values at this intersection
5. Use either of the following equations to determine the new
pump speed (or use equations involving D to determine the trim
on the impeller):
⎛Q ⎞ H2
Nnew = Nold ⎜ 2 ⎟ or, Nnew = Nold (243)
⎝ Q3 ⎠ H3
6. Now your actual operating point will be the desired operating

point (at least until the pump wears appreciably or other
physical changes occur)
• You cannot directly apply any of the affinity laws in this case because you will
either get the right discharge and wrong head, or the right head and wrong
discharge
Apply Affinity
Law from Here
e
rv
Cu
m
te
ys
3 S Operating Point
Head
Desired without Adjustment

Operating Point
2
correct head
Pu
incorrect discharge
rve
m
p
Cu
g!
C
ron
y
nc
ur
W
cie
ve
ffi
l E
ua
Eq
0
0 Flow Rate
III. Specific Speed
• The specific speed is a dimensionless index used to classify pumps

• It is also used in pump design calculations
Pump Type Specific Speed

Centrifugal (volute case) 500 - 5,000
Mixed Flow 4,000 - 10,000
Axial Flow 10,000 - 15,000
• To be truly dimensionless, it is written as:
2πN Q
Ns = (244)
(gH)0.75
where the 2π is to convert revolutions (dimensional) to radians

(dimensionless)
• Example: units could be N = rev/s; Q = m3/s; g = m/s2; and H = m

• However, in practice, units are often mixed, the 2π is not included, and even
g may be omitted
• This means that Ns must not only be given numerically, but the exact
definition must be specified
IV. Cavitation
• Air bubbles will form (the water boils) when the pressure in a pump or
pipeline drops below the vapor pressure
• If the pressure increases to above the vapor pressure downstream, the
bubbles will collapse
• This phenomenon is called “cavitation”
• Cavitation often occurs in pumps, hydroelectric turbines, pipe valves, and
ship propellers
• Cavitation is a problem because of the energy released when the bubbles
collapse; formation and subsequent collapse can take place in only a few
thousandths of a second, causing local pressures in excess of 150,000 psi,
and local speeds of over 1,000 kph
• The collapse of the bubbles has also been experimentally shown to emit
small flashes of light (“sonoluminescence”) upon implosion, followed by rapid
expansion on shock waves
• Potential problems:
1. noise and vibration

2. reduced efficiency in pumps
3. reduced flow rate and head in pumps
4. physical damage to impellers, volute case, piping, valves
• From a hydraulics perspective cavitation is to be avoided

• But, in some cases cavitation is desirable. For example,
1. acceleration of chemical reactions

2. mixing of chemicals and or liquids
3. ultrasonic cleaning
• Water can reach the boiling point by:
1. reduction in pressure (often due to an increase in velocity)

2. increase in temperature
• At sea level, water begins to boil at 100°C (212°F)

• But it can boil at lower temperatures if the pressure is less than that at mean
sea level (14.7 psi, or 10.34 m)
Patmospheric
Pvapor
container with water

• Pump inlets often have an eccentric reducer (to go from a larger pipe
diameter to the diameter required at the pump inlet:
1. Large suction pipe to reduce friction loss and increase NPSHa, especially
where NPSHa is already too close to NPSHr (e.g. high-elevation pump
installations where the atmospheric pressure head is relatively low)
2. Eccentric reducer to avoid accumulation of air bubbles at the top of the
pipe
• See the following figure…
Required NPSH
• Data from the manufacturer are available for most centrifugal pumps
• Usually included in this data are recommendations for required Net Positive
Suction Head, NPSHr
• NPSHr is the minimum pressure head at the entrance to the pump, such that
cavitation does not occur in the pump
• The value depends on the type of pump, its design, and size
• NPSHr also varies with the flow rate at which the pump operates
• NPSHr generally increases with increasing flow rate in a given pump
• This is because higher velocities occur within the pump, leading to lower
pressures
• Recall that according to the Bernoulli equation, pressure will tend to
decrease as the velocity increases, elevation being the same
• NPSHr is usually higher for larger pumps, meaning that cavitation can be
more of a problem in larger pump sizes
Available NPSH
• The available NPSH, or NPSHa, is equal to the atmospheric pressure minus

all losses in the suction piping (upstream side of the pump), vapor pressure,
velocity head in the suction pipe, and static lift
• When there is suction at the pump inlet (pump is operating, but not yet
primed), the only force available to raise the water is that of the atmospheric
pressure
• But, the suction is not perfect (pressure does not reduce to absolute zero in
the pump) and there are some losses in the suction piping
V2
NPSHa = hatm − hvapor − hf − hlift − (245)
2g
Vapor Pressure Head
Atmospheric Pressure Head

Friction Loss
Static Lift
Velocity Head
Available NPSH
• If the pump could create a “perfect vacuum” and there were no losses, the
water could be “sucked up” to a height of 10.34 m (at mean sea level)
• Average atmospheric pressure is a function of elevation above msl
perfect
vacuum
10.34 m
sea level
water
• 10.34 m is equal to 14.7 psi, or 34 ft of head

• Vapor pressure of water varies with temperature
11
10
9
Vapor Pressure Head (m)
8
7
6
5
4
3
2
1
0
0 10 20 30 40 50 60 70 80 90 100
Water Temperature (C)
• Herein, when we say “vapor pressure,” we mean “saturation vapor pressure”

• Saturation vapor pressure head (as in the above graph) can be calculated as
follows:
⎛ 17.27 T ⎞
hvapor = 0.0623exp ⎜ ⎟ (246)
⎝ T + 237.3 ⎠
for hvapor in m; and T in ºC
• Mean atmospheric pressure head is essentially a function of elevation above

mean sea level (msl)
• Two ways to estimate mean atmospheric pressure head as a function of
elevation:
Straight line:
hatm = 10.3 − 0.00105 z (247)
Exponential curve:
5.26
⎛ 293 − 0.0065 z ⎞
hatm = 10.33 ⎜ ⎟ (248)
⎝ 293 ⎠
where hatm is atmospheric pressure head (m of water); and z is elevation
above mean sea level (m)
10.50
10.00
Mean atmospheric pressure (m)
9.50
Straight Line (m)
Exponential Curve (m)
9.00
8.50
8.00
7.50
7.00
6.50
0 500 1000 1500 2000 2500 3000
Elevation above msl (m)
V. Example Calculation of NPSHa
1. Head Loss due to Friction
ε 0.2 mm
= = 0.000556 (249)
D 360 mm
viscosity at 20°C, ν = 1.003(10)-6 m2/s
flow velocity,
Q 0.100 m3 / s
V= = = 0.982 m/s (250)
π
A
( 0.36 ) 2
4
Reynold’s Number,
VD ( 0.982 )( 0.36 )
NR = = = 353,000 (251)
ν 1.003(10)−6
Darcy-Weisbach friction factor, f = 0.0184
velocity head,
V 2 (0.982)2
= = 0.049 m (252)
2g 2g
head loss in suction pipe,
L V2 ⎛ 8.1 ⎞
(hf )pipe =f = 0.0184 ⎜ ⎟ ( 0.049 ) = 0.0203 m (253)
D 2g ⎝ 0.36 ⎠
local losses, for the bell-shaped entrance, Kr = 0.04; for the 90-deg elbow, Kr =
0.14. Then,
(hf )local = ( 0.04+0.14 )(0.049 ) = 0.0088 m (254)
finally,
(hf )total = (hf )pipe + (hf )local = 0.0203 + 0.0088 = 0.0291 m (255)
2. Vapor Pressure
for water at 20°C, hvapor = 0.25 m
3. Atmospheric Pressure
at 257 m above msl, hatm = 10.1 m
4. Static Suction Lift
• the center of the pump is 3.0 m above the water surface

• (the suction lift would be negative if the pump were below the water
surface)
5. Available NPSH
V2
NPSHa = hatm − hvapor − (hf )total − hlift −
2g (256)
NPSHa = 10.1 − 0.25 − 0.0291 − 3.0 − 0.049 = 6.77 m
VI. Relationship Between NPSHr and NPSHa
• If NPSHr < NPSHa, there should be no cavitation

• If NPSHr = NPSHa, cavitation is impending
• As the available NPSH drops below the required value, cavitation will
become stronger, the pump efficiency will drop, and the flow rate will
decrease
• At some point, the pump would “break suction” and the flow rate would go to
zero (even with the pump still operating)
Lecture 12
Center Pivot Design & Operation
I. Introduction and General Comments
• Center pivots are used on about half of the sprinkler-irrigated land in the USA
• Center pivots are also found in many other countries
• Typical lateral length is 1,320 ft (400 m), or ¼ mile
• The lateral is often about 10 ft above the ground
• Typically, 120 ft pipe span per tower (range: 90 to 250 ft), often with one-
horsepower electric motors (geared down)
• At 120 ft per tower, a 1,320-ft lateral has about 10 towers; with 1-HP motors,
that comes to about 10 HP just for moving the pivot around in a circle
• The cost for a ¼-mile center pivot is typically about $55,000 (about $435/ac
or $1,100/ha), plus about $20,000 (or more) for a corner system
• For a ½-mile lateral, the cost may be about $75,000 (w/o corner system)
• In the state of Nebraska there are said to be 43,000 installed center pivots,
about 15% of which have corner systems
• Center pivots are easily (and commonly) automated, and can have much
lower labor costs than periodic-move sprinkler systems
• Center pivot maintenance costs can be high because it is a large and fairly
complex machine, operating under “field” conditions
• The typical maximum complete rotation is 20 hrs or so, but some (120-acre
pivots) can go around in only about 6 hrs
• IPS 6” lateral pipe is common (about 6-5/8 inches OD); lateral pipe is
generally 6 to 8 inches, but can be up to 10 inches for 2,640-ft laterals
• Long pivot laterals will usually have two different pipe sizes
• Typical lateral inflow rates are 45 - 65 lps (700 to 1,000 gpm)
• At 55 lps with a 6-inch pipe, the entrance velocity is a bit high at 3 m/s
• Typical lateral operating pressures are 140 - 500 kPa (20 to 70 psi)
• The end tower sets the rotation speed; micro switches & cables keep other
towers aligned
• Corner systems are expensive; can operate using buried cable; corner
systems don’t necessarily irrigate the whole corner
• Without a corner system or end gun, π/4 = 79% of the square area is
irrigated
• For a 1,320-ft lateral (without an end gun), the irrigated area is 125.66 acres
• For design purposes, usually ignore soil WHC (WaZ); but, refill root zone at
each irrigation (even if daily)
• Center pivots can operate on very undulating topography
• Some center pivots can be moved from field to field
• Below are some sample center pivot arrangements
• Some pivots have an end gun that turns on in the corners, in which all other
sprinklers shut off via individual solenoid-actuated valves. The pivot stops in
the corner while the end gun runs for a few minutes.
• Others just slow down in the corners, turning on an end gun, but leaving the
other sprinklers running (at lower discharges)
• Many farmers like extra capacity in the center pivot so they can shut off
during windy times of the day, and still complete the irrigations in time
• Corner systems have angle detectors so that sprinklers in the corner arm
turn on and off individually (or in groups) as the arm swings out and then
back in again
• Center pivots have safety switches to shut the whole thing off if any tower
gets too far out of alignment. Some also have safety switches to shut them
off if the temperatures gets below freezing (ice builds up and gets heavy,
possibly collapsing the structure). Some have safety switches connected to
timers: if a tower has not moved in a specified number of minutes, the
system shuts down. There may also be safety switches associated with the
chemical injection equipment at the lateral inlet location.
• Center pivots on rolling terrain almost always have pressure regulators at

each sprinkler
• Some engineers claim that center pivots can have up to about 90%
application efficiency
II. System Capacity
• The general center pivot design equation for system capacity is based on Eq.
5.4 from the textbook:
Ad R2d R2Udk f
Qs = K = = (257)
fT k1fT k1TEpa
where,
K is 2.78 for metric units and 453 for English units

k1 is (3,600 s/hr)/π = 1,146 for metric units; 30.6 for English units
kf is the peak period evaporation factor (Table 14.1 in the textbook)
A is area (ha or acre)
d is gross application depth (mm or inch)
f is frequency in days per irrigation
T is operating time (hrs/day)
R is the effective radius (m or ft)
Ud is the peak-use ET rate of the crop (mm/day or inch/day)
Qs is the system capacity (lps or gpm)
• The gross application depth, d, is equal to dn/Epa, where Epa is the design
application efficiency, based on uniformity and percent area (pa) adequately
irrigated
• The operating time, T, is generally 20-22 hrs/day during the peak-use period
• R is the effective radius, based on the wetted area from the center pivot
• The effective radius is about 400 m for many pivots
• R ≈ L + 0.4w, where L is the physical length of the lateral pipe, and w is the
wetted diameter of the end sprinkler
• This assumes that approximately 0.8 of the sprinkler radius beyond the
lateral pipe is effective for crop production
• Note that, for center pivots, Qs is proportional to Ud, and d and f are generally
not used, which is similar to drip irrigation design
III. Gross Application Depth
• If a center pivot is operated such that the water holding capacity of the soil is
essentially ignored, and water is applied frequently enough to satisfy peak-
use crop water requirements, then use dn/f = Ud, and
k f Ud k f Ud
d' = = (258)
Epa DEpaReOe
where d' is the gross application depth (mm/day or inches/day); and kf is a
peak-use period evaporation factor, which accounts for increased soil and
foliage evaporation due to high frequency (daily) irrigation
• When LR > 0.1, the LR can be factored into the equation as:
0.9k f Ud
d' = (259)
(1 − LR)DEpaR eOe
which is the same as Eq. 14.1b from the textbook, except that DEpa, Re and
Oe are all as fractions (not percent)
• Values of kf can be selected for the peak period from Table 14.1 of the
textbook for varying values of frequency, f
• Values for non-peak periods can be computed as described in the textbook
on page 314:
(100 − PT ') / PT '

k'f = ( k f − 1) + 1.0 (260)
(100 − PT) / PT
where kf and PT are for the peak-use period (Table 14.1), and k'f and PT' are
the frequency coefficient and transpiration percentage (PT) for the non-peak
period
T
PT = (261)
ET
• PT and PT' can be thought of as the basal crop coefficient (Kcb), or perhaps
Kcb - 0.1 (relative to alfalfa, as per the note in Table 14.1)
• It represents the transpiration of the crop relative to an alfalfa reference
IV. Water Application along the Pivot Lateral
• A major design difficulty with a center pivot is maintaining the application rate
so that it is less than the intake rate of the soil
• This is especially critical near the end of the lateral where application rates
are the highest
• As one moves along the center pivot lateral, the area irrigated by each unit
length of the lateral (each 1 ft or 1 m of length) at distance r from the pivot
point can be calculated as:
a = π(r + 0.5)2 − π(r − 0.5)2 = 2πr (262)
which is equal to the circumference at the radial distance r

• The portion of Qs (called q) which is applied to the unit strip at distance r is:
q a 2 πr 2r
= = = (263)
Qs A πR 2 R 2
or,
2rQs
q= (264)
R2
where q can be in units of lps per m, or gpm per ft
• This gives the amount of water which should be discharging from a specific
unit length of lateral at a radial distance r from the pivot point
• The q value at the end of the lateral (r = R) per ft or m is:
2Qs
qend = (265)
R
• Use q to select the nozzle size, where qnozzle = q Se
V. End-Gun Discharge
• This last equation is very similar to Eq. 14.20a, except for the omission of the
Sj term
• Equation 14.20b is for the end gun discharge, assuming that the end gun is
used primarily to compensate for the lack of pattern overlap at the end of the
lateral
• Equation 14.20b can be justified as follows:
• Assuming the “basic” circle discharge, Qb, includes the end gun discharge,
qg, we can write:
Qb qg
≈ (266)
πL2 ∆L ( 2πL' )
or, perhaps more precisely,
Qb qg
≈ (267)
πL2 ∆L ( 2π (L '+ ∆L / 2 ) )
but ∆L/2 is generally very small compared to L’, and this is ostensibly
assumed in Eq. 14.20b, so after solving the above for qg you will arrive at Eq.
14.20b:
2L' ∆L
qg ≅ Qb ; for ∆L < 0.03L (268)
L2
VI. Application Rate
• For a center pivot, Se = 1 (based on a unit distance along the lateral) and Sl =
w (wetted width in the tangential direction), so the average application rate
(called AR) at a distance r along the lateral is:
k3 2r QsReOe 2πr k f dReOe 2πr U'd ReOe

AR = 2
= = (269)
R w 60 f T w 60 T w
where AR is the average application rate over width w (mm/min or inch/min);

k3 is 1.61 for English units and 60 for metric units; and f is the time to
complete one revolution (days)
• w is equal to the wetted diameter of the spray or sprinkler nozzles on the

lateral
• U’d is the gross daily irrigation water requirement (mm/day or inch/day) and
includes the effect of kf
k f d k f (Ud − Pe )
U'd = = (270)
f DEpa
• The three forms of the above equation assume a rectangular application

pattern across the width w (that is, the application rate is uniform across w)
• Note that AR is proportional to r and is at a maximum at the end of the lateral
• Note that if w could be equal to 2πr, the application rate would be equal to
the gross application depth divided by the hours of operation per day (just
like a fixed or solid-set sprinkler system) – but this is never the case with a
center pivot machine
• At the end of the lateral (r = R), the average application rate can be
calculated as:
2πRU'd ReOe
ARr =R = (271)
60 T w
again, where a rectangular application pattern is assumed
VII. Application Rate with an Elliptical Pattern
• If the application pattern perpendicular to the lateral were elliptical in shape:
4 ⎛ 2k 3r QsReOe ⎞ 4 ⎛ 2 π r U'd R eOe ⎞ r U'd R eOe

AR x = ⎟ = π⎜ ⎟ = 7.5 T w
π ⎜⎝
(272)
2
R w ⎠ ⎝ 60 T w ⎠
where ARx is the maximum application rate (in the center of the pattern) (ARx
is in mm/min for U’d in mm/day)
• In the above equation, k3 is 1.61 for English units, or 60 for metric units
• It is usually a better approximation to assume an elliptical pattern under the
sprinklers than to assume a rectangular pattern, even though both are only
approximations
• For example, if U'd = 9 mm/day (which includes kf), T is 22 hrs/day, w is 30

m, R is 400 m, Re is 0.95 and Oe is 1.0, and the sprinkler application pattern
is elliptical, then the maximum application rate at the far end of the lateral is:
(400)(9)(0.95)(1.0)
AR x = = 0.69 mm / min (273)
(7.5)(22)(30)
• ARx is the peak AR (at the top of the ellipse, or directly beneath the lateral),
so an “average” (ARav) can be calculated, representing the average AR
beneath the wetted area perpendicular to the lateral pipe
• The calculated value of 0.69 mm/min is 41.4 mm/hr, which could be tolerated
only by a very sandy soil
• For a rectangular pattern, ARav = ARx
• For an elliptical pattern, ARav = (π/4)ARx
• Therefore, in the example, ARav = (π/4)(0.69) = 0.54 mm/min
• If d were 10 mm, it would take tt = 10/0.54 = 18 minutes to apply the water at
the rate ARav. (may want to use d Re Oe instead of just d in such a
calculation)
• Re can be taken from Fig. 6.8 or from examples in Table 14.3
• Guidelines for determining CI are given in Table 14.4
• The center pivot speed (at the end of the lateral) is w/tt, where tt is the time of
wetting
• In the preceding example, w is 30 m and tt is 18 min
• Therefore, the speed should be about 30 m/18 min = 1.7 m/min at the end
• Note that with spray booms, w is larger, and AR is smaller for the same q
value
Lecture 13
Center Pivot Nozzling
I. Center Pivot Nozzling
• The wetted width of the application package can be reduced closer to the
pivot point because the towers are moving at a slower speed at inner points;
therefore, the application intensity (AR) is less (qr ∝ r)
• Generally, if spray booms are required near the end of the center pivot, spray
drops can be used toward the center, and the spray drops nearest the pivot
point will produce something like a fine mist
• At the far end of the lateral the application may be more like a torrential rain
• Generally, impact and spray sprinklers would not be mixed on a center pivot
because the pressure requirements are substantially different
• The minimum wetted width at any radius r along the pivot (for an elliptical
pattern) can be calculated as:
8r Ud
wr = (274)
60 T AR x DEpa
where ARx is the maximum permissible application rate (mm/min) according

to limits imposed by the soil, slope, and vegetative cover; Ud is in mm/day; T
is in hrs/day; and DEpa is expressed as a fraction
• Note that wr approaches zero for r → 0

• A suitable application device can then be selected for radius r such that the
wetted diameter of the device, wd, is greater or equal to wr (wd ≥ wr)
• The actual application rate (mm/min) at radius r at the ground using a device
with wetted diameter wd should be:
r U'dReOe
ARr ≤ (275)
7.5 T w d
• The term “ReOe” is included in the above equation to account for evaporation
and wind drift losses, and pipe leakage
• Note that 60/8 = 7.5 and that we are using f = 1 day
• Divide by Re in the above equation to obtain AR at the nozzle
• The wetting time at any radius r (assuming an elliptical pattern) along the
lateral is:
4D f
( t t )r = (276)
π ARr
where Df is the total cumulative application (d Re Oe)
II. Center Pivot Nozzling Example
Given:
Ud = 8 mm/day
Pe = 0 mm/day
T = 22 hrs/day
Qs = 73.6 lps
R = 400 m
Speed = 21.6 hrs/revolution
ARx = 2.3 mm/min (allowable)
kf = 1.02
DEpa = 0.74
Re = 0.94
Oe = 0.99
Calculations:
• Calculate Ud’:
k f (Ud − Pe ) 1.02 ( 8 − 0 )
U'd = = = 11.0 mm/day (277)
DEpa 0.74
• Tangential speed of the pivot at distance “r” along the lateral:
2πr
Sr = (278)
t
where Sr is the speed in m/min; t is the minutes per full-circle revolution; and
r is the radius from the pivot point in m
• The wetting time is, then:
wr
τ= (279)
Sr
where τ is wetting time in minutes; and wr is the minimum wetted width in m.
• The values of wd can be selected from available boom lengths, which in this
case is 6, 8, 10, and 12 m. For less than 4-m width, no boom is required.
Select wd values such that ARx is not exceeded.
• Spreadsheet calculations:
r q wr wd ARr tau Qr
(m) (lps/m) (m) (m) (mm/min) (s) (lps)
0 0.0000 73.6
40 0.0368 1.14 3 0.829 928 72.9
80 0.0736 2.28 3 1.658 464 70.7
120 0.1104 3.42 6 1.244 619 67.0
160 0.1472 4.56 6 1.658 464 61.8
200 0.1840 5.70 6 2.073 371 55.2
240 0.2208 6.84 8 1.866 413 47.1
280 0.2576 7.98 8 2.177 354 37.5
320 0.2944 9.12 10 1.990 387 26.5
360 0.3312 10.26 10 2.239 344 14.0
400 0.3680 11.39 12 2.073 371 0.0
• Notes about the variables:
“q” is the sprinkler discharge per unit length of lateral at r

“wr” is the minimum wetted width at r
“wd” is the actual (selected) wetted width at r
“ARr” is the actual application rate at r
“wetting” is the wetting time at a given location
Qr is the flow rate in the lateral pipe at a given location
• Most manufacturers prefer to specify the actual nozzle sizing and spacing
along center pivots at the factory (rather than have the buyer specify these)
for reasons of liability (they have specialized computer programs which
attempt to maximize uniformity)
• Therefore, the designer will generally only specify the flow rate, pressure,
and type of nozzle (spray drop, booms, impacts, etc.), and the manufacturer
will specify individual nozzle sizes
• The following figure shows a center pivot with booms (the booms are greatly
exaggerated in width to show the concept)
III. Sprinkler/Nozzle Configurations
• Because the required discharge per unit of lateral is linearly proportional to

the distance along the lateral, discharge rates of sprinklers (or the sprinkler
spacing) need to be varied to match this linear variation in flow rate
• Various configurations of locating sprinklers along a center pivot lateral are
used
1. Uniform Sprinkler Spacing
• qa of nozzles increases with r

• Large heads are required near the end of the lateral
• A large w (due to large heads) results near the end which is good, but large
drop sizes may cause soil surface sealing and infiltration reduction
• The next figure shows a center pivot lateral with uniform sprinkler spacing
and increasing w values toward the downstream end of the lateral
2. Uniform Sprinkler Size where the distance between sprinklers decreases
with r
• Sprinkler density is higher near the downstream end of the lateral

• May allow the use of small nozzles with lower pressures and less soil
sealing
• The value of w is smaller, so the soil must support a higher application rate
• See the following figure
3. Combination of 1 and 2
• This is the most common configuration because it

is easier to match the required q/m of the lateral
• Smaller drop size
• Intermediate wetted width, w
• May have uniform spacing of outlets on lateral spans, but unused outlets
are plugged (w/o any sprinkler), so pipes in each span are identical and
interchangeable
Sizing Individual Nozzles:
1. Determine q for each spacing along the lateral (see Eq. 14.20a):
2Qs
qr = r Sr (280)
R2
where R is the maximum effective radius of the center pivot (approximately
equal to L + 0.4w); Sr is the sprinkler spacing at a distance r from the pivot
point; and r, Sr and R have the same units (m or ft)
2. Beginning at the design pressure at the end of the lateral, L (where q is
known), determine Pr:
Pr = (Pa )end + ( hfr )end + ( ∆He )end−r

(281)
where (hfr)end is the friction loss from point r to the far (downstream) end of
the lateral. Note that (∆He)end-r often averages out to zero as the pivot makes
its way around the circle, if the field slope is uniform (see the next figure).
3. Select the best nozzle size to provide qr at a pressure of Pr
4. Return to Step 1 and repeat for the next r-Sr location
5. The required pressure at the pivot point is Pr = 0
no slope
uniform slope
centered on a hill
centered on a depression
IV. Trajectory Angles of Impact Sprinklers
• For center pivots, sprinklers with 6° to 18° trajectory angles (low angle) are
preferred because drift losses are minimized (see Table 14.3 in the textbook)
• Other things being the same, wind drift and evaporation losses can be higher
with center pivots than with other types of sprinkler systems because of the
relative height of the sprinklers above the ground
• But, you can use drop-down sprayers on a “goose-neck” pipe – some of
these may be only a few centimeters from the mature crop canopy
V. End Guns
• The discharge for an end gun can be computed as (see Eq. 14.21):
⎛ 2Sr Q ⎛ Sr ⎞⎞
⎜ 2 ⎜ L + ⎟⎟
R ⎝ ⎠⎠
Qg = ⎝
2
(282)
0.93
where L is the lateral length; R is the effective length of the pivot (R = L + Sr);
Qg is the end gun discharge; Q is the total center pivot flow rate (includes
Qg); Sr is R - L, which equals the effective wetted radius (or 75% of the gun
radius)
• The above equation is essentially Eq. 14.20a with r = L + Sr/2

• The 0.93 factor (taken as Oe) is to account for ineffective discharge beyond
75% of the gun's wetted width, w (see page 349 of the textbook)
• Also, Qg can be approximated as:
⎛ L2 ⎞
Qg ≈ Q ⎜ 1 − 2 ⎟ (283)
⎜ R ⎟
⎝ ⎠
where Q includes Qg and R = L + Sr

• Note that you can derive this last equation by substituting R - L for Sr in the
previous equation as follows:
2 (R − L ) Q [L + 0.5(R − L)]
=
0.93R2
2QL (R − L ) + Q (R − L )
2
(284)
2
0.93 R
Q ⎛ L2 ⎞
= ⎜ 1 − ⎟
0.93 ⎜⎝ R2 ⎟⎠
where the only difference is the 0.93 in the denominator
• A part circle rotation (typically about 150°) is generally used to achieve best
uniformity under the area covered by the gun sprinkler, which is beyond the
end of the lateral pipe
• If the rotation of the end gun covered 180° or more, it might make it too
muddy for the wheels of the end tower – so with 150° (or so) the path in front
of the end tower stays relatively dry
Booster Pump
• If an end gun is used primarily on corners, and if

Qg > 0.2Q, then a booster pump in parallel or
series may be necessary
• Without a booster, when the gun is turned on (at
the corner), the pressure along the lateral will
drop, and individual sprinkler flow rates will be
approximately q[Q/(Q+Qg)]
• An alternative to a booster pump is to
automatically decrease the center pivot speed at the corners: Scorner =
S[Q/(Q+Qg)], where S is the speed in all places except the corners. That is,
turn the gun on in the corners without any booster pump. This can help
provide for equivalent, more nearly uniform application in all parts of the field
(friction losses will be greater when the gun is operating; therefore, uniformity
will not be perfect)
• If the end gun requires higher pressure than the nozzles along the center
pivot lateral, then an electric-motor-driven booster pump may be mounted on
the last tower, upstream of the end gun. This pump will increase the
pressure to the end gun only. All other nozzles will be operated at lower
pressure.
Center Pivot Hydraulic Analysis
I. Center Pivot Hydraulics
• The total discharge in the lateral pipe (not the flow rate from sprinklers at r) at
any point r is approximately:
⎛ r2 ⎞
Qr = Qs ⎜ 1 − 2 ⎟ (285)
⎜ R ⎟
⎝ ⎠
where Qs is total system capacity (possibly including an end gun); r is

measured radially from the pivot point; and R includes Sr
Friction Loss
(1) For a center pivot without an end gun:
1.852
⎛Q⎞
hf = khFpL ⎜ ⎟ D−4.87 (286)
⎝C⎠
where kh is 10.50 for hf and L in ft, Q in gpm, and D in inches; kh is
16.42(10)4 for hf and L in m, Q in lps, and D in cm
• Fp is the multiple outlet friction factor for a center pivot (see Fig. 14.12)
• Fp = 0.555 for a center pivot with a “large” number of outlets and no end gun
when using the Hazen-Williams equation.
• Other sources suggest using Fp = 0.543
• The value of C is about 130 for galvanized steel, or 145 for epoxy-coated
steel
(2) For a center pivot with an end gun:
Compute friction loss as though the center pivot were R m long rather than L,
and then subtract the non-existent friction past the point L, where R is the
effective (wetted) radius and L is the physical length of the lateral pipe.
A traditional way to consider the effects of an end gun on friction loss is:
1.852 1.852
⎛Q⎞ −4.87 ⎛ Qg ⎞
hf = khFpR ⎜ ⎟ D − khFg (R − L) ⎜ ⎟ D−4.87
⎝C⎠ ⎝ C ⎠ (287)
where Q is the total flow rate of the pivot plus the end gun; and Qg is the flow
rate of the end gun
• Fg should represent Fp for the distance R - L

• Base this on the bottom of page 355 of the textbook, where the number of
outlets is (R - L)/Se, where Se is the sprinkler spacing on the last
(downstream) span
• This Fg value will be conservative (it will underestimate the imaginary friction
loss, therefore, will overestimate the total friction in the pivot lateral), because
Fg will be greater than that stated
• Note that, near the end gun, the change in Q with distance is small; therefore,
the value of Fg will usually be near unity
• A different method for calculating friction loss is suggested in the textbook

(Eq. 14.26a)
• This may be a better method than that given above
Dual Pipe Sizes
• A center pivot may be assembled with dual pipe sizes (8- and 6-inch pipe, or
8- and 6 5/8-inch, for example)
• One requirement to note is that the tower spacing must be closer for the 8-
inch pipe due to the added weight of water in the 8-inch pipe, and traction
problems from towers which are too heavy and sink into the soil
• Therefore, balance the savings in hf with added cost for the system
• Tower spacing is often 100 ft (30 m) for 8-inch pipe, and 150 ft (45 m) for 6-
inch pipe
• Weight per tower = Wt of tower + Wt of 1 span (steel) + Wt of water in the
span
Pressure Balance Equation
Hl = Ha + hf + ∆He + Hr + Hmin or (288)
where Hl is the pressure head required at ground level, at the pivot point; Ha is
the pressure head requirement of the last nozzle (or end gun); hf is the total
friction loss along the pivot lateral; ∆He is the elevation increase between the
pivot point and lateral end; Hr is the height of the lateral pipe less the vertical
length of any drop tubes; and Hminor is the sum of all minor losses along the
lateral
• Ha is, then, the pressure head requirement of the last nozzle

• If end guns require higher pressure than the nozzles along the lateral, then
an electric booster pump can be installed at the last tower on the pivot
• See Chapter 14 of the textbook for the hydraulic effect of turning end guns or
corner systems on and off, and for additional elevation effects
II. An Alternative Method for Determining Hf and Hl
• An alternative method to compute the exact value of Hl is to start at the pivot

end and progress toward the pivot, adding friction losses
• This is a "stepwise" computational procedure, and is generally more accurate
than using the JFpL equations
• Friction loss, hf, in pipe segments between nozzles on the pivot can be
computed using the Hazen-Williams or Darcy-Weisbach equations with Fp =1
• In this case, include ∆He if Hend is the minimum pressure head at the highest
elevation position of the lateral
n
( )
Hl = Hend + ∑ ⎡ hfi + ∆Hei + Hr + Hmin or ⎤
⎣ ⎦
i=0 (289)
where Hend is the desired nozzle pressure head at the pivot end; i is the outlet
number along the lateral (i = 0 at the end, and i = n at the pivot point); n is
the number of outlets (sprinklers) on the lateral; ∆Hei is the elevation
difference between two adjacent points (i and i+1) along the lateral; and hfi is
the friction loss between outlet i and i+1, where i+1 is upstream of i. This last
term is defined as:
Ji ∆Li
hfi = (290)
100
where,
1.852
12 ⎛ Qi
⎞
Ji = 1.21(10) ⎜ ⎟ Di−4.87 (291)
⎝C⎠
with Qi in lps and Di in mm
and,
i
Qi = ∑ q j (292)
j= 0
in which j is the outlet number (j = 0 at the downstream end of the lateral);

and qj is a function of hj and Kd
• Hminor includes short hose connections between pipe segments (at towers)
• Therefore, actual computed hj values should be used with the selected
nominal nozzle size (or FCN size), where hj is the pressure head at outlet j
• The desired qj is:
⎛ 2Q ⎞ ⎛ 2Q ⎞
q j = rjSrj ⎜ 2 ⎟ = rj∆Li ⎜ 2 ⎟ (293)
⎝R ⎠ ⎝R ⎠
where ∆Li is the distance between outlet i and outlet i+1.
• ∆Li is constant for constant spacing, variable nozzle size

• ∆Li is variable for variable spacing, constant nozzle size
Point of “Average” Elevation
The point of “average” elevation along a lateral may be determined by weighting

elevations according to areas as (Allen 1991):
L 2π r (He ) r,α
Hew = ∫ ∫ 2πL2
dr dα (294)
r =0 α
where r is a weighting term; and (He)r,α is the elevation at radius r and pivot
rotation angle α.
• Use α = 0 for the whole irrigated area

• Note that on uniform slopes, the weighted elevation, Hew, equals (He)pivot
• Note also that you probably won’t have data for elevations in polar
coordinates, so this equation may be rather “academic”
• See the textbook for additional equations which consider elevation effects
Lecture 14
Center Pivot Uniformity Evaluation
I. Introduction
• The calculation of an application uniformity term must take into account the
irrigated area represented by each catch container
• It is more important to have better application uniformity further from the pivot
point than nearer, because the catch containers at larger distances represent
larger irrigated areas
• If the catch containers are equally spaced in the radial direction, the area
represented by each is directly proportional to the radial distance
II. Equation for Center Pivot CU
• The equation for CU proposed by Heermann and Hein is (ASAE/ANSI S436):
⎛ ⎛ ⎞⎞
∑ i=1( diri ) ⎟⎟ ⎟⎟
n
⎜ n ⎜
⎜ ∑ r d −
i=1⎜ i i
∑i=1ri ⎟⎠ ⎟⎟
n
CU = 100 ⎜ ⎜ (295)
⎜ 1.0 − ⎝
⎜ ⎟
∑ i=1( diri )
n
⎜ ⎟
⎝ ⎠
where CU is the coefficient of uniformity; di is the depth from an individual

container; ri is the radial distance from the pivot point; and n is the number of
containers
• First calculate the summations:
∑i=1ri ∑i=1( dri i )

n n
and, (296)
• Then, perform the outer summation to determine the CU value

• That is, don’t recalculate the inner summation values for every iteration of the
outer summation – it isn’t necessary
• It is usually considered that a center pivot CU should be greater than 85%
• If the radial distances, ri, are equal, the sequence number of the can
(increasing with increasing radius) can be used instead of the actual distance
for the purpose of calculating application uniformity
• Consider the following two figures:
slope?
catch containers
leg # #1
3 g
le
slope?
leg #
2
catch containers
level field?
no wind?...
CU = 100%
III. Standard Uniformity Values
• You can also calculate the “standard” CU or DU if you weight each catch
value by multiplying it by the corresponding radial distance
• To obtain the low ¼, rank the unweighted catches, then start summing radii
(beginning with the radius for the lowest catch value) until the cumulative
value is approximately equal to ¼ of the total cumulative radius
• This may or may not be equal to ¼ of the total catch values, because each
catch represents a different annular area of the field
• Finally, divide the sum of the catches times the radii for this approximately ¼
area by the cumulative radius
• This gives the average catch of the low ¼
• Don’t rank the weighted catches (depth x radius) because you will mostly get
the values from the low r values (unless the inner catches are relatively high
for some reason), and your answer will be wrong
• Don’t calculate the average of the low ¼ like this…
(because the lowest ¼ of the catches generally represents
something different than ¼ of the irrigated area):
• Actually, the equation at the right is all right, except for the
value “n/4”, which is probably the wrong number of ranked
values to use in representing the low ¼
• You can set up a table like this in a spreadsheet application:
Ranked Center Pivot Catches

Radius, r Cumulative r Depth, d d*r Cumulative d*r
smallest
largest
Totals: ---- ----
• Note that when you rank the depths, the radius values should stay with the
same depth values (so that the radius values will now be “unranked”; all
mixed up)
• To get the average weighted depth for the whole pivot area, divide the total
“Cumulative d*r” by the total “Cumulative r” (column 5 divided by column 2)
• Find the row corresponding closest to ¼ of the total “Cumulative r” value, and
take the same ratio as before to get the weighted average of the low ¼ area
• Look at the example data analysis below:
Ranked catches
Radius, r Cum. r Depth, d d*r Cum. d*r
120 120 0.52 62.6 62.6
900 1,020 0.95 851.9 915
160 1,180 1.29 205.8 1,120
340 1,520 1.31 445.6 1,566
1000 2,520 1.46 1,456.3 3,022
1040 3,560 1.46 1,514.6 4,537
240 3,800 1.48 355.3 4,892
800 4,600 1.50 1,203.9 6,096
860 5,460 1.53 1,315.0 7,411
480 5,940 1.58 757.3 8,168
1280 7,220 1.58 2,019.4 10,188
980 8,200 1.60 1,569.9 11,758
540 8,740 1.63 878.2 12,636
360 9,100 1.65 594.2 13,230
460 9,560 1.67 770.4 14,000
880 10,440 1.70 1,495.1 15,496
320 10,760 1.72 551.5 16,047 1/4 area (11,055)
1140 11,900 1.75 1,992.2 18,039
1160 13,060 1.75 2,027.2 20,067
280 13,340 1.82 509.7 20,576
720 14,060 1.82 1,310.7 21,887
1300 15,360 1.82 2,366.5 24,253
200 15,560 1.84 368.9 24,622
420 15,980 1.84 774.8 25,397
440 16,420 1.89 833.0 26,230
1020 17,440 1.89 1,931.1 28,161
1200 18,640 1.92 2,301.0 30,462
600 19,240 1.94 1,165.0 31,627
640 19,880 1.94 1,242.7 32,870
1060 20,940 1.94 2,058.3 34,928
1100 22,040 1.94 2,135.9 37,064
220 22,260 1.97 432.5 37,497 1/2 area (22,110)
1080 23,340 1.97 2,123.3 39,620
380 23,720 1.99 756.3 40,376
740 24,460 1.99 1,472.8 41,849
920 25,380 1.99 1,831.1 43,680
1220 26,600 1.99 2,428.2 46,108
300 26,900 2.01 604.4 46,713
180 27,080 2.03 364.8 47,077
820 27,900 2.04 1,671.8 48,749
1260 29,160 2.04 2,568.9 51,318
660 29,820 2.06 1,361.7 52,680
1180 31,000 2.06 2,434.5 55,114
680 31,680 2.09 1,419.4 56,534
940 32,620 2.11 1,985.0 58,519
560 33,180 2.14 1,196.1 59,715
260 33,440 2.16 561.7 60,276
1120 34,560 2.18 2,446.6 62,723
700 35,260 2.23 1,563.1 64,286
760 36,020 2.23 1,697.1 65,983
100 36,120 2.25 224.5 66,208
960 37,080 2.26 2,167.0 68,375
520 37,600 2.28 1,186.4 69,561
620 38,220 2.28 1,414.6 70,976
1240 39,460 2.28 2,829.1 73,805
500 39,960 2.33 1,165.0 74,970
140 40,100 2.35 329.6 75,300
400 40,500 2.40 961.2 76,261
780 41,280 2.52 1,968.9 78,230
40 41,320 2.57 102.9 78,333
80 41,400 2.57 205.8 78,538
840 42,240 2.79 2,344.7 80,883
580 42,820 2.82 1,633.0 82,516
60 42,880 3.23 193.7 82,710
1320 44,200 3.79 4,998.1 87,708
20 44,220 3.83 76.7 87,784
• Notice that the depth values (3rd column) are ranked from low to high
• Notice that the maximum value of cumulative r is 44,220 & maximum
cumulative d*r is 87,784. Then, the weighted average depth for the entire
center pivot is equal to 87,784/44,220 = 1.985 (whatever units)
• One quarter of 44,220 is equal to 11,055 which corresponds most closely to
the row in the table with depth = 1.72. For the same row, divide the two
cumulative columns (Col 5/Col 2) to get 16,047/10,760 = 1.491, which is
approximately the average of the low ¼.
• Finally, estimate the distribution uniformity for this data set as:
⎛ 1.491 ⎞
DU ≅ 100 ⎜ ⎟ ≅ 75% (297)
⎝ 1.985 ⎠
• Note that in this example, the average of the low ¼ was, in fact, based on
approximately the first n/4 ranked values
• Consider the weighed catch-can data plotted below:
14000
12000
Weighted Catch Values
10000
8000
6000
4000
2000
0
0 10 20 30 40 50 60 70
Distance from Pivot Point
• As in any application uniformity evaluation, there is no “right” answer. The

results are useful in a comparative sense with evaluations of other center
pivots and other on-farm irrigation systems.
• However, a plot of the catches can give indications of localized problems
along the center pivot radius
IV. The Field Work
• It may take a long time for the full catch in containers near the pivot point,
and because these represent relatively small areas compared to the total
irrigated area, it is usually acceptable to ignore the inside 10% or 20% of the
radius
• The pivot quickly passes the outer cans, but takes longer to completely pass
the inner cans, so you can collect the data from the outer cans sooner
• The pivot should not be moving so fast that the application depth is less than
about 15 mm
• Catch containers can be placed beyond the physical length of the lateral
pipe, but if they are so far out that the catches are very low, these can be
omitted from the uniformity calculations
• Catch containers should be spaced in the radial direction no further than
about 30% of the average wetted diameter of the sprinklers
• There is often an access road leading to the pivot point for inspection,
manual operation, maintenance, and other reasons
• If the crop is dense and fairly tall (e.g. wheat or maize) it will be difficult to
perform the evaluation unless the cans are placed on the access road
• Otherwise, you can wait until the crop is harvested, or do the test when the
crop is still small
• Some people recommend two radial rows of catch cans, or even two parallel
rows, to help smooth out the effects of the non continuous movement of
towers (they start and stop frequently to keep the pivot lateral in alignment)
• Some have used troughs instead of catch cans to help ameliorate this
problem.
• Note that if the field is sloping or undulating, the results from one radial row
of catch cans may be quite different from those of a row on another part of
the irrigated circle
• See Merriam and Keller (1978)
Linear Move Systems
I. Introduction
• Mechanically, a linear move system is essentially the same as a center pivot

lateral, but it moves sideways along a rectangular field, perpendicular to the
alignment of the lateral pipe
• The variation of flow rate in a linear move lateral is directly proportional to
distance along the lateral pipe, whereas with center pivots it is proportional to
a function of the square of the distance from the pivot point
• A center or end tower sets the forward speed of the machine, and the other
towers just move to keep in line with the guide tower (this is like the far end
tower on a center pivot)
• Usually, each tower is independently guided by cables and micro-switches as
for a center pivot – this keeps the lateral pipe in a straight line (aligned with
itself)
• Alignment with the field is usually not mechanically “enforced”, but it is
“monitored” through switches in contact with a straight cable along the center
of the field, or along one end of the field
• The center tower has two "fingers", one on each side of the cable, usually
slightly offset in the direction of travel (they aren’t side by side). The fingers
should be in constant contact with the cable – if one is lifted too far a switch
will be tripped, shutting the system down (because the whole lateral is
probably getting out of alignment with the field)
• If the cable is broken for any reason, this should also shut the system down
because the fingers will lose physical contact
• If the lateral gets out of alignment with the field and shuts off, it will be
necessary to back up one side and or move the other side forward until it is
in the correct position
• This can involve electrical “jumps” between contacts in the control box, but
some manufacturers and some installers put manual switches in just for this
purpose
• Some linear moves are fitted with spray nozzles on drop tubes or booms
• If they are spaced closely along the lateral, it may be necessary to put booms
out beyond the wheels at tower locations, either in back of the lateral or on
both sides of the lateral
Water Supply
• Water is usually supplied to the lateral via:
1. a concrete-lined trapezoidal-sectioned ditch, or

2. a flexible hose (often 150 m in length), or
3. automatic hydrant coupling devices with buried mainline
• Hose-fed systems require periodic manual reconnection to hydrants on a
mainline – it is kind of like a period-move system, and you have to ask
yourself whether the linear move machine is worth the cost in this case!
• With the automatic hydrant coupling machines (see Fig. 15.3) there are two
arms with pipes and an elbow joint that bends as the linear move travels
down the field. The two arms alternate in connecting to hydrants so as not to
disrupt the irrigation nor the forward movement of the machine. These are
mechanically complex.
• The advantage of hose-fed and automatic coupling linear moves is that you
don’t need to have a small, uniform slope in the direction of travel, because
water is supplied from a pressurized mainline instead of an open channel
• On ditch-fed systems there can be a structure at the end of the field that a
switch on the linear move contacts, shutting down the pump and reversing
the direction of movement so that it automatically returns to the starting end
of the field.
• The advantages and disadvantages of the ditch-feed system are:
Pros
• Low pressure (energy) requirement
• Totally automated system
• More frequent irrigations than hose-fed, since no one needs to be
available to move the hose
Cons
• Trash and seeds and sediment pass through screen and may plug
nozzles
• The pump must be on (move with) the lateral, causing extra weight
• Should have uniform slope along the lateral route
Pros and Cons Compared to a Center Pivot
Pros
• Easy irrigation of a rectangular field (important if land is expensive, but
not important if land is cheap and water is scarce)
• Application rate is uniform over length of lateral, rather than twice the
average value at the end of the center pivot
• No end gun is required
Cons
• The lateral does not end up right back at the starting point immediately
after having traversed the irrigated area – you either have to
“deadhead” back or irrigate in both directions
• May be more expensive than a pivot due to extra controls, pump on
ditch feed, or more friction loss in the flexible feed hose (the hose is
fairly expensive)
II. System Costs
Relative costs for linear move systems are:
• $50,000 for a 1,280-ft hose-fed machine (perhaps 160 acres)

• $55,000 for a winch-tow hose fed 1,280-ft machine (perhaps 160 acres)
• $55,000 for a center pivot (1,280 ft) with a corner system (about 150 acres)
• $140,000 for a ½-mile (2,560-ft) linear with automated hydrant coupling
system (no ditch or hose required). The mainline is down the middle (a ¼-
mile lateral on each side). Perhaps 320 acres or more irrigated.
III. System Design
• A main strategy in linear move design is to minimize the cost per unit area.
This is done by maximizing the area covered per lateral (length of field)
• Generally, the lateral length is limited to 400 to 800 m. Therefore, the major
difficulties and objectives in linear design are to:
1. Maximize the irrigated area per lateral (this minimizes $/area). In other
words, how large a field can be irrigated by one machine?
2. Prevent runoff by matching ARx with Isoil + SS/ti (this tends to limit the field
length, because if AR is small, it won’t be possible to finish in f’ days),
where SS is the allowable surface storage in (mm or inches); and, ti is the
time of irrigation
3. Determine whether spray nozzles can be used without causing runoff
4. Minimize labor (for moving hoses and supervising)
as long as possible
• The allowable surface storage, SS, is the maximum amount of ponding
without incurring surface runoff
• SS is a function of the general topography and the microtopography, and of
the amount of foliar interception (water can “pond” on the crop leaves too)
• SS is usually less than about 5 mm unless small basins are created along
furrows, for example
• ARx limits the field length because it corresponds to some minimum time to
finish an irrigation, for a given gross application depth, whereby a maximum
interval (f) is calculated in the preliminary design steps
Lateral Inlet Head
• This is the same as for periodic-move systems

• The pressure balance equation for linear move systems is similar to set
systems (both are linear, with uniform discharge from each outlet).
Pa 3 1
Hl = + hf + Hr + ∆He + (hf )minor + (hf )hose (298)
γ 4 2
• Or, if using flow control nozzles, with a minimum pressure required at the end
(assuming the minimum pressure occurs at the end):
Pend
Hl = + hf + Hr + ∆He + (hf )minor + (hf )hose (299)
γ
where Hr is the height of the lateral or spray boom above the ground; and,
(hf)minor are the hydrant coupler and tower connection losses.
• The parameter (hf)hose is the loss in the flexible hose connection on a hose-
fed system
• Note that (hf)hose may be a major loss, since the hose diameter is usually less
than 5” or 6".
1.852
⎛Q⎞
hf = khFL ⎜ ⎟ D−4.87 (300)
⎝C⎠
where kh = 10.50 for hf and L in ft, Q in gpm, and D in inches; Kh = 1.21(10)10
for hf and L in m, Q in lps, and D in mm. F is the multiple outlet friction factor
for a linear move (F ≈ 0.36).
• For hose-fed systems, the maximum hose length for dragging the hose is
220 ft. Therefore, there could be about 400 ft between hydrants.
• For hose-fed systems with a cable/winch system for assisting in dragging the
hose (towers only have a moderate amount of tractive power), the maximum
hose length is 330 feet (640 feet between hose hydrants).
• Flexible hoses normally come in 5-inch ($18/ft) and 6-inch ($25/ft) diameters
• The Hazen-Williams C value for the hose can usually be taken as 150
Lecture 15
Maximizing Linear Move Field Length
I. The Procedure
• The following procedure for maximizing field length is from Allen, 1983, Univ.
Idaho and Allen, 1990 (Irrig. Symp. Paper), and is used in the USUPIVOT
computer program
• The basic strategy is to examine different application depths and different w
values to maximize the area covered by the sprinkler system, and or to
minimize labor requirements
1. Calculate the maximum application depth per irrigation (dx ≤ MAD*Z*Wa).

Note that the maximum application depth may be less than MAD*Z*Wa
with an automatic system to maintain optimal soil water conditions and to
keep soil water content high in case of equipment failure (i.e. don’t need
to take full advantage of TAW)
f' = dx/Ud (round down to even part of day)
2. Calculate net and gross application depths:
dn = f (Ud)
d = dn/Epa
3. Calculate the (presumed) infiltrated depth per irrigation:
(Df)max = d.Re
where (Df)max is the maximum depth to be evaluated, and assuming no

runoff
4. For a series of 10 or so infiltration depths, df, beginning with df equal to

some fraction (say 1/10) of (Df)max:
df = (i/10)(Df)max where i = 1 to 10
and,
f' = df DEpa/(100 Ud)
f = f' - days off (days off may be zero because the system is automatic),
where f' = irrigation frequency for depth df. DEpa is used here (in percent)
because Ud is net, not gross
5. Determine the maximum ARx for a particular df value using the following
two equations (assuming an elliptical pattern):
⎛ ( SF )( AR x ) ⎞ ⎛ n+1 n ⎞
1 n
⎜1− ⎜
⎟ k (n + 1)n+1 (D − SS − c )n+1 ⎟
⎝ k ⎠⎝⎜ ⎟
AR x = ⎠ (301)
2 2
⎛π⎞ ⎛ D ⎞
1.05 − 1.6 ⎜ ⎟ ⎜ − 0.5 ⎟
⎝ 2 ⎠ ⎝ df ⎠
where,
−n−1
⎛⎛ −0.5 ⎞
2 2⎞ ⎛ 2 ⎞
⎜⎜ ⎛ π ⎞ ⎛ D ⎞ ⎛ π ⎞ ⎛ D ⎞ ⎟
2 2 ⎟ ⎜ −1.6 AR2x ⎜ ⎟ ⎜ − 0.5 ⎟ ⎟ ⎟
⎜ ⎜ 1.05 AR x − 1.6 AR x ⎜⎝ 2 ⎟⎠ ⎜ d − 0.5 ⎟ ⎟ ⎜
⎜⎝ ⎝ f ⎠ ⎠ ⎝ ⎝ 2 ⎠ ⎝ df ⎠ ⎟⎠ ⎟
D=⎜ ⎟
⎛ ( SF )( AR x ) ⎞ ⎛ −1 ⎞ ⎛ 1 ⎞
⎜ ⎟
⎜
df n ⎜ 1 − ⎟ ⎜⎜ ( n + 1) n+1 ⎟⎟ ⎜ k n+1 ⎟ ⎟
⎜ ⎝ k ⎠⎝ ⎠ ⎝⎜ ⎟
⎠ ⎟
⎝ ⎠
+SS + c
(302)
and ARx is the peak application per pass (mm/min); D is the applied depth
at time t = ∫ (AR) dt (mm); SS is the allowable surface storage (after
ponding) before runoff occurs (usually less than about 5 mm); c is the
instantaneous soil infiltration depth, from SCS soil intake families (mm); k
is the coefficient in the Kostiakov-Lewis equation; and df is the total depth
of water applied to the ground surface (mm)
• The parameter “n” is defined as: n = a -1, where “a” is the Kostiakov
exponent (see NRCS soil curves at www.wcc.nrcs.usda.gov/nrcssirrig)
• Note that SS is a function of the field topography and micro-topography,
and is affected by foliar interception of applied water
• These last two equations have π in them because there is an inherent
assumption of an elliptical water application profile from the sprinklers or
sprayers
• Recall that ARav = (π/4)ARx for an elliptical pattern
• SF is a relative sealing factor (in terms of soil water infiltration), and may
have values in the range of 0 to about 0.36
• The higher values of SF tend to be for freshly tilled soils, which are
generally most susceptible to surface sealing from the impact of water
drops
• Lower values of SF are for untilled soils and vegetative cover, such as
alfalfa or straw, which tend to reduce the impact of water drops on the soil
and help prevent runoff too
• If the linear move irrigates in both directions (no deadheading), then df is
one-half the value from these two equations
6. Compute the total wetting time, ti, in minutes:
df
ti = (303)
π
( AR x )
4
7. Compute the speed of the system for the required ti:
S = w/ti (m/min) (w is for a specific nozzle type)
If S ≥ Smax (this may occur for a high intake soil or for a very light
application with surface storage) then reduce the application rate and
increase time as follows:
w
ti = (304)
Smax
4 df
AR x = (305)
π ti
Thus,
S = Smax (306)
8. Calculate maximum field length, X:
8(a). For irrigation in one direction, only (dry return, or deadheading):
60 f T − 2 treset
X= (307)
⎛ 1 1 t ⎞
⎜⎜ + + hose ⎟
⎟
⎝ S wet Sdry 100 ⎠
where,
X = maximum length of field (m);
f = system operating time per irrigation (days);
T = hours per day system is operated (21-23);
treset = time to reset lateral at each end of the field (min);
those = time to change the hose (min/100 m);
Swet = maximum speed during irrigation (m/min); and
Sdry = maximum dry (return) speed (m/min)
labor =
(
2 treset + 0.01X those + 2 t super ) (308)
60 f
where labor is in hrs/day; and tsuper is minutes of supervisory time per 100
m of movement
8(b). For irrigation in both directions (no deadheading):
60 f T − 2 treset
X= (309)
⎛ 1 t ⎞
2⎜ + hose ⎟
⎝ S wet 100 ⎠
and labor is calculated as above in 8(a)
9. Calculate the irrigated area:
XL
Areamax = (310)
10,000
where Areamax is in ha; and L is the total lateral length (m)
10. Labor per hectare per irrigation, Lha:
labor
Lha = (311)
Areamax
11. Repeat steps 5-10 for a different value of df
12. Repeat steps 4-11 for a new w (different application device or different
operating pressure)
13. Select the nozzle device and application depth which maximizes the field
length (or fits available field length) and which minimizes labor
requirements per ha
14. System capacity:
π AR x w L
Qs = (312)
4k 3 Re
where k3 = 96.3 for L and w in ft, Qs in gpm, and ARx in in/hr; and k3 = 60
for L and w in m, Qs in lps, and ARx in mm/min
The system capacity can also be computed as:
df w L
Qs = (313)
ti k 3 R e
II. Assumptions & Limitations of the Above Procedure
• In the above procedure (and in the USUPIVOT computer program), when

designing for a system which irrigates in both directions, the second pass is
assumed to occur immediately after the first pass, so that the infiltration
curve is decreased due to the first pass before the ARx of the second pass is
computed
• This will occur near the ends of the field, where the design is most critical.
The proposed procedure assumes that:
• There is no “surge” effect of soil surface sealing due to a brief time period
between irrigation passes (when irrigating in both directions)
• The infiltration curve used represents soil moisture conditions
immediately before the initiation of the first pass
• The infiltration curve used holds for all frequencies (f) or depths (df)
evaluated, while in fact, as f↑, θ↓, so that the Kostiakov coefficients will
change. Therefore, the procedure (and field ring infiltration tests) should
be repeated using coefficients which represent the Kostiakov equation for
the soil moisture condition which is found to be most optimal in order to
obtain the most representative results.
Linear Move Design Example
I. Given Parameters
• Hose-fed linear move, irrigating in only one direction in a 64-ha field (400 m
wide and 1,600 m long)
• The pressure is 140 kPa (20 psi) for spray booms with a preliminary w if 10
m (33 ft)
• The soil infiltration characteristics are defined for the Kostiakov-Lewis
equation as:
Z = 5.43τ0.49 (314)
with Z in mm of cumulative infiltrated depth; and τ is intake opportunity time

in minutes. Other design parameters:
Ud = 7.7 mm/day
MAD = 50%
Z = 0.9 m
Wa = 125 mm/m
Oe = 1.00
Re = 0.94
Epa = 85%
• Maximum dry (returning) speed = 3.5 m/min

• Maximum wet (irrigating) speed = 3.0 m/min
• Reset time = 0.5 hours per end of field
• Hose Reset time = 10 min/100 m of travel distance
• Supervisory time = 5 min/100 m of travel distance
II. One Possible Design Solution
• This design will consider only spray booms with w = 10 m

• Note that the full procedure would normally be performed with a computer
program or spreadsheet, not by hand calculations
1. Calculate the maximum application depth per irrigation [dx = MAD(z)(Wa),

or less]
dx = (0.5)(0.9)(125) = 56 mm (315)
dx 56
f' = = = 7.3 ⇒ f ' = 7 days (316)
Ud 7.7
2. Net and gross application depths:
dn = f Ud = (7)(7.7) = 54 mm (317)
dn 54
d= = = 64 mm (318)
Epa 0.85
3. Infiltrated depth at each irrigation:
D f = dR e = (64)(0.94) = 60 mm (319)
4. For a series of 10 infiltration values, calculate df, beginning with df = Df

/10:
⎛ i ⎞
df = D f ⎜ ⎟ (320)
⎝ 10 ⎠
where i = 1 to 10. For this example, let i = 4 and, df = (0.4)(60 mm) = 24

mm. Then,
df DEpa (24)(0.85 / 0.94)

f' = = = 2.8 days (321)
Ud 7.7
Assume no days off (no down time during the peak use period)
f = f '− days off = 2.8 − 0 = 2.8 days (322)
5. Determine the maximum ARx for the particular df depth:
From Eq. 282:
ARx = 0.97 mm/min
ARx reaching the soil surface = 0.97 (Re) = 0.91 mm/min
6. Compute the total wetting time, ti, in minutes:
4 df 4(24)
ti = = = 34 min (323)
π AR x π (0.91)
7. Compute the speed of the system for the required ti:

w 10
S= = = 0.3 m / min (324)
ti 34
Thus, S < Smax (3.0 m/min), so this is OK.
8. Calculate maximum field length, X:
For irrigation in one direction, only (deadhead back):
60 f T − 2 treset
X= =
⎛ 1 1 t ⎞
⎜⎜ + + hose ⎟
⎟
⎝ S wet Sdry 100 ⎠ (325)
60(2.8)(22) − 2(30)
= = 970 m
⎛ 1 1 10 ⎞
⎜ + + ⎟
⎝ 0.3 3.5 100 ⎠
and, the labor requirements are:
(
2 treset + 0.01 those + 2 t sup er X ) =
60 f
(326)
2(30) + 0.01[10 + 2(5)][970]
= = 1.5 hrs / day
60(2.8)
where treset is the reset time at the end of the field (min); those is the hose
reconnection time (min/100 m); and tsuper is the “supervisory” time
(min/100 m)
9. Maximum irrigated area:
Areamax = XL/10000 = 970(400)/10000 = 38.8 ha
which is only about half of the actual field area!
10. Labor per ha per irrigation, L/ha:
L/ha = (labor/area)max = 1.5/38.8 = 0.039 hr/ha/day
11. Repeat steps 5 - 10 for a new df (not done in this example)
12. Repeat steps 4-11 for a new w (different application device or operating
pressure). (not done in this example).
13. Select the nozzle, device and application depth that maximizes the field
length (or fits the available field length), and which minimizes labor
requirements per ha.
Note: 38.8 ha << 64 ha, which is the size of the field, (970 m << 1600 m
which is the length of the field). Therefore, it is important to continue
iterations (steps 11 and 12) to find an application depth and or new w
(different sprinkler or spray device) to reach 1600 m and 64 ha, if
possible.
Additional Observations:
• For a 6-m spray boom, applying a 12-mm depth per each 1.4 days
would almost irrigate the 64 ha. However, the labor requirement is
doubled, as the machine must be moved twice as often. This
additional cost must be considered and weighed against the larger
area irrigated with one linear move machine.
• If larger spray booms were used (w = 16 m rather than 10 m) (these
would be more expensive) then 18 mm could be applied each 2.1
days, and all 64 ha could be irrigated with one machine.
• If low pressure impact sprinklers were used (these would be less
expensive than spray booms, but energy costs would be higher), then
w = 22 m, and 30 mm could be applied each 3.5 days (more water
can be applied since the application rate is spread over a wider area
from the lateral), and all 64 ha could be irrigated. In addition, ETc
would be less since the soil would be wetted less often. Also, the soil
intake rate would be higher each irrigation because of a drier
antecedent moisture at the time of irrigation.
• Notice that required wetting time for rotation times (f) greater than 2
days are identical between all types of spray devices. This is
because, for the large depths applied, a minimum wetting time is
required. The system speed is adjusted to fit the w value of the water
application device.
• If no acceptable solution for this problem were found, then
alternatives to be evaluated would be to irrigate in both directions, or
to consider a ditch-fed linear move (this requires a leveled ditch, but
does not required time for moving hoses and hose friction losses).
• You could also consider a “robot” controlled machine that
automatically connects alternating arms to hydrants on a buried
mainline (but this is a very expensive alternative)
• You might begin to wonder whether an investment in a linear move
machine is justifiable when there is a significant labor requirement for
reconnecting the supply hose, resetting at the end of the field, and
supervising operation. That is, why not put in a center pivot or a side
roll system instead?
• If one linear move cannot cover the entire field length in the available
period, “f” (days), you could consider two linear move machines for
the same field
14. System Capacity:
π AR x w L π (0.91)(10)(400)
Qs = = = 51 lps (809 gpm)
4k 3Re 4(60)(0.94) (327)
alternatively,
df wL (24)(10)(400)
Qs = = = 51 lps (809 gpm) (328)
ti k 3R e (33.6)(60)(0.94)
Note that the computed Qs is larger than one based strictly on Ud and T,
because the machine is shut off during reset and hose moving
For Qs based only on f, A, d and T, with no consideration for those,
Ad (38.3)(24)
Qs = 2.78 = 2.78 = 44 lps (700 gpm) (16)
fT (2.8)(22)(0.94)
But this flow rate is too low – it does not consider hose moving and
reset time. So, the 51 lps system capacity should be used for design
Lecture 16
Trickle Irrigation System Components & Layout
I. Introduction and Descriptions
• “Trickle” and “drip” are terms used to describe what can be generally called
“micro-irrigation systems”, in which water is applied in relatively precise
quantities and precise times and at precise locations
• Land-leveling costs notwithstanding, trickle irrigation systems are usually the
most expensive types of on-farm water application system to install
• They can also be expensive to operate and maintain
• Usually, trickle irrigation systems are installed in areas where water is scarce
and or expensive, crop value is very high, or topographical and other
conditions might preclude the successful use of other types of irrigation
systems
• Not all micro-irrigation systems are complex and expensive
• Labor-intensive forms of micro-irrigation continue to be practiced in many
areas of the world, especially for vegetable and other “cash” crops
• For example, people may carry water in buckets or shoulder harnesses to
carefully pour at each plant in a field
• Or, porous pots are buried at regular intervals along rows and filled with
water individually, which seeps out into the surrounding soil
• Sometimes water is merely splashed onto crop beds by hand
II. Advantages and Disadvantages of Trickle Systems
Advantages
1. Significant water, fertilizer, and operating cost (labor and

power) savings are possible
2. Ease of field operations due to reduced weed problems and non-wetted soil
surface (e.g. strawberries)
3. Ability to apply saline water because of frequent (daily) irrigation; thus, soil
water salinity is nearly the same as the irrigation water salinity
4. Ability to operate on steep slopes and rough terrain
5. The ratio of crop yield to evapotranspiration can be higher under trickle
irrigation because of reduced soil surface evaporation, continuously high soil
water (near FC), and lower root zone salinity due to frequent application
6. Relatively easy to automate the system
7. Can be less labor intensive than some other irrigation systems
Disadvantages
1. Systems are expensive to purchase and install ($1,000 to $6,000 per ha)
2. Susceptibility to clogging of emitters, which usually have very small
openings – so, it is important to spend time and money on maintaining the
system, applying chemicals, and keeping filters clean
3. Possibly low distribution uniformity due to low operating pressures and
possibly due to steep slopes, especially along laterals, and due to clogging
4. Where laterals are on steep slopes, the water will drain out the downhill end
at every startup and shut-down.
5. Soils with very low intake rates will exhibit ponding and runoff
6. Salt tends to accumulate at the soil surface and around the wetted area --
when it rains, these accumulated salts may be driven into the root zone
7. These systems tend to require more capable and diligent management
because of the susceptibility to clogging, and because the systems are
usually designed to operate continuously during peak ET periods (can’t
afford to let the system shut down during these periods). These systems do
not usually take full advantage of the soil storage (buffer) capacity.
III. Trickle System Components
• The following is a list of many

of the common components
found in modern trickle
irrigation systems:
1. Pump & motor

2. Control head
• Valves
• Filters and
Screens
• Chemical Injection
Equipment
• Flow Rate or Volume Meters (U/S of acid injection)
3. Mainline
• Submains
• Manifolds
• Laterals
4. Water applicators
• Emitters
• Bubblers
• Sprayers
• Misters
• Others
5. Other equipment
• Valves & air vents
• Vacuum Relief Valves
• Pressure Relief Valves
• Various Pipe Fittings and Appurtenances
• Not all trickle irrigation systems will have all of these components
• For example, some systems are gravity-fed and require no pumping
• Simple systems may not have submains and manifolds
• Some systems do not have pressure relief or other types of safety valves
• Systems with relatively dirty water will have multiple levels of filtration, others
may have only minimal screening
IV. Types of Water Applicators
• The basic purpose of water applicators in trickle irrigation systems is to

dissipate energy
• This is because lateral pressures must be high enough to provide adequate
uniformity, yet emitters must yield small flow rates
• Otherwise, one could simply punch holes in a plastic pipe (in some cases this
is exactly what people do)
1. Drip Emitters
• Emitters are the typical water application device in many systems

• Can be “on-line” (usually with barb) or “in-line” (inserted into tubing
by the tubing manufacturer)
• Various approaches are used to provide
energy dissipation: long-path (“spaghetti
tubing”), tortuous path (labyrinth), orifice,
vortex,
• Some emitters have flushing, or continuous
flushing features to help prevent clogging while
still having small flow rates (some designs are
very clever)
• Some emitters have pressure compensating
features to provide a more constant flow rate
over a range of operating pressures
• Some emitters are designed with multiple outlets
2. Line Source Tubing
• These are typically buried

• Single chamber with small orifices
• Double chamber with orifices between chambers and orifices to
discharge water into the soil (acts something like a manifold to
control pressures and provide greater uniformity)
• Can be removed and reused next year (typically 4-5 years life)
• Can be “disked up” and left in the field as chunks of plastic
• Porous or “leaky” pipe, made from old tires or new materials
3. Micro-Sprayers
• These are like small sprinklers, but

may not overlap enough to wet the
entire ground surface
• Sometimes referred to as “spitters”
• A “gray area” between micro and
sprinkle irrigation – they have both
precise application and aerial +
soil distribution of water
• Larger wetted area per
applicator, compared to non-
spray emitters
• Less susceptible to clogging than most emitters
V. Types of Pipe Materials
• Laterals are usually polyethylene (PE)

• Laterals may be made from polybutylene (PB) or PVC
• Metal pipes are not used because of corrosion problems from injected
chemicals
• Non-buried lateral tubing should be black to discourage growth of algae and
other organic contaminants (don’t allow light to enter the tubing)
• Non-buried pipes preferably have ultraviolet light (UV) protection to prevent
rapid deterioration from exposure to sunlight
VI. Typical Trickle System Layouts
• Chemical injection should be upstream of filters to prevent system clogging

• Chemical tank and valves are often plastic to avoid corrosion and freeze-up
• A check valve will help prevent contamination of the water supply
• Manifolds are often the basic operational subunits
• Valves for flushing at ends of manifolds and laterals are often manually
operated
• Various mainline-manifold-header arrangements are possible
VII. Emitter Flow Rate versus Pressure
• The discharge equation for emitters is similar to that used for sprinkler
nozzles, but the exponent on the head or pressure term is variable
• An exponent of ½ corresponds to orifice flow, which is how some, but not all,
emitters are designed
• The general emitter equation is:
q = K dHx (329)
where q is the volumetric flow rate; Kd is the discharge coefficient; H is the

head (or pressure); and x is the exponent
• Note that the value of Kd is different depending on whether units of head or

pressure are used in the equation
• The exponent in the above equation is a function of the emitter design
• For purely turbulent orifice flow the exponent is ½, assuming the pressure
head is fully converted to velocity head
• Pressure compensating emitters have x ≤ ½. A fully pressure compensating
emitter would have x = 0 (but these don’t really exist)
• Long-path, laminar-flow emitters typically have x ≈ 0.7 (if the flow were
completely laminar, the exponent would be 1.0)
• Vortex-type emitters typically have x ≈ 0.4
VIII. Emitter Design Objectives
• The two basic emitter design objectives, other than energy dissipation, are:
1. Have a low value of the exponent, x

2. Have flushing properties to prevent clogging
• The first objective provides for pressure compensating features, if the

exponent is less than 0.5 (i.e. compensates better than a simple orifice)
• The first and second objectives tend to be conflicting, because the more
pressure compensating the less ability to flush particles
IX. Field-Measured Uniformity
• To measure emission uniformity in the field you can use an equation

equivalent to the Distribution Uniformity (DU), as applied to sprinkler
systems:
⎛ q'n ⎞
EU' = 100 ⎜ ⎟ (330)
⎜ qa ⎟
⎝ ⎠
where EU’ is the field test emission uniformity; qn’ is the average discharge of
the low ¼ emitters from the sampling; and qa is the average discharge of all
emitters sampled
• EU’ should be at least 95% for properly designed and properly maintained
trickle irrigation systems
• Note that it is impossible to calculate EU’ based on field measurements if the
system is being designed (hasn’t been installed yet) – in this case there are
other equations to approximate EU (recall the design efficiency for sprinkler
systems)
• Most nonuniformity in micro irrigation systems is caused by: (1) emitter
plugging, wear, and manufacturing variations; and, (2) nonuniform pressure
distribution in pipes and hoses
X. Manufacturer’s Coefficient of Variation
• Emitters of the same type and manufacture have variations in discharge (at
the same operating pressure) due to small differences from manufacturing
tolerances. Some variation is allowed in the interest of cost savings.
• The manufacturer’s coefficient of variation is defined as:
s
ν= (331)
qa
where the standard deviation, s, is calculated from at least 50 samples (all at

the same pressure) of the same emitter type and model; and the
denominator is the average discharge of all emitters
XI. Emitter Layouts
• There are a number of configurations designed to increase the percent

wetted area, and still be economical
• There are tradeoffs between flow per lateral and total length of pipe and
tubing
• Below are some common emitter layouts
XII. Valves & Automation
• Valve automation can be accomplished with individual timers/devices, or with

a central controller
1. Volumetric Valves
• Manually turned on
• Automatically turn off
2. Sequential Operation
• Manually turned on
• Automatic sequencing from low to high elevation
3. Fully Automatic
• Time-based: soil moisture sensors determine whether to turn the

system on at a given time each day, then the system runs for fixed
duration in each subunit
• Volume-based: uses a flow meter to measure a specified volume
delivered to a subunit, then turns water off to that subunit
• Soil moisture-based: uses tensiometers, resistance blocks, or other
device to determine when to irrigate and for how long
• Note that time-based systems may give varying application depths over time
if the system flow rate changes due to clogging of filters
• This can be partially corrected by using pressure compensating emitters
• However, the use of a volume-based system with a flow meter may be best
because the flow rate measurement also gives an indication about filter
clogging
Lecture 17
Filtration for Trickle Irrigation Systems
I. Introduction
• Water for trickle irrigation systems can

come from open reservoirs, canals, rivers,
groundwater, municipal systems, and other
sources
• Solid contaminants can include both organic
and inorganic matter
• Examples of inorganic matter are sand, silt
and clay (soil particles), and trash floating in
the water
• Examples of organic matter are bacteria,
algae, moss, weeds & weed seeds, small fish, insect larvae, snails, and
others (see the figure below)
• Solid contaminants need to be removed from trickle irrigation systems

because they:
1. Cause clogging of emitters, which can lead to serious water deficits

and non-uniformity of the applied water
2. May cause wear on pump impellers, emitter outlets, and other
hardware
3. Can provide nutrients which support the growth of bacteria in the
pipes
4. Can accumulate at the ends of pipelines and clog valves
5. Can contain weed seeds which aggravate weed control in the irrigated
area
6. Cost the farmer money
• All of the above problems translate to direct costs to the farmer
• These filters do not remove salts from the irrigation water (unless reverse
osmosis membranes are used, which are very uncommon in irrigation
systems and are not covered here)
• Groundwater usually requires less filtration than

surface water, but even groundwater should be
filtered
• The maximum allowable particle size in trickle
irrigation water is usually between 0.075 mm and
0.2 mm, so the water must be quite clean
• Filtration is almost always complemented by the
injection of various chemicals into the water to help
prevent clogging due to bacterial growth and
precipitation of solids from the water
• Solid particles smaller than emitter outlets can cause clogging when they
bridge at the opening (see the figure above)
• Some consultants recommend the removal of all particles larger than 1/10 of
the minimum outlet diameter for drip emitters, or about 1/7 of the minimum
outlet diameter for spitters, misters, and microsprayers
• Larger particles may be allowed with spitters, misters, and microsprayers
because of “shorter pathways” and sometimes larger openings
II. Types of Filtration
• The basic types of filtration used in trickle systems are:
1. Reservoirs (settling ponds)

2. Pre-screening devices
3. Sand separators
4. Sand (media) tanks
5. Gravity overflow screens
6. Tubular screens
7. Disc filters
III. Use of Reservoirs in pre-Filtration
• Some of the benefits of an open reservoir upstream of the pumps in a trickle

irrigation system are:
1. To buffer differences in supply and demand rates. The supply from a canal
or well seldom coincides exactly with the system requirements (flow rate and
duration), and the system requirements can change due to different numbers
of stations in operation, “down time”, and duration of sets.
2. To allow for settling of some of the suspended particles. In these cases the
reservoir serves as a “settling basin”. Precipitated sediment can be
periodically removed from the reservoir with equipment or manual labor.
Soil Texture Particle Size Vertical Settling

(microns) Velocity (mm/min)
Coarse sand > 500 38,000 (1½ mph)
Medium sand 250 - 500 22,000
Fine sand 100 - 250 5,000
Very fine sand 50 - 100 900
Silt 2 - 50 15
Clay <2 0.6 (very slow!!)
The barchart below has a logarithmic scale on the ordinate:
100,000
10,000
Settling velocity (mm/min)
1,000
100
10
0
Coarse sand Medium sand Fine sand Very fine Silt Clay
sand
Longer settling basins will allow more time for suspended particles to fall to
the bottom before arriving at the pump intake.
3. To aerate water pumped from wells, thereby oxidizing and precipitating
manganese and iron out of the water (some groundwater has manganese
and iron, and these can cause plugging of emitters). Only 1.5 ppm of either
manganese or iron can cause severe clogging problems in trickle laterals
and emitters (see Table 18.1 in the textbook).
4. To allow for air to escape when the water comes from a “cascading” well, in
which air becomes entrained into the water. Air in pipelines can dampen the
effects of water hammer, but also causes surges and blockages of flow.
5. To allow oils to collect on the water surface. Oils can cause rapid clogging of
most types of filters, requiring special cleaning with solvents and possible
replacement of sand media. When pumping from a reservoir the inlet is
below the water, and oil does not enter.
IV. Pre-Screening Devices
• These screens are intended to

prevent fish, large debris and trash
from entering the pipe system,
upstream of the other filtration
devices
• Pre-screening is not necessary for
groundwater or municipal water
supplies
• Pre-screening devices often have
“self-cleaning” features, otherwise
they can clog up rapidly
• Horizontal grills
• Screen plates
• Rotating and self-cleaning screens
• Gravity screen filters
• If the inlet line upstream of a pre-screening device is pressurized, you will

lose all of the pressure and have to repressurize downstream of the screens
– this is a major disadvantage to pre-screening in such cases
V. Sand Separators
• Sand separators are used to remove sand (but not organic matter) from the
water
• Most work by spinning the water in an enclosed column (or cone) to remove
sand through a centrifuge-type action
• There are no moving parts
• Solid particles with a density of approximately 1.5 g/cm3 can be removed by
these devices (most sand has a density of about 2.65 gm/cm3)
• Can remove from 70 to 95% of dense particles
• Periodic purging of accumulated sand (manual
or automatic) is necessary to maintain
performance
• Must have the correct flow rate through the
sand separator for proper operation, otherwise
less sand will be removed from the water
• Most sand separators have a pressure loss of
between 5 and 12 psi, from inlet to outlet. This
pressure loss does not change with time, only
with flow rate.
• Some sand separators are designed to fit down
into wells to protect the impellers and pump
bowls, but they are not as efficient as above-
ground sand separators
• Sand separators cannot remove all of the sand, and may pass large amounts
when the system is starting or stopping
• Therefore, screen filters should be installed downstream
• Sand separators are available but are not used as much as they were in the
past because people are using media tanks and other filters instead
• When taking water from a deep well, an alternative to using a sand separator
is to properly develop the well and use a good quality well screen
VI. Sand Media Filters
1. Introduction
• This type of filter is filled with sand, or some

other particles such as crushed (makes it
angular, traps debris better) granite,
crushed silica material (e.g. garnet)
• Some designers go by a uniformity
coefficient for the media, defined as the
ratio of 40% retained size (larger) to the
90% retained size (smaller) from a sieve analysis
• This uniformity should usually be between 1.0 (perfect!) and 1.5
• In most media filters the water passes through vertically, from top to bottom
• A drain screen (of which there are various types) at the bottom allows clean
water to exit, but prevents the media from leaving the filter tank.
• Many media filters have a layer of gravel around the bottom drain
• The inside top of the tank usually has some kind of “diffuser” to help spread
the dirty water evenly over the media surface
• Standard tank sizes are 24, 36 and 48 inches in diameter, with a media
depth of about 30 to 40 cm
• Tank walls are usually 3 - 5 mm thick, depending on the brand
• The tanks are often made out of carbon steel, type 304 stainless steel, or
type 316 stainless steel (if water has high salt content)
• Every installation should have at least two tanks so that back-flushing can
occur during operation, but many designers recommend at least three tanks
in which only one is back-flushed at a time
• New media should be rinsed with clean water before placing it in the tanks
because it may have dust and other particles in it
• Some tanks have not performed well when the installers failed to rinse the
media first (resulting in fine particles passing into the irrigation system when
the tanks are first put into use)
2. Applicability of Media Filters
• These filters are very good for removing relatively large amounts of organic
and inorganic matter, but some pre-screening is usually necessary with
surface water supplies
• High volume filtration at 20 to 30 gpm/ft2 (1.3 to 2.0 cm/s)
• Some silt and clay particles can also be removed by sand media filters, but
not by most screen-type filters. However, much silt and clay can pass
through a media filter too.
• Large volumes of particle contaminants can be collected in the sand media
before the media must be cleaned, or “back-flushed”
• In some cases the water must be pre-cleaned before entering the sand tanks
to prevent rapid accumulation of particle contaminants
• Media filters can also remove some sand from the supply water, but this
sand cannot always be effectively back-flushed from the media -- for large
amounts of sand, there should be a sand separator upstream of the media
tanks
• Industrial media filters are often five feet deep (or more), but have smaller
flow rates and less frequent back-flushing than agricultural media filters,
which may be only 14 inches deep
• Many of the particles captured by agricultural media filters stay within the
upper few inches of the sand because they are back-flushed frequently
3. Back-Flushing the Tanks
• Back-flushing is required to clean the tanks

• Back-flushing can be performed manually or automatically, based on elapsed
time and or on a pressure differential limit across the tanks
• Typical pressure differential triggers are 5 to 10 psi (35 to 70 kPa) greater
than the pressure differential when the media is clean (clean media typically
has a pressure differential of 3 to 5 psi, or 20 to 35 kPa)
• Often, a timer is set to back-flush at least one time per day, even if the
pressure differential criteria (for flushing) is not met
• Automatic back-flushing is recommendable, because labor is not always
reliable
• Some installations have view ports on the back-flush pipes so that an
operator can see if the water is clean (to know if the duration of the back-
flush is sufficient) and to see if any of the sand media is escaping during
back-flush
• The pressure differential during a backflush operation should be 7 - 10 psi
(50 - 70 kPa) – if it is greater than 10 psi, the flow rate is too high
VII. Secondary Filters
1. Tubular Screen Filters
• These are conventional screen filters, with two-dimensional surfaces and

little capacity to accumulate debris
• There are many different kinds and variations of these filters
• Primarily used as backup (safety) filters downstream of the primary filters
• If the screen becomes dirty and is not cleaned, the pressure differential can
become great enough to burst the screen. Or, the screen may stretch until
the openings expand enough to pass some of the debris (which is not
desirable)
• Flow through the filter is usually from inside to outside (debris is trapped on
the inside surface during operation) to prevent collapse of the screens
• Cleaning can be manual or automatic, and there are many varieties of
automatic cleaning methods
• Some filter designs have a rotating suction mechanism to clean the dirty
(inner) side of the screen element
• Manually-cleaned filters can have slow or quick release cover latches -- the
slow release latches are preferred because the quick release version can
“explode” if opened while the system is at operating pressure (dangerous to
personnel)
2. Disc Filters
• Similar to a tubular screen filter, but using tightly packed plastic

disks for the filter media, with a deeper filter area
• Holds more contaminants than a regular screen filter without
clogging
• Often installed in banks (several filters in parallel)
• Often have automatic back-flushing features, requiring higher
pressure (about 45 psi minimum pressure) than normally available
for system operation, so there is a special booster pump for
cleaning
• Cleaning is often performed when the pressure differential across the filter
reaches 6 psi
• The pressure differential for a clean filter should be about 1 to 4 psi (unless
too much water is being pumped through the filter)
• These filters are not designed to remove sand from the water (sand gets
stuck in the grooves)
• These filters can have clogging problems with some kinds of stringy algae
VIII. Chemical Injection for System Maintenance
• It is usually necessary to use chemicals to maintain a trickle irrigation system

-- if not, the system will eventually become clogged
• Some chemicals are for the addition of plant nutrients, or fertilizers; this is
called “fertigation”
• Different chemicals may be used for pest control: herbicides, insecticides,
fungicides
• Other chemicals are used to kill bacteria and other organic contaminants,
and maintain a sufficiently low pH
• The uniformity at which chemicals are applied to the field can be assumed to
be equal to the emission uniformity of the emitters (assuming the chemicals
are water soluble, which they should be if injected into the irrigation system)
• Following are guidelines for clogging hazard of irrigation water in trickle
systems with emitter flow rates of 2 to 8 lph (after Bucks and Nakayama
1980):
Kind of Problem Hazard Level
low moderate severe
suspended solids 50 ppm 50-100 ppm > 100 ppm
pH 7.0 7.0-8.0 > 8.0
salts 500 ppm 500-2,000 ppm > 2,000 ppm
bicarbonate -- 100 ppm --
manganese 0.1 ppm 0.1-1.5 ppm > 1.5 ppm
total iron 0.2 ppm 0.2-1.5 ppm > 1.5 ppm
hydrogen sulfide 0.2 ppm 0.2-2.0 ppm > 2.0 ppm
bacteria count 10,000/liter 10,000-50,000/liter > 50,000/liter
• Types of clogging in trickle systems that can be managed through the

injection of chemicals:
1. Slimy bacteria
These can grow inside pipes and inside emitters.
The chemicals used to kill this bacteria are
chlorine, ozone, and acids.
2. Iron and manganese oxides

Some kinds of bacteria can oxidize iron and manganese. Only small
amounts of iron or manganese are necessary to support bacterial
growth in the water. This can be treated with chlorine injection, acid
and chlorine injection, linear phosphate injection, and aeration in a
pond.
3. Iron and manganese sulfides

These are problematic with some groundwater. Dissolved iron and
manganese form a black, insoluble material. Manganese is toxic to
most plants in small concentrations, and this may become a problem
before clogging occurs. This can be treated by aeration,
chlorination, and acid injection.
4. Precipitation of calcium and magnesium carbonates

These are typically quantified by SAR (or adjusted SAR). You can
use a dropper to put some hydrochloric or muriatic acid on selected
drip emitters to test for the presence of carbonates and
bicarbonates. If present, the acid will react and cause a “fizzing”
sound and bubbles. One method to control precipitation of
carbonates is to continuously inject carbon dioxide gas into the
system, creating carbonic acid and lowering the pH. Sulfuric and
phosphoric acids are also used, as are SO2 generators. All of these
are designed to maintain the pH near (or slightly below) 7.0.
5. Plant root entry into underground emitters
This is mainly a problem in permanent (several years) buried trickle
irrigation laterals. Can use acid injection at end of each season for
perennials to kill roots that are in the buried drip tubing. Or, use
herbicides to kill roots in the tubing without damaging the plants.
Some emitters and plastic drainage pipe have herbicide in the plastic
to discourage roots from entering.
• It is dangerous to experiment with chemical mixtures in trickle systems,

because some mixtures can cause clogging of the emitters
• Test the mixture in a glass container first, checking it after several hours or
more, and determining whether the chemical mixture is water soluble
• Injection of chemicals should be after the system starts, and stopping before
the system is turned off
• As a rule of thumb, one can assume an average pipe flow velocity of 1 fps, or
0.3 m/s, divide this into the longest pipe distance in the system (from pump to
farthest emitter), and determine the time
• This is the time to wait after starting the pump, and the time to allow for
flushing before turning the pump off
For example, if the farthest emitter is 600 ft from the

pump, the travel time can be estimated at 600 s, or 10
min. Thus, you should wait 10 min before beginning
chemical injection, and discontinue chemical injection 10
min before stopping the system, or before changing
stations.
• Chemicals should be injected on a mass basis per set, not time. Thus, one
would want to apply a certain number of lbs or kg of a chemical in an
irrigation set, and it does not matter that it is all applied quickly or over a long
time (provided that the starting and stopping delay discussed above is
adhered to)
• The minimum injection rate can be put into equation form:
qc =
Fr A
=
(kg / ha )(ha ) = lph (332)
ctr Ta ( kg / liter )( 0.8 )( hrs / set )
where Fr is the mass application rate per unit area; A is the area irrigated per
set; c is the concentration of the chemical; tr is some kind of uniformity ratio,
taken to be 0.8; and Ta is the hours per set, or hours of chemical injection, if
shorter than the set time
Lecture 18
Trickle Irrigation Planning Factors
I. Soil Wetted Area
• Trickle irrigation systems typically apply small amounts of water on a

frequent basis, maintaining soil water near field capacity
• But, usually not all of the soil surface is wetted, and much of the root zone is
not wetted (at least not by design) by the system
• Recall that the system is applying water to each individual plant using one or
more emission points per plant
Widely-Spaced Crops
• These include orchards and vineyards, for example

• According to Keller & Bliesner, for widely-spaced crops, the percent wetted
area, Pw, should normally be between 33% and 67%
• The value of Pw from the irrigation system can fall below 33% if there is
enough rainfall to supplement the water applied through the trickle system
• Lower values of Pw can decrease the irrigation system cost because less
emitters per unit area are required
• Lower values of Pw can allow more convenient access (manual labor &
machinery) for cultural practices during irrigation
• Lower values of Pw can also help control weed growth in arid and semi-arid
regions, and reduce soil surface evaporation
• Lower values of Pw carry the danger that the soil will dry to dangerously low
levels more quickly in the event the irrigation system goes “off-line” for any
reason (power failure, broken pipe, pump problems, labor shortage, etc.)
• With lower values of Pw, there is less storage of applied water in the root
zone, especially with light-textured soils (sandy soils)
• With tree crops, low values of Pw can lead to “root anchorage” problems, in
which root extension is insufficient to support the trees during winds
Closely-Spaced Crops
• These include most row crops

• Actual Pw values may be near or at 100% with row crops and subsurface drip
irrigation systems (in the USA rows are typically spaced from 30 inches to 60
inches)
• Larger values of Pw usually mean more extensive root development, and
enhanced ability for the plant to make use of any rain water that may come
• Figure 19.1 in the textbook shows a generalized relationship between Pw,
amount of rainfall, and crop production level – the figure implies that
maximum crop yield may be higher under a trickle irrigation system than with
other methods
• Figure 19.1 indicates that 100% crop yield might be obtained, in general, with
Pw ≥ 33%
Wetted Soil Area, Aw
• The wetted soil area, Aw, is not measured at the soil surface, but from a
horizontal plane about 30 cm below the soil surface (actually, it depends on
root depth and soil type)
• The same is true for Pw
• The reason we are interested in Pw is to calculate the application depth “dx,”
as discussed in the following lecture
• This wetted area is distorted for sloping terrain, but the distortion is uniform
for uniform slopes (all other factors being the same)
• Wetted soil area can be estimated from empirical relationships and tables
(Table 19.1 in the textbook), but it is best to have site-specific field data in
which potential emitters are operated in the design area
• That is, test the emitter(s) and spacings in the field before completing the
irrigation system design
• Calculate percent wetted area, Pw, as follows:
⎛ N pS e w ⎞
Pw = 100 ⎜ ⎟ , for Se < 0.8w (333)
⎜ S pS rPd ⎟
⎝ ⎠
where Np is the number of emission points (emitters) per plant; Se is the
spacing of emitters along a lateral; w is the wetted width along the lateral; Sp
is the spacing of plants along a row; Sr is the spacing between rows; and Pd
is the fraction (not percent) of area shaded (see Lecture 19)
• Note that the numerator of Eq. 333 is wetted area, and the denominator is
actual plant area
• Note also that some emitters have multiple emission points
• Se is the spacing between emitters on the lateral; however, if Se is greater
than 0.8w, then use 0.8w instead:
⎛ 0.8N p w 2 ⎞
Pw = 100 ⎜ ⎟ , for Se ≥ 0.8w (334)
⎜ S pS r Pd ⎟
⎝ ⎠
• Note that w is a function of the soil type

• Se’ is the “optimal” emitter spacing,
defined as 0.8w
• There are practical limitations to the value
of Se with respect to Sp, otherwise there
may not be enough emitters per plant
(perhaps less than one)
• Sample calculation:
• Suppose Sr = Sp = 3.0 m, Pd =
80%, and w = 1.1 m
• Determine Np for Pw ≥ 33%
Se ' = 0.8w = 0.8(1.1) = 0.88 m (335)
Np (0.88)(1.1)
0.33 = (336)
(3.0)(3.0)(0.80)
whereby Np = 2.45. Then,
3(0.88)(1.1)
Pw = = 0.40 (337)
(3.0)(3.0)(0.80)
• For double-lateral trickle systems, spaced Se’ apart, Pw is calculated as

follows (see Eq. 19.4):
⎛ N S' ( S' + w ) ⎞
Pw = 100 ⎜ ⎟ , for Se ≤ 0.8w
p e e
( )
(338)
⎜ 2 Pd S pS r ⎟
⎝ ⎠
or,
⎛ 1.44 w 2 N ⎞ 72 w 2 N
Pw = 100 ⎜ ⎟=
p p
, for Se ≤ 0.8w
( ) ( )
(339)
⎜ 2Pd S pS r ⎟ Pd S pS r
⎝ ⎠
Se’=0.8w
0.5w
w
Se ’
0.5w
Se Se
Double laterals
• As in the previous equation, if Se > Se’, use Se’ instead of Se in the above
equation for double laterals
• In the above equation, the denominator has a “2” because Np for double
lateral systems is always at least 2
• For micro-spray emitters, the wetted area is greater than that measured at
the surface (because it is measured below the surface):
⎡ ⎛ Se ⎞ ⎤
⎢ N p ⎜ A s + (PS ) 2 ⎟ ⎥
Pw = 100 ⎢ ⎝ ⎠ ⎥ , for S ≤ 0.8 w (340)
e
⎢ S pS rPd ⎥
⎢ ⎥
⎣ ⎦
where As is the surface area wetted by the sprayer; and PS is the perimeter
(circumference) of the wetted surface area
• In the above equation for Pw, the term in the inner parenthesis is:
Se π w 2 π w Se π w ⎛ w ⎞
A s + (PS ) = + = ⎜ + Se ⎟ (341)
2 4 2 2 ⎝2 ⎠
where w is the diameter corresponding to As, assuming a circular area
• See Fig. 19.4 on sprayers
Salinity in Trickle Irrigation
I. Salinity in Trickle Systems
• Salinity control is specialized with trickle irrigation because (usually) less than
100% of the area is wetted, and because water movement in the soil has
significant horizontal components
• Irrigation water always contains salts, and fertilizers add salt to the crop root
zones -- salinity management in the crop root zone is a long-term
management consideration with trickle systems, as it is with any other
irrigation method
• Salts tend to accumulate, or “build up”, at the periphery of the wetted bulb
shape under the soil surface
1. Rain can push salts near the surface down into the crop root area (but
a heavy rain can push them all the way through the root zone)
2. If and when the irrigation system is not operated for a few days, there
can be pressure gradients in the soil that pulls salts from the periphery
up into the root zone
• The crop is depending on frequent irrigations (perhaps daily) to keep salt

build-ups from moving into the root mass
• It may be necessary to operate the trickle system immediately following a
light rain to keeps salts away from roots (even if the soil is at field capacity)
• Annual leaching with surface irrigation or sprinklers (on a trickle-irrigated
field) may be necessary to clean salts out of the root zone, unless there is a
rainy period that provides enough precipitation to leach the soil
• If the irrigation water has high salinity, trickle systems can provide for higher
crop production because the frequent irrigations maintain the soil salinity
nearer to the ECw (this is often not the case with sprinklers and surface
irrigation systems - salinity concentrates due to ET processes between water
applications)
II. Yield Effects of Salinity
• According to Keller, the relative crop yield can be estimated as (Eq. 19.6):
Yactual (ECe )max − ECw

Yr = = (342)
Ypotential (ECe )max − (ECe )min
• This is the relative crop yield (or production) in terms of soil water salinity
only
• ECw is the electrical conductivity of the irrigation water
• (ECe)max is the zero yield point, and (ECe)min is the 100% yield threshold
value
• (ECe)max may be as high as 32, and (ECe)min can be as low as 0.9
• This is based on the linear relationship between relative yield and salinity as
adopted years ago by FAO and other organizations
• Of course, calculated Yr values must be between 0 and 1
• Salinity of the soil extract, ECe, is measured by taking a soil sample to the
laboratory, adding pure water until the soil is saturated, then measuring the
electrical conductivity -- most published crop tolerance and yield relationships
are based on the ECe as a standard reference
• Crops don’t instantly die when the salinity approaches (ECe)max; the osmotic
potential increases and roots cannot extract the water that is there
• There can also be specific toxicity problems with minerals at high salinity
levels
• According to Allen, the relative yield will be near 100% for ECw less than
about 2(ECe)min, provided that frequent irrigations are applied (maintaining
salinity concentrations in root zone)
III. Leaching Requirement
• According to Keller & Bliesner, the leaching requirement under a trickle

system in an arid or semi-arid region does not consider effective rainfall (arid
regions often have more serious salinity problems, but tropical regions are
also subject to salinity in low areas)
• Look at Eq. 19.7:
ECw
LR t = (343)
ECdw
where LRt is the leaching requirement under trickle irrigation (fraction); and
ECdw is the electrical conductivity of the “drainage water”, which means the
water that moves downward past the root zone
• ECdw can be replaced by 2(ECe)max for daily or every-other-day irrigations

(keep water moving through the root zone), still obtaining Yr = 1.0
ECw
LR t = (344)
2 (ECe )max
IV. Allen’s Equation for LRt
• R.G. Allen suggests a more conservative equation for calculating the

leaching requirement under trickle irrigation:
1. For continuous trickle system operation (daily or once every two days),
the soil water in the root zone is maintained near field capacity, which can be
taken as approximately 50% saturation (θv) for many soils. Thus,
ECe = 0.5ECsoil (345)
(recall that ECe is measured after adding distilled water to the soil sample
until it is saturated)
2. Suppose the average ECsoil is taken as (0.667ECw + 0.333ECdw). Then,

for 100% relative yield at field capacity:
(ECe )min = 0.5 (0.667ECw + 0.333ECdw ) (346)
solving for ECdw,

ECdw = 6 (ECe )min − 2EC w (347)
3. Substitute this last equation into Eq. 19.7 from the textbook to obtain:
EC w
LR t = (348)
6 (ECe )min − 2ECw
this is similar to the leaching requirement as calculated for sprinkler
irrigation in Eq. 3.3 (coefficients 5 and 1 instead of 6 and 2), except that
(ECe)min is for 100% yield rather than 10% reduction in yield
Lecture 19
Water Requirements in Trickle Irrigation
I. Trickle Irrigation Requirements
1. Daily Use Rate
• The daily transpiration rate under a trickle system is based on Ud and the
percent area shaded (covered) by the plant leaves. Eq. 19.9:
Td = 0.1 Ud Pd (349)
where Ud is as previously defined and Pd is the percent (0 to 100) shaded

area when the sun is overhead (or most nearly overhead, in temperate
zones)
• Note that when Pd = 0, Td = 0

• Note that when Pd = 100%, Td = Ud
• Note that Td is called “transpiration,” but it really includes evaporation too
1.0
0.9
0.8
0.7
Closely-spaced crops
0.1*sqrt(Pd)
0.6
0.5
pi/4 = 0.785
0.4
0.3
0.2
0.1
0.0
0 10 20 30 40 50 60 70 80 90 100
Pd
• The reduction from Ud is justified by considering the typical reduction in wet
soil evaporation with trickle irrigation
• The maximum Pd for a mature orchard is usually about π/4 (0.785), which is
the ratio of the area of a square and the circle it encloses:
• Tree spacing is generally such that the trees do not compete for sunlight, and
the area of each tree is equal to the square of the spacing between them (for
a square spacing)
2. Seasonal Water Use
• This is calculated as for the peak daily use in Eq. 19.9:
Ts = 0.1 U Pd (350)
3. Seasonal Water Deficit
• To determine the seasonal water deficit, to be supplied from the irrigation

system, consider effective rainfall and initial soil moisture, in addition to
percent shaded area:
(
Dn = (U − Pe − Ms ) 0.1 Pd ) (351)
where U is used instead of Ts because Pe (effective precipitation) and Ms

(initial soil water content) are over the entire surface area
4. Net Depth per Irrigation
• This is the same as for sprinkle irrigation (or surface irrigation), but with an
adjustment for percent wetted area. Eq. 19.12 is:
MAD Pw
dx = Wa Z (352)
100 100
• Essentially, the same net volume of water is applied as with other irrigation
methods, but on a smaller area of the surface (and subsurface)
• Then, the maximum irrigation interval is:
dx
fx = (353)
Td
and f’ (round down from fx to get whole number of days) is less than or equal
to fx, but often assumed to be 1 day for trickle system design purposes.
Then,
dn = Tdf ' (354)
II. Gross Irrigation Requirements
• The transmission ratio (peak use period) takes into account the two-
dimensional infiltration pattern, or bulb shape, under trickle irrigation
• Even if the net depth is exactly right, there will almost always be some deep
percolation (more than that which may be required for leaching purposes)
• The transmission ratio, Tr, is used as a factor to increase required gross
application depth from dn
• The transmission ratio is equivalent to the inverse of the distribution
efficiency, DEpa, as given in Chapter 6 of the textbook
• The transmission ratio is lower for heavy-textured (“fine”) soils because there
is more lateral water movement in the soil, and the bulb shape is flatter; thus,
potentially less deep percolation losses
• Table 19.3 gives approximate values of Tr for different soil textures and root
depths (1.0 < Tr < 1.1) – obtain more representative values from the field, if
possible
• Then, for LRt < 0.1, or Tr > 1/(1-LRt), Eq. 19.15a:
⎛d T ⎞
d = 100 ⎜ n r ⎟ (355)
⎝ EU ⎠
where EU is the emission uniformity (%), which can be taken as a field-

measured value for existing trickle systems, or as an assumed design value
• EU takes into account pressure variations due to friction loss and elevation
change, and the manufacturer’s variability in emitter production
• If f’ = 1 day, then dn can be replaced by Td in Eq. 19.15a
• For LRt > 0.1, or Tr < 1/(1-LRt), Eq. 19.15c:
100dn
d= (356)
EU (1.0 − LR t )
• The difference in the above two equations is in whether LRt or Tr dominates

• If one dominates, it is assumed that the other is “taken care of” automatically
Gross Volume of Water per Plant per Day
• Equation 19.16:
d
G= SpSr (357)
f'
with d in mm; Sp and Sr in m; and G in liters/day
• Note that millimeters multiplied by square meters equals liters

• This equation does not use Pw because d is calculated for the entire surface
area, and each plant occupies an SpSr area
• Other versions of this equation are given in the textbook for gross seasonal
volume of water to apply
Required Application Time During Peak-Use Period
• Equation 20.11:
G
Ta = (358)
Npqa
where Ta is the required application (irrigation) time during the peak-use

period (hr/day), with G in litres/day, and qa in litres/hr
III. Coefficient of Variation
• This is a statistical index to quantify discharge variations in emitters, at the

same operating pressure, due to differences in the emitter construction
• The coefficient of variation is important in trickle system design and
evaluation because it can significantly affect the adequacy of the system to
irrigate the least watered areas of a field
• For statistical significance, there should be at least 50 measurements of
discharge from 50 individual emitters of the same design and manufacture
2
1⎛ n ⎞
∑ (qi2 )
n
− ⎜ ∑ qi ⎟
i=1
n ⎜⎝ i=1 ⎟⎠ σ
ν= = (359)
⎛1 n ⎞ qavg
n − 1 ⎜ ∑ qi ⎟
⎜n ⎟
⎝ i=1 ⎠
or,
n
∑ ( qi − qavg )
2
1 i=1
ν= (360)
qavg n −1
where n is the number of samples; σ is the standard deviation; qi are the

individual discharge values; and qavg is the mean discharge value of all
samples
• Standard classifications as to the interpretation of ν have been developed

(Soloman 1979):
Classification Drip & Spray Emitters Line-Source Tubing

Excellent ν < 0.05 ν < 0.1
Average 0.05 < ν < 0.07 0.1 < ν < 0.2
Marginal 0.07 < ν < 0.11 ---
Poor 0.11 < ν < 0.15 0.2 < ν < 0.3
Unacceptable 0.15 < ν 0.3 < ν
• For a large sample (n > 50) the data will usually be normally distributed
(symmetrical “bell-shaped” curve) and,
68% of the discharge values are within............... (1±ν)qavg

95% of the discharge values are within............. (1±2ν)qavg
99.75% of the discharge values are within........ (1±3ν)qavg
IV. System Coefficient of Variation
• The system coefficient of variation takes into account the probability that the
use of more than one emitter per plant will cause an effective decrease in the
combined discharge variability per plant due to differences in the emitters
(not due to pressure variability due to pipe friction losses and elevation
changes)
• On the average, discharge variability due to manufacturer tolerances will
tend to balance out with more emitters per plant
ν
νs = (361)
N'p
where Np’ is the minimum number of emitters from which each plant receives
water (see page 493 of the textbook)
• For a single line of laterals per row of plants,
L w = w + (N − 1)Se (362)
where Lw is the length of the wetted strip; and N is the number of emitters
(assumed to be evenly spaced). Then,
⎛L − w⎞
N = 1+ ⎜ w ⎟ (363)
⎝ Se ⎠
or,
⎛ Sp − w ⎞
Np ' ≈ 1 + ⎜ ⎟ (364)
⎝ Se ⎠
V. Design Emission Uniformity
• In new system designs it is not possible to go out to the field to measure the
EU’ (Eq. 17.2) – a different approach is required to estimate EU
• The design EU is defined as (Eq. 20.13):
qn
EU = 100 (1 − 1.27 νs ) (365)
qa
where qn is the minimum emitter discharge rate in the system, corresponding

to the emitter with the lowest pressure; and qa is the average emitter
discharge rate in the system, corresponding to the location of average
pressure in the system
• qn and qa are calculated (not measured) in new system designs by knowing

the topography, system layout, pipe sizes, and Qs
• Note that the value in parenthesis in Eq. 20.13 corresponds to the low one-
quarter emitter discharge
• EU gives a lower (more conservative) value than EU’, and the equation is
biased toward lower discharges to help ensure that the least watered areas
will receive an adequate application
• Graphical interpretations of these relationships are given in Figs. 20.9 and
20.10
VI. Average of the low ¼
• Note that the inclusion percentages for 1, 2 and 3 standard deviations

correspond to any normally distributed data
• Note also that ν qavg = σ
• The textbook says that for a normal distribution, the average flow rate of the
low one-quarter of measured q samples is approximately (1 - 1.27 ν)qavg
• The 1.27 coefficient can be determined from the equation for the normal
distribution and tabular values of the area under the curve
• The equation is:
2
1 ⎛ q−qavg ⎞
− ⎜ ⎟
2⎝ σ ⎠
e
occurrences = (366)
σ 2π
• To use the tabular values of area under the curve (e.g. from a statistics
book), it is necessary to use qavg = 0 and σ = 1 (the alternative is to integrate
the above equation yourself, which can also be done)
• Actually, qavg never equals zero, but for the determination of the 1.27
coefficient it will not matter
• In the tables, for area = 75%, the abscissa value (q, in our case) is about
0.675
• The same tables usually go up to a maximum abscissa of 3.49 (recall that
99.75% of the values are within ±3σ, so 3.49 is usually far enough)
• Anyway, for 3.49, the area is about 99.98%, and that is from -∞ to +3.49 (for
qavg = 0 and σ = 1), for the high ¼
• For the low ¼, take the opposite, changing to q = -0.675 and q = -3.49
• In this case (qavg = 0 and σ = 1), the equation reduces to:
2
e −0.5q
occurrences = (367)
2π
• For q = 0.675, occurrences = 0.31766718
• For q = 3.49, occurrences = 0.0.00090372
• Finally,
0.31766718 − 0.00090372
= 1.268 (368)
0.9998 − 0.7500
• Therefore, (1 - 1.268 ν)qavg is the average of the lowest 25% of measured

discharge values, for any given values of ν & qavg, and given normally-
distributed data
V. System Capacity
• The system capacity of a trickle system can be calculated by Eq. 20.15:
A N pq a
Qs = 2.78 (369)
N s S pS r
where Ns is the number of stations (sets); and A is the total net irrigated area.
Or,
A qa
Qs = 2.78 (370)
N s S eS l
where the coefficient “2.78” is for Qs in lps; A in ha; qa in lph; and Sp, Sr, Se,
and Sl in m (10,000 m2/ha divided by 3,600 s/hr = 2.78)
VI. Operating Hours per Season
• The approximate number of hours the system must operate per irrigation
season (or per year, in many cases) is equal to the required gross seasonal
application volume, divided by the system flow rate:
Vs
Ot = K (371)
Qs
where K = 2,778 for Vs in ha-m; and Qs in lps; and Vs is gross seasonal

volume of irrigation water
Lecture 20
Emitter Selection & Design
I. Introduction
• There are hundreds of models, sizes and types of emitters, sprayers,

bubblers, and others, available from dozens of manufacturers
• Prices of emitters can change frequently
• Some emitters have longer life than others, but cost more
• Some emitters have better pressure compensating features, but cost more
• Some emitters have better flushing capabilities, but cost more
• It is very difficult to know which is the “correct” emitter for a particular design,
and usually there are a number of emitters that could work and would be
acceptable for a given system
• Thus, the selection of an emitter involves knowledge of the different types,
their prices, their availability, and their performance
• Experience on the designer’s part is valuable, and emitter selection will often
involve a process of elimination
II. Long-Path Emitters
• So-called “spaghetti” tubing is a typical example of a long-path emitter

• Long-path emitters also come in spiral configurations (Fig. 20.1 of the
textbook)
• These can be represented by an equation used for capillary flow under
laminar conditions:
g π D4 H
Ac = (372)
ν qK
where lc is the length of the flow path; D is the inside diameter; H is the
pressure head; ν is the kinematic viscosity (a function of water temperature);
q is the flow rate; K is for units conversion; and g is the ratio of force to mass
• The above equation is only approximately correct for long-path emitters

• The above equation is based on circular cross-sections, which is typical
• The above equation assumes laminar flow, which may not be the case
• Note that the flow rate is proportional to the fourth power of the diameter, so
the diameter is a very important dimension
• Note also that the flow rate is inversely proportional to the length (double the
length and get half the flow rate)
• When is it valid to assume laminar flow? Consider that a Reynolds number
of 4,000 is probably as high as you can go without transitioning from laminar
to turbulent flow:
VD 4Q
= < 4,000 (373)
ν π νD
or, Q < 15D @ 10°C, with Q in lph and D in mm
cool warm hot
short
long
longest
• In black PE lateral hose, sunlight warms the water significantly as the velocity
slows down, and water viscosity decreases
• Long-path emitters would ideally be progressively longer along the lateral to
compensate and provide a more uniform discharge along the lateral
III. Tortuous- and Short-Path Emitters
• Tortuous-path emitters also have long paths, but not laminar flow. This is
because the path has many sharp bends, and is in the form of a maze
• Tortuous-path emitters tend to behave hydraulically like orifices, and so do
many short-path emitters
• Flow rate is nearly independent of the viscosity, at least over typical ranges
in viscosity
• Many short-path emitters have pressure compensating features
IV. Orifice Emitters
• Many drip emitters and sprayers behave as orifices

• The orifice(s) are designed to dissipate energy and reduce the flow rate to an
acceptable value
• Flow rate is approximately proportional to the square root of the pressure
V. Line Source Tubing
• Single-chamber tubing provides less uniformity than dual-chamber tubing

• In dual-chamber tubing, much of the head loss occurs through the orifices
between the two chambers. The outer chamber is somewhat analogous to a
manifold or header.
• The flow rate equation for dual-chamber tubing can be expressed as:
2gHno2
q = a'K
( )
(374)
1 + no2
where a’ is the area of the outer orifice; K is an empirical coefficient; H is the

pressure head; and no is the number of outer orifices per inner orifice (no >
1.0)
• See Fig. 20.2 in the textbook
VI. Vortex and Sprayer Emitters
• Vortex emitters have a whirlpool effect in which the water must exit through
the center of the whirlpool
• Energy is dissipated by the friction from spinning in a chamber, and from
exiting through an orifice in the center
• As mentioned in a previous lecture, the exponent on the pressure head is
approximately equal to 0.4 (in the discharge equation). Thus, these can
usually be considered to be (partially) pressure compensating
VII.Pressure Compensating Emitters
• Pressure compensating emitters usually have some flexible or moving parts

• These types of emitters tend to need replacement or repair more often than
most of the simpler emitter designs, therefore incurring higher maintenance
cost
• Figure 20.3 of the textbook shows one design approach for a pressure
compensating emitter
• As defined previously, pressure compensating emitters always have a
pressure head exponent of less than 0.5 (otherwise they aren’t considered to
be pressure compensating)
VIII.Self-Flushing Emitters
• In this category there are continuous and periodic flushing emitters

• Periodic flushing emitters perform their self cleaning when the lateral is filled
(before it reaches full operating pressure), and when the lateral is emptied.
In other words, they typically flush once per day.
• Continuous-flushing emitters have flexible parts that can stretch to allow solid
particles to pass through
• Fig. 20.4 in the textbook shows an example of one such design
• These can be sensitive to temperature changes and are not normally
pressure compensating
IX. Calculating the Discharge Exponent
• You can calculate the exponent, x, based on a pair of measured flow rates
and pressure heads
• Recall a rule of logarithms: log (ax) = x (log a)
• The solution can be obtained graphically, but is more quickly accomplished
with calculators and electronic spreadsheets
• If you have more than two pairs of q and H, then you can take the logarithmic
transformation of the equation and perform linear regression; however, the
regression will be mathematically biased toward the smaller values
Design Approach & Example
I. Review of Example Designs
• We will review example designs in Chapter 21 of the textbook, and discuss

design alternatives and parameters affecting efficiency, etc
II. Summarized Trickle Irrigation Design Process
• These are 15 basic steps, following the material presented in Chapters 17-24
of the textbook, that can be followed for the design of many trickle systems
• These are basic steps and represent a summary of the generalized design
process, but remember that each design situation will have some unique
features
1. Collect data on the crop, climate, soil, topography, and irrigation water
quality, field shape & size, water availability.
2. Select an emitter and determine an emission point layout such that 33% <
Pw < 67%. This will determine the number of emitters per plant, Np.
Emitter selection may involve field testing to determine the wetted width
(or diameter), w.
3. Calculate dx, fx, and Td. Note that fx will almost always be greater than
1.0.
4. Select a target value for EU (usually 70-95%; see Table 20.3) and
estimate the peak-use transmission ratio, Tr (usually 1.00-1.10; see Table
19.3).
5. Calculate the leaching requirement, LRt, based on crop type and irrigation
water quality.
6. Let f = 1 day (usually), then dn = Td. Calculate the gross application

depth, d.
7. Calculate the gross volume of water required per plant per day, G.
⎛ dS pS r ⎞
G = K⎜ ⎟ (375)
⎝ f ⎠
8. Calculate the daily hours of operation, Ta, (per station, or subunit) during
the peak-use period.
G
Ta = (376)
N pq a
9. Determine the number of operation stations based on Ta (with more

stations, the system capacity is lower).
If Ta = 24 hrs, then Ns = 1
If Ta = 12 hrs, then Ns = 1 or 2
If Ta = 8 hrs, then Ns = 2 or 3, and so on
10. Adjust Np and qa so that TaNs is equal to, or slightly less than, 90%(24
hrs/day) = 21.6 hrs/day. First, try adjusting qa because this is usually less
expensive than increasing Np. If the emitter is pressure compensating, or
if qa must be greatly altered, you may need to change Np (or you may
need to select a different emitter).
11. Having determined the value of qa, calculate the minimum allowable
emitter discharge, qn
q a EU
qn = (377)
100 (1.0 − 1.27 νs )
Note that if EU is high and νs is high, it could be that qn > qa (but this
would not be a reasonable calculation result!)
12. Calculate the average (nominal) and minimum lateral pressure heads
1/ x
⎛ q ⎞
h=⎜ ⎟ (378)
⎝ Kd ⎠
1/ x
⎛q ⎞
hn = ha ⎜ n ⎟ (379)
⎝ qa ⎠
13. Calculate the allowable change in pressure head in an operating station
∆Hs = 2.5 (ha − hn ) (380)
14. Calculate Qs, Vs, and Ot.
15. Finally, size the laterals, headers, manifolds and mainline(s) according to
hydraulic design criteria.
Lecture 21
Pipe Specifications & Trickle Lateral Design
I. Plastic Pipe Specifications
• Trickle and sprinkle irrigation systems are

commonly built with plastic pipe, of which there
are various types and specifications
• It is important to understand how the technical
specifications affect design decisions (pipe
sizing)
• Standards for the design and operation of
pipelines are available from various professional
organizations such as ASAE (American Society
of Agricultural Engineers) and AWWA (American
Water Works Association)
• Some of the material below is taken from ASAE
standard S376.1 OCT92
• ASAE standard S435 pertains to the use of PE pipe for microirrigation
laterals
• Plastic pipe is now commonly used in irrigation and other pipelines

• Some of the most common types are PVC (polyvinyl chloride), ABS
(acrylonitrile-butadiene-styrene), and PE (polyethylene)
• PVC pipes are usually white, while ABS and PE are usually black
• ABS pipes are often used for buried drains and drainage pipes
• All of these pipe materials are called “thermoplastic” because the material
can be repeatedly softened by increasing the temperature, and hardened by
a decrease in temperature
• The pressure rating of plastic pipe (especially PVC) decreases rapidly with
increasing temperature of the pipe and or water
• For example, at about 43°C (109°F) the PVC pressure rating
drops to one-half of the nominal value at 23°C (73°F), and
almost the same amount for PE
• PE pipe temperature can easily reach 43°C on a sunny day
• Unlike most metal pipes, these plastic pipe materials are immune to almost
all types of corrosion, whether chemical or electrochemical
• The resistance to corrosion is a significant benefit when chemigation is
practiced in a pressurized irrigation system
• The dimension ratio (DR) of a plastic pipe is the ratio of average diameter (ID
or OD) to wall thickness
• PVC, ABS and some PE are OD-based, while other PE pipe is ID-based
• Plastic pipe is currently manufactured up to a maximum diameter of 54
inches
• There are several standard dimension thickness
ratios (SDR) for several values, each with
its own pressure rating (at 23°C)
• Different types of PVC, ABS and PE
compounds exist, some of which are
stronger than others diameter
• Some plastic pipe is manufactured with
non-standard dimension ratios; in these
cases the ratio is called “DR” rather than
“SDR”
• Some pipe sizes are correspond to iron
pipe size (IPS), plastic irrigation pipe (PIP), and others
• These are different standards for indirectly specifying pipe dimension ratios
and pressure ratings
• The relationship between SDR, hydrostatic design stress (S in psi), and

pressure rating (PR in psi) for OD-based pipe is defined by ISO standard
161/1-1978
• The pressure rating (PR), which is the maximum recommended operating
pressure, is determined by the following equations:
2S 2S
PR = = (OD-based) (381)
SDR − 1 ⎛ OD ⎞
⎜ t − 1⎟
⎝ ⎠
2S 2S
PR = = (ID-based) (382)
SDR + 1 ⎛ ID ⎞
⎜ t + 1⎟
⎝ ⎠
where t is the pipe wall thickness
• Values of S can be obtained from published tables, as can values of PR for

given SDR and pipe material (plastic compound)
• Values of S vary from 6900 to 13,800 kPa for PVC, and from 3400 to 5500
kPa for PE
• Common terms used in the industry for PVC pipe include Class 160, Class
200, Schedule 40, Schedule 80 and Schedule 120 (in increasing strength
and decreasing SDR)
• With the “schedule” classification, the higher the schedule, the thicker the
walls, for a given nominal pipe diameter
• The maximum allowable operating pressure is approximately equal to:
(schedule)SE
P= (383)
1000
where P is the operating pressure (psi); S is the allowable stress in the pipe
material (psi); E is the “joint efficiency”; and “schedule” is the schedule
number (e.g. 40, 80, 120, etc.)
• Joint efficiency (or “joint quality factor”) for PVC is approximately 1.00, due to
the fact that it is seamless
• Class 160 and 200 refer to 160 psi and 200 psi ratings, respectively
• The Schedule 40 and 80 specifications have carried over from classifications
used in iron pipes
• Schedule 80 is seldom used in irrigation because its pressure rating is much
higher than the maximum pressures found in most irrigation systems
• Schedule 40 is commonly used in irrigation
• Some specifications for the design and protection of pipelines depend on

whether the pressure is “low” or “high”
• Low pressure pipelines are generally considered to have operating pressures
less than about 80 psi
• The maximum working pressure in a plastic pipe should normally be about
70% of the pipe’s pressure rating, unless special care is taken in design and
operation such that surges and excessive pressure fluctuations will not occur
• Manufacturers and testing centers provide data on minimum bursting

pressures
• Depending on the SDR value, the minimum burst pressure for plastic pipes
should be between about 900 and 1800 kPa (130 and 260 psi), otherwise the
pipe does not meet standard specifications
• Below is a glossary of common pipe abbreviations and terms:
Abbreviation Meaning
ABS Acrylonitrile-Butadiene-Styrene
DR Dimension Ratio
ID Inside Diameter
IPS Iron Pipe Size
ISO International Organization for Standardization
OD Outside Diameter
PE Polyethylene
PIP Plastic Irrigation Pipe
PR Pressure Rating
PVC Polyvinyl Chloride
SDR Standard Dimension Ratio
II. Trickle Irrigation Laterals
• Laterals are often above ground, but may be buried

• Drip “tape”, single- and dual-chamber laterals are usually buried a few
centimeters below the ground surface
• Above-ground laterals may be on the ground surface, or suspended above
the surface (e.g. in vineyards)
• Black polyethylene (PE) plastic pipe (or “hose”) is usually used for trickle
irrigation laterals
• Lateral pipes are typically about 0.5 or 1.0 inches in diameter
• Standard PE sizes are usually ID based and come in standard dimension
ratio (SDR) values of 15, 11.5, 9, 7 and 5.3
• Nominal PE pipe sizes for laterals are ½-inch, ¾-inch, 1-inch, and 1¼-inch
(all iron pipe size, or IPS)
• Laterals are usually single-diameter, but can be dual-sized

• Dual-sized lateral hydraulic analysis is essentially the same as previously
discussed for dual-sized sprinkler laterals
• To start a new system design, Keller & Bliesner recommend limiting the
lateral pressure variation to 0.5∆Hs, where ∆Hs is calculated from Eq. 20.14
• Then, 0.5∆Hs remains for the manifolds (if manifolds are subunits, or
“stations”)
• In lateral designs, the pipe diameter is usually chosen (not calculated), and if
the pressure variation or loss is “out of range”, then a different size can be
selected
• There are usually only a few lateral diameters to choose from
III. Trickle Lateral Hydraulics
• Friction loss gradients in plastic lateral pipe can be approximated by

combining the Darcy-Weisbach and Blasius equations (Eq. 8.7a):
7 Q1.75
J = 7.83(10) (384)
D4.75
for J in m/100 m; Q in lps; and D in mm
• The Blasius equation estimates the D-W “f” factor for smooth pipes
• If you want to calculate relative roughness, use ε = 1.5(10)-6 m
• It may be necessary to increase the J value because of emitter losses within

the lateral hose (barb, etc.) (see Fig. 20.8)
• Equation 22.1 is:
⎛S +f ⎞
J' = J ⎜ e e ⎟ (385)
⎝ Se ⎠
where fe is an equivalent length of lateral hose for each emitter, spaced
evenly at a distance of Se
• The fe pipe length is one way that minor hydraulic losses are calculated in
pipes
• From Eq. 8.7a, a dimensionless friction loss equation can be developed (see
Fig. 8.2), which is useful in semi-graphical hydraulic design work for trickle
irrigation laterals
• This is discussed in detail in the following lectures
• For a given lateral pipe size, lateral length, emitter spacing, and nominal
discharge per emitter, the lateral inlet pressure must be determined such that
the average lateral pressure is “correct”
• Then, the manifold can be designed to provide this lateral inlet pressure with
as little variation (with distance) as possible
• Figure 22.1 shows four different hydraulic cases for single lateral designs
• The design of pairs of laterals is essentially a compound single lateral
problem, with the added criterion that the minimum pressure be the same in
both laterals
• Not including riser height, the required lateral inlet pressure is (Eq. 22.6):
H l = H a + kh f + 0.5 ∆he (386)

where k is 0.75 for single pipe size laterals, or 0.63 for dual pipe size laterals
(as in the design of sprinkler laterals); and ∆he is positive for laterals running
uphill
• The minimum pressure in a lateral is given by Eq. 22.7:
H'n = Hl − (h f + ∆he ) − ∆H c
(387)
H'n = H c − ∆H c
where Hc is the pressure head at the closed end of the lateral
IV. References (plastic pipe)

http://www.uni-bell.org/lit.cfm
http://www.dpcpipe.com/ag/pipirrig.html
ASAE Standards (1997). American Soc. of Agric. Engineers, St. Joseph, MI.
Handbook of PVC Pipe. (1979). Uni-Bell Plastic Pipe Association, Dallas, TX.
Trickle Manifold Location
I. Optimal Manifold Location
• If the ground slope along the direction of the laterals is less than 3% or so, it
is usually recommendable to run laterals off both sides (uphill and downhill)
of each manifold
• If the ground slope along the direction of the laterals is more than 3%, it may
be best to run the laterals only in the downhill direction
• The design objective for a pair of laterals is to have equal values of minimum
pressure, Hn’, in uphill and downhill laterals
• This means that the downhill lateral will always be longer for laterals of equal
pipe size on sloping ground
• The manifold should be located in-between rows of plants (trees), not over a
row
• For laterals on flat ground, the manifold goes in the center of the field (the
trivial solution)
II. Sample Graphical Solution for Manifold Location
• Use the dimensionless friction loss curves (Fig. 8.2) to locate the optimal
manifold position in a sloping field
• The laterals run along the 0.021 m/m slope
• The combined uphill + downhill lateral length is 315 m
• The spacing of plants (trees) is
Sp = 4.5 m
• The spacing of emitters is Se =

1.5 m (thus, Np = 3 for single
line)
• The equivalent emitter loss is
fe = 0.12 m
• The nominal emitter discharge
is qa = 3.5 lph at 10 m head
(68.9 kPa)
• The lateral pipe ID is 14.7 mm
Solution for Manifold Location:
1. Number of emitters for the pair of laterals is:
315 m
= 210 emitters (388)
1.5 m/emitter
2. Total nominal discharge for the pair of laterals is:
Qpair =
( 210 emitters )( 3.5 lph/emitter ) = 12.25 lpm (389)
( 60 min/hr )
3. From Table 8.2 (page 141), J ≅ 13.3 m/100m. The adjusted J is:
⎛ S e + fe ⎞ ⎛ 1.5 + 0.12 ⎞
J' = J ⎜ ⎟ = 13.3 ⎜ ⎟ = 14.4 m/100 m (390)
⎜ Se ⎟ ⎝ 1.5 ⎠
⎝ ⎠
4. Multiple outlet factor, F = 0.36 for 210 outlets
5. Friction loss for the pair of laterals:
J'FL (14.4 )( 0.36 )( 315 )

(hf )pair = = = 16.3 m (391)
100 100
6. Elevation change for the pair of laterals:
( ∆he )pair = ( 315 m)( 0.021) = 6.62 m (392)
7. Ratio of elevation change to friction loss for the pair:
⎛ ∆he ⎞ 6.62
⎜ ⎟ = = 0.41 (393)
⎝ f ⎠pair
h 16.3
8. From the nondimensional graphical solution (Fig. 8.2): x/L = 0.69. Then, x
= (0.69)(315 m) = 217 m. Look at the figure below:
How was this done?
• Looking at the above figure, a straight line was drawn from the origin
(0, 0) to (1.0, 0.41), where 0.41 is the ratio calculated above
• The nondimensional curve was overlapped and shifted vertically so
that the curve was tangent to the same straight line, then traced onto
the graph
• The nondimensional curve was then shifted vertically even more so
that the inverse half-curve (dashed) intersected the (1.0, 0.41) point,
also tracing it onto the graph
• The intersection of the two traced curve segments gave an abscissa
value of about 0.69, which is the distance ratio
9. Finally, adjust x for tree spacing,
(217 m)/(4.5 m/tree) = 48.2 trees
• Therefore, round to 48 trees

• Then, x = (48 trees)(4.5 m/tree) = 216 m
• This way, the manifold lays buried halfway between two rows of trees,
not on top of a row
• This manifold position give the same minimum pressure in both the uphill
and downhill laterals
• Minimum pressure in the downhill lateral is located approximately
(0.35)(315 m) = 110 m from the closed end, or 216 - 110 = 106 m from
the manifold.
• This graphical solution could have been obtained numerically

• But the graphical solution is useful because it is didactic
• If you like computer programming, you can set up the equations to solve
for the lateral hydraulics based on non-uniform emitter discharge (due to
pressure variations in the laterals), non-uniform ground slope, etc.
• Note that this procedure could also be used for sprinklers, but it would
probably only be feasible for solid-set, fixed systems
Lecture 22
Numerical Solution for Manifold Location
I. Introduction
• In the previous lecture it was seen how the optimal manifold location can be
determined semi-graphically using a set of non-dimensional curves for the
uphill and downhill laterals
• This location can also be determined numerically
• In the following, equations are developed to solve for the unknown length of
the uphill lateral, xu, without resorting to a graphical solution
II. Definition of Minimum Lateral Head
• In the uphill lateral, the minimum head is at the closed end of the lateral
(furthest uphill location in the subunit)
• This minimum head is equal to:
hn ' = hl − hfu − xuS (394)
where hn’ is the minimum head (m); hl is the lateral inlet head (m); hfu is the
total friction loss in the uphill lateral (m); xu is the length of the uphill lateral
(m); and S is the slope of the ground surface (m/m)
• Note that S must be a positive value

manifold
location
(hfd)1 hfu
(hfd)2
xm xu
h n’ h n’
• In the downhill lateral, the minimum head may be anywhere from the inlet to
the outlet, depending on the lateral hydraulics and the ground slope
• The minimum head in the downhill lateral is equal to:
hn ' = hl − (hfd )1 + (hfd )2 + x mS (395)
where (hfu)1 is the total friction loss in the downhill lateral (m); (hfu)2 is the
friction loss from the closed end of the downhill lateral to the location of
minimum head (m); and xm is the distance from the manifold (lateral inlet) to
the location of minimum head in the downhill lateral (m)
• Combining Eqs. 1 and 2:
hfu + S ( xu + xm ) − ( hfd )1 + ( hfd )2 = 0

(396)
III. Location of Minimum Head in Downhill Lateral
• The location of minimum head is where the slope of the ground surface, S,
equals the friction loss gradient, J’:
S = J' (397)
where both S and J’ are in m/m, and S is positive (you can take the absolute
value of S)
• Using the Hazen-Williams equation, the friction loss gradient in the downhill
lateral (at the location where S = J’) is:
⎡ Se + fe ⎤ ⎡ ⎛ qa (L − xu − xm ) ⎞ ⎤
1.852
⎢ 10 −4.87 ⎥
J' = ⎢ ⎥ 1.212(10) ⎜ ⎟ D (398)
⎣ Se ⎦ ⎢⎣ ⎝ 3,600Se C ⎠ ⎥
⎦
where: J’ is the friction loss gradient (m/m);

Se is the emitter spacing on the laterals (m);
fe is the equivalent lateral length for emitter head loss (m);
qa is the nominal emitter discharge (lph);
L is the sum of the lengths of the uphill and downhill laterals (m);
xu is the length of the uphill lateral (m);
xm is the distance from the manifold to the location of minimum head
in
the downhill lateral (m);
C is approximately equal to 150 for plastic pipe; and
D is the lateral inside diameter (mm);
• The value of 3,600 is to convert qa units from lph to lps

• Note that xd = L - xu, where xd is the length of the downhill lateral
• Note that qa(L - xu - xm)/(3,600 Se) is the flow rate in the lateral, in lps, at the
location of minimum head, xm meters downhill from the manifold
• Combining the above two equations, and solving for xm:
⎛ 0.0129SeCD2.63 ⎞ ⎛ SeS ⎞0.54
xm = L − xu − ⎜ ⎟⎟ ⎜ ⎟ (399)
⎜ q ⎝ S + f ⎠
⎝ a ⎠ e e
where the permissible values of xm are: 0 ≤ xm ≤ xd
• Combine Eqs. 396 & 399, and solve for xu by iteration

• Alternatively, based on Eq. 8.7a from the textbook, xm can be defined as:
0.571
⎡ 3,600Se ⎤ ⎡⎛ SD4.75 ⎞ ⎛ Se ⎞ ⎤
xm = L − xu − ⎢ ⎥ ⎢⎜⎜ 5 ⎟⎟ ⎜ S + f ⎟
⎥ (400)
⎣ qa ⎦ ⎢⎣⎝ 7.89(10) ⎠ ⎝ e e ⎠ ⎥⎦
IV. Definition of Head Loss Gradients
• In the uphill lateral, the head loss is:
hfu = Ju 'Fu xu (401)
• In the downhill lateral, the head losses are:
(hfd )1 = Jd1 'Fd1 (L − xu ) (402)

and,
(hfd )2 = Jd2 'Fd2 (L − xu − xm ) (403)
• The above three “F” values are as defined by Eq. 8.9 in the textbook
• The friction loss gradients (in m/m) are:
Ju ' = K J ( xu )
1.852
(404)
Jd1 ' = K J (L − xu )
1.852
(405)
Jd2 ' = K J (L − xu − xm )
1.852
(406)
where, for the Hazen-Williams equation,
⎛ Se + fe ⎞ ⎡
1.852 ⎤
− ⎛ q ⎞
⎟ ⎢1.212(10) D ⎥
10 4.87
KJ = ⎜ ⎜
a
⎟ (407)
⎝ S e ⎠ ⎢ ⎝ 3,600S C
e ⎠ ⎥
⎣ ⎦
V. Solving for Optimal Manifold Location
• Using the definitions above, solve for the length of the uphill lateral, xu
• Then, xd = L - xu
• Note that you might prefer to use the Darcy-Weisbach and Blasius equations
for the manifold calculations; they may be more accurate than Hazen-
Williams
• The “OptManifold” computer program uses the Darcy-Weisbach & Blasius
equations
Where do these Equations Come From?
I. Derivation of Nondimensional Friction Loss Curves
• The nondimensional friction loss curves are

actually one curve, with the lower half
laterally inverted and shown as a dashed line
(Fig. 8.2)
• The dashed line is simply for flow in the
opposite direction, which for our purposes is
in the uphill direction
• We know from the previous lectures and
from intuition that the uphill segment of
lateral pipe will not be more than ½ the total
length, because it is equal to ½ for the case
where the ground slope is zero
• Following is the derivation for Eq. 8.10b, from

which Fig. 8.2 was plotted
• Darcy-Weisbach equation for circular pipes:
L V2
hf = f (408)
D 2g
• Blasius equation, for estimating f for small diameter (D < 125 mm) “smooth
pipes” (e.g. PE & PVC), and based on more complete equations that are
used to plot the Moody diagram
−0.25
f ≅ 0.32NR (409)
where NR is the Reynolds number, which for circular pipes is:
VD 4Q
NR = = (410)
ν νπD
• The kinematic viscosity, ν, is equal to about 1.003(10)-6 m2/s for water at
20°C
• Then, for this kinematic viscosity,
−0.25 −0.25
⎛ 4Q ⎞ ⎛Q⎞
f ≅ 0.32 ⎜ ⎟ ≈ 0.0095 ⎜ ⎟ (411)
⎝ νπD ⎠ ⎝D⎠
• Putting the above into the Darcy-Weisbach equation:
−0.25
⎛Q⎞ L V2
hf = 0.0095 ⎜ ⎟ (412)
⎝D⎠ D 2g
or:
Q1.75
hf ≅ 0.00079L (413)
D4.75
where hf is in m; L is in m; Q is in m3/s; and D is in m
• Eq. 8.7a is obtained by having Q in lps, and D in mm, whereby the above
coefficient changes to 7.9(10)7
• Finally, in the above, use L(x/L) instead of L, and Q(x/L) instead of Q, and
call it “hfx”:
⎡Q ( x / L ) ⎤⎦
1.75
hfx ≅ 0.00079L(x / L) ⎣ (414)
D4.75
• Then,
hfx
= (x / L) ( x / L ) = (x / L)
1.75 2.75
(415)
hf
which is Eq. 8.10b and the basis for the nondimensional friction loss curves,
valid for plastic pipes with D < 125 mm
II. Derivation of Equation for ∆Hc
• The difference between the minimum pressure head and the pressure head
at the closed end of a lateral, ∆Hc, is used to calculate the minimum head in
the lateral, Hn’
• This is because the pressure head at the end of the lateral is easily
calculated as:
H c = H l − h f − ∆he (416)
where ∆he is negative for downhill slopes
• But the minimum pressure head does not necessarily occur at the end of the
lateral when the lateral runs downhill
• Thus, in general,
H'n = H c − ∆H c (417)
• The above is from Eq. 22.7 in the textbook

• These concepts can also be interpreted graphically as in Fig. 22.1
• Following is a derivation of an equation for ∆Hc (based on Keller and Rodrigo
1979)
1. The minimum pressure in the lateral occurs where the ground slope (for a
uniform slope) equals the slope of the friction loss curve. The dimensionless
friction loss curve is defined as (Eq. 8.10b or Eq. 22.3b):
2.75
⎛ hfx ⎞ ⎛x⎞
⎜ ⎟ =⎜ ⎟ (418)
⎝ hf ⎠pair ⎝ L ⎠
2. The slope of this friction loss curve is:
⎛h ⎞
d ⎜ fx ⎟
1.75
⎝ hf ⎠pair ⎛x⎞
= 2.75 ⎜ ⎟ (419)
⎛x⎞ ⎝L⎠
d⎜ ⎟
⎝L⎠
3. The uniform ground slope on the dimensionless graph is:
⎛ ∆he ⎞ SL 100S
⎜ ⎟ = = (420)
⎝ hf ⎠pair ⎛ J'FL ⎞ J 'F
⎜ 100 ⎟
⎝ ⎠
4. Then,
100S
= 2.75 ( y )
1.75
(421)
J'F
in which y is the value of x/L where the minimum pressure occurs (0 ≤ y ≤ 1);
S is the ground slope (m/m); J’ is the friction loss gradient for the flow rate in
the pair of laterals (m/100 m); and F is the reduction coefficient for multiple
outlet pipes (usually about 0.36)
5. Solve for y:
1/ 1.75
⎛ 100S ⎞
y=⎜ ⎟ (422)
⎝ 2.75 J'F ⎠
or,
1/ 1.75
⎛ 100S ⎞
y≈⎜ ⎟ (423)
⎝ J' ⎠
where F ≈ 0.36
6. Referring to the figure on the previous page, the following equality can be
written:
∆Hc 100 y S
y 2.75 + = (424)
(hf )pair J'F
solving for ∆Hc,

⎛ 100 y S ⎞
∆Hc = (hf )pair ⎜ − y 2.75 ⎟ (425)
⎝ J'F ⎠
where,
J'FL
(hf )pair = (426)
100
and y can be approximated as in step 5 above (for F = 0.36)
7. After manipulating the equation a bit, the following expression is obtained:
−0.57
∆Hc = 8.9LS1.57 ( J' ) (427)
for ∆Hc in m; L in m; S in m/m; and J’ in m/100 m. Note that J’ and L are for
the pair of laterals, not only uphill or only downhill
III. Derivation of Equation for α
• The parameter α is used in the calculation of inlet pressure for a pair of

laterals on sloping ground where (Eq. 22.17):
⎛x ⎞
Hl = Ha + α (hf )pair + ⎜ − 0.5 ⎟ ( ∆he )pair (428)
⎝L ⎠
with,
1/ x
⎛q ⎞
Ha = ⎜ a ⎟ (429)
⎝ Kd ⎠
J'FL
(hf )pair = (430)
100
100S
( ∆he )pair = (hf )pair = SL (431)
J'F
• Note that (∆he)pair must be a negative number

• The ratio x/L is the distance to the manifold, where L is the length of the pair
of laterals
• The following derivation is based on equations presented by Keller and
Rodrigo (1979):
1. Given that for a single lateral approximately ¾ of the friction loss occurs
from the inlet to the point where the average pressure occurs (multiple
outlets, uniform outlet spacing, constant discharge from outlets, single
lateral pipe size) we have the following:
(hf )downhill ⎛⎜ ⎞⎟ + (hf )uphill ⎛⎜ 1 − ⎞⎟

3 x 3 x
α (hf )pair = (432)
4 ⎝L⎠ 4 ⎝ L⎠
The above equation is a weighted average because the uphill lateral is

shorter than the downhill lateral
2. Recall that,
2.75
⎛ hfx ⎞ ⎛x⎞
⎜ ⎟ =⎜ ⎟ (433)
⎝ hf ⎠pair ⎝ L ⎠
Then,
2.75
⎛x⎞
(hf )downhill =⎜ ⎟ (hf )pair
⎝L⎠
(434)
2.75
(hf )uphill = ⎛⎜ 1 −
x⎞
(hf )pair
⎝ L ⎟⎠
3. Combining equations:
3 ⎡⎛ x ⎞ ⎛ x ⎞2.75 ⎛ x ⎞⎛ x⎞
2.75 ⎤
α (hf )pair = (hf )pair ⎢⎜ ⎟ ⎜ ⎟ + ⎜1− ⎟ ⎜1− ⎟ ⎥ (435)
4 ⎢⎣⎝ ⎠ ⎝ ⎠
L L ⎝ L ⎠⎝ L ⎠ ⎥⎦
3 ⎡⎛ x ⎞
3.75 3.75 ⎤
⎛ x⎞
α = ⎢⎜ ⎟ + ⎜1− ⎟ ⎥ (436)
4 ⎢⎝ L ⎠ ⎝ L⎠ ⎥⎦
⎣
• This last equation for α is Eq. 22.25 from the textbook
• See the figure below
Lecture 23
Manifold Hydraulic Design
I. Introduction
• Manifolds in trickle irrigation systems often have multiple pipe sizes to:
1. reduce pipe costs

2. reduce pressure variations
• In small irrigation systems the reduction in pipe cost may not be significant,
not to mention that it is also easier to install a system with fewer pipe sizes
• Manifold design is normally subsequent to lateral design, but it can be part of
an iterative process (i.e. design the laterals, design the manifold, adjust the
lateral design, etc.)
• The allowable head variation in the manifold, for manifolds as subunits, is
given by the allowable subunit head variation (Eq. 20.14) and the calculated
lateral head variation, ∆Hl
• This simple relationship is given in Eq. 23.1:
( ∆Hm )a = ∆Hs − ∆Hl (437)
• Eq. 23.1 simply says that the allowable subunit head variation is shared by
the laterals and manifold
• Recall that a starting design point can be to have ∆Hl = ½∆Hs, and ∆Hm =
½∆Hs, but this half and half proportion can be adjusted during the design
iterations
• The lateral pressure variation, ∆Hl, is equal to the maximum pressure minus
the minimum pressure, which is true for single-direction laterals and
uphill+downhill pairs, if Hn’ is the same both uphill and downhill
II. Allowable Head Variation
• Equation 20.14 (page 502 in the textbook) gives the allowable pressure head
variation in a “subunit”
• This equation is an approximation of the true allowable head variation,
because this equation is applied before the laterals and manifold are
designed
• After designing the laterals and manifold,
the actual head variation and expected
EU can be recalculated
• Consider a linear friction loss gradient (no

multiple outlets) on flat ground:
In this case, the average head is
equal to Hn plus half the difference in the maximum and minimum heads:
(
H max − H n = 2 H a − H n ) (438)
• Consider a sloping friction loss gradient (multiple outlets) on flat ground:
In this case, the average head occurs

after about ¾ of the total head loss
(due to friction) occurs, beginning
from the lateral inlet. Then,
(
H max − H n = 4 H a − H n ) (439)
• For a sloping friction loss gradient (multiple outlets) on flat ground with dual
pipe sizes, about 63% of the friction head loss occurs from the lateral inlet to
the location of average pressure. Then 100/(100-63) = 2.7 and,
(
H max − H n = 2.7 H a − H n ) (440)
• In summary, an averaging is performed to skew the coefficient toward the

minimum value of 2, recognizing that the maximum is about 4, and that for
dual-size laterals (or manifolds), the coefficient might be approximately 2.7
• The value of 2.5 used in Eq. 20.14 is such a weighted average
• With three or four pipe sizes the friction loss gradient in the manifold will
approach the slope of the ground, which may be linear
• Thus, as an initial estimate for determining allowable subunit pressure
variation for a given design value of EU, Eq. 20.14 is written as follows:
(
∆Hs = 2.5 H a − H n ) (441)
• After the design process, the final value of ∆Hs may be different, but if it is
much different the deviation should be somehow justified
III. Pipe Sizing in Manifolds
• Ideally, a manifold design considers all of the following criteria:
1. economic balance between pipe cost (present) and pumping

costs (future)
2. allowable pressure variation in the manifold and subunit
3. pipe flow velocity limits (about 1.5 - 2.0 m/s)
• From sprinkler system design, we already know of various pipe sizing

methods
• These methods can also be applied to the design of manifolds
• However, the difference with trickle manifolds is that instead of one or two
pipe sizes, we may be using three or four sizes
• The manifold design procedures described in the textbook are:
1. Semi-graphical
2. Hydraulic grade line (HGL)
3. Economic pipe sizing (as in Chapter 8 of the textbook)
Semi-Graphical Design Procedure
• The graphical method uses “standard” head loss curves for different pipe
sizes and different flow rates with equally-spaced multiple outlets, each outlet
with the same discharge
• The curves all intersect at the origin (corresponding to the downstream
closed end of a pipe)
• Below is a sample of the kind of curves given in Fig. 23.2 of the textbook
• Instead of the standard curves, specific curves for each design case could be
custom developed and plotted as necessary in spreadsheets
• The steps to complete a graphical design are outlined in the textbook
• The graphical procedure is helpful in understanding the hydraulic design of
multiple pipe size manifolds, but may not be as expedient as fully numerical
procedures
3.8
3.6
3.4
3.2
3.0
Friction Head Loss (ft)
2.8
2.6
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.25 inch
1.0
1.50 inch
0.8
2.00 inch
0.6
2.50 inch
0.4 3.00 inch
0.2 4.00 inch
0.0
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320
Flow Rate (gpm)

• The following steps illustrate the graphical design procedure:
Step 1:
flow direction
(∆Hm)a
So ∆Em
1
0 manifold flow rate Qm
xd
Step 2:
flow direction
(∆Hm)a
So ∆Em
1
xd
Step 3:
flow direction
(∆Hm)a
So ∆Em
1
xd
Step 4:
flow direction
(∆Hm)a
So ∆Em
1
xd
Step 5:
flow direction
(∆Hm)a
So ∆Em
1
xd
Step 6:
flow direction
(∆Hm)a
So ∆Em
1
xd
HGL Design Procedure
• The HGL procedure is very similar to the graphical procedure, except that it
is applied numerically, without the need for graphs
• Nevertheless, it is useful to graph the resulting hydraulic curves to check for
errors or infeasibilities
• The first (upstream) head loss curve starts from a fixed point: maximum
discharge in the manifold and upper limit on head variation
• Equations for friction loss curves of different pipe diameters are known (e.g.
Darcy-Weisbach, Hazen-Williams), and these can be equated to each other
to determine intersection points, that is, points at which the pipe size would
change in the manifold design
• But, before equating head loss equations, the curves must be vertically
shifted so they just intersect with the ground slope curve (or the tangent to
the first, upstream, curve, emanating from the origin)
• The vertical shifting can be done graphically or numerically
Economic Design Procedure
• The economic design procedure is essentially the same as that given in

Chapter 8 of the textbook
• The manifold has multiple outlets (laterals or headers), and the “section flow
rate” changes between each outlet
• The “system flow rate” would be the flow rate entering the manifold
IV. Manifold Inlet Pressure Head
• After completing the manifold pipe sizing, the required manifold inlet pressure
head can be determined (Eq. 23.4):
Hm = Hl + k hf + 0.5∆Em (442)
where k = 0.75 for single-diameter manifolds; k = 0.63 for dual pipe size
laterals; or k ≈ 0.5 for three or more pipe sizes (tapered manifolds); and ∆El
is negative for downward-sloping manifolds
• As with lateral design, the friction loss curves must be shifted up to provide
for the required average pressure
• In the case of manifolds, we would like the average pressure to be equal to
the calculated lateral inlet head, Hl
• The parameter ∆El is the elevation difference along one portion of the
manifold (either uphill or downhill), with positive values for uphill slopes and
negative values for downhill slopes
V. Manifold Design
• Manifolds should usually extend both ways from the mainline to reduce the
system cost, provided that the ground slope in the direction of the manifolds
is less than about 3% (same as for laterals, as in the previous lectures)
• As shown in the sample layout (plan view) below, manifolds are typically
orthogonal to the mainline, and laterals are orthogonal to the manifolds
• Manifolds usually are made up of 2 to 4 pipe diameters, tapered
(telescoping) down toward the downstream end
• For tapered manifolds, the smallest of the pipe diameters (at the downstream
end) should be greater than about ½ the largest diameter (at the upstream
end) to help avoid clogging during flushing of the manifold
D1 D2 D3>0.5D1
• The maximum average flow velocity in each pipe segment should be less
than about 2 m/s
• Water hammer is not much of a concern, primarily because the manifold has
multiple outlets (which rapidly attenuates a high- or low-pressure wave), but
the friction loss increases exponentially with flow velocity
VI. Trickle Mainline Location
• The objective is the same as for pairs of laterals: make (Hn)uphill equal to
(Hn)downhill
• If average friction loss slopes are equal for both uphill and downhill manifold
branches (assuming similar diameters will carry similar flow rates):
Downhill side:
( ∆Hm )a = hfd − ∆E ⎛⎜
x⎞
⎟ = hfd − Y∆E (443)
⎝ ⎠
L
Uphill side:
L−x⎞
( ∆Hm )a = hfu + ∆E ⎛⎜ ⎟ = hfu + (1 − Y)∆E (444)
⎝ L ⎠
where x is the length of downhill manifold (m or ft); L is the total length of the
manifold (m or ft); Y equals x/L; and ∆E is the absolute elevation difference of
the uphill and downhill portions of the manifold (m or ft)
• Note that in the above, ∆E is an absolute value (always positive)

• Then, the average uphill and downhill friction loss slopes are equal:
Juphill = Jdownhill
hfu h (445)
= fd
L−x x
where J-bar is the average friction loss gradient from the mainline to the end
of the manifold (J-bar is essentially the same as JF)
Then,
hfd = Jx
(446)
hfu = J(L − x)
and,
( ∆Hm )a = Jx − Y∆E
(447)
( ∆Hm )a = J(L − x) + (1 − Y) ∆E
then,
( ∆Hm )a + Y∆E
=J
x (448)
( ∆Hm )a − (1 − Y) ∆E
=J
L−x
• Equating both J-bar values,
( ∆Hm )a + Y∆E ( ∆Hm )a − (1 − Y) ∆E

= (449)
x L−x
• Dividing by L and rearranging (to get Eq. 23.3),
( ∆Hm )a + Y∆E ( ∆Hm )a − (1 − Y) ∆E

= (450)
Y 1− Y
or,
∆E 2Y − 1
= (451)
( ∆Hm )a 2Y(1 − Y)
• Equation 23.3 is used to determine the lengths of the uphill and downhill
portions of the manifold
• You can solve for Y (and x), given ∆E and (∆Hm)a = ∆Hs - ∆Hl
• Remember that ∆Hs ≈ 2.5(Ha – Hn), where Ha is for the average emitter and
Hn is for the desired EU and νs
• Equation 23.3 can be solved by isolating one of the values for Y on the left
hand side, such that:
⎛ 2Y − 1 ⎞ ⎛ ( ∆Hm ) ⎞
Y = 1− ⎜ ⎟⎜ a
⎟⎟ (452)
⎜ 2Y ⎟ ⎜ ∆ E
⎝ ⎠⎝ ⎠
and assuming an initial value for Y (e.g. Y = 0.6), plugging it into the right
side of the equation, then iterating to arrive at a solution
• Note that 0 ≤ Y ≤ 1, so the solution is already well-bracketed
• Note that in the trivial case where ∆E = 0, then Y = 0.5 (don’t apply the above
equation, just use your intuition!)
• A numerical method (e.g. Newton-Raphson) can also be used to solve the
equation for Y
VII. Selection of Manifold Pipe Sizes
The selection of manifold pipe sizes is a function of:
1. Economics, where pipe costs are balanced with energy costs

2. Balancing hf, ∆E, and (∆Hm)a to obtain the desired EU
3. Velocity constraints
VIII. Manifold Pipe Sizing by Economic Selection Method
• This method is similar to that used for mainlines of sprinkler systems

• Given the manifold spacing, Sm, and the manifold length, do the following:
(a) Construct an economic pipe size table where Qs = Qm

(b) Select appropriate pipe diameters and corresponding Q values at locations
where the diameters will change
(c) Determine the lengths of each diameter of pipe (where the Q in the manifold
section equals a breakeven Q from the Economic Pipe Size Table (EPST)
⎛ Qbeg − Qend ⎞
LD = L ⎜ ⎟ (453)
⎝ Q m ⎠
where Qbeg is the flow rate at the beginning of diameter “D” in the EPST (lps
or gpm); Qend is the flow rate at the end of diameter “D” in the EPST, which is
the breakeven flow rate of the next larger pipe size) (lps or gpm); L is the
total length of the manifold (m or ft); and Qm is the manifold inflow rate (lps or
gpm). (see Eq. 23.7)
(d) Determine the total friction loss along the manifold (see Eq. 23.8a):
FLK ⎛ Q1a Qa2 − Q1a Q3a − Qa2 Qa4 − Q3a ⎞

hf = ⎜ + + + ⎟ (454)
100Qm ⎜⎝ D1c Dc2 D3c Dc4 ⎟⎠
where,
a= b+1 (for the Blasius equation, a = 2.75)

c= 4.75 for the Blasius equation (as seen previously)
Q1 = Q at the beginning of the smallest pipe diameter
Q2 = Q at the beginning of the next larger pipe diameter
Q3 = Q at the beginning of the third largest pipe diameter in the manifold
Q4 = Q at the beginning of the largest pipe diameter in the manifold
F= multiple outlet pipe loss factor
• For the Hazen-Williams equation, F equals 1/(1.852+1) = 0.35

• For the Darcy-Weisbach equation, F equals 1/(2+1) = 0.33
L= the total length of the manifold

D= inside diameter of the pipe
K= 7.89(10)7 for D in mm, Q in lps, and length in m
K= 0.133 for D in inches, Q in gpm, and length in ft
hf = friction head loss
• The above equation is for four pipe sizes; if there are less than four sizes, the
extra terms are eliminated from the equation
• An alternative would be to use Eq. 23.8b (for known pipe lengths), or
evaluate the friction loss using a computer program or a spreadsheet to
calculate the losses section by section along the manifold
• Eq. 23.8b is written for manifold design as follows:
a −1 ⎛ a
FK Qm x1 x a2 − x1a x 3a − x a2 x a4 − x 3a ⎞
hf = ⎜ + + + ⎟ (455)
100La −1 ⎜⎝ D1c Dc2 D3c Dc4 ⎟⎠
where, x1 = length of the smallest pipe size

x2 = length of the next smaller pipe size
x3 = length of the third largest pipe size
x4 = length of the largest pipe size
• Again, there may be up to four different pipe sizes in the manifold, but in
many cases there will be less than four sizes
(e) For s ≥ 0 (uphill branch of the manifold),
∆Hm = hf + S xu (456)
For s < 0 (downhill branch of the manifold),
⎛ 0.36 ⎞
∆Hm = hf + S ⎜ 1 − xd
n ⎟⎠
(457)
⎝
where n is the number of different pipe sizes used in the branch; and S is the
ground slope in the direction of the manifold (m/m)
• The above equation estimates the location of minimum pressure in the

downhill part of the manifold
(f) if ∆Hm < 1.1 (∆Hm)a, then the pipe sizing is all right. Go to step (g) of this
B B B B B B
procedure. Otherwise, do one or more of the following eight adjustments:
(1) Increase the pipe diameters selected for the manifold
• Do this proportionately by reselecting diameters from the EPST using a

larger Qs (to increase the energy “penalty” and recompute a new
B B
EPST). This will artificially increase the break-even flow rates in the
table (chart).
• The new flow rates to use in re-doing the EPST can be estimated for s >
0 as follows:
1/ b
⎛ hf ⎞
Qnew old
= Qs ⎜ ⎟ (458)
s ⎜ ( ∆Hm ) − ∆Em ⎟
⎝ a ⎠
and for s < 0 as:
1/ b
⎛ ⎞
⎜ hf ⎟
old ⎜ ⎟
Qnew
s = Qs (459)
⎜ ∆H ⎛ 0.36 ⎞ ⎟
⎜ ( m )a − ∆El ⎜ 1 − n ⎟ ⎟
⎝ ⎝ ⎠⎠
•The above two equations are used to change the flow rates to compute
the EPST
• The value of Qm remains the same B B
• The elevation change along each manifold (uphill or downhill branches)

is ∆El = sL/100
(2) Decrease Sm B B
• This will make the laterals shorter, Qm will decrease, and ∆Hl may B B B B
decrease
• This alternative may or may not help in the design process
(3) Reduce the target EU

• This will increase ∆Hs B B
(4) Decrease ∆Hl (use larger pipe sizes) B B
• This will increase the cost of the pipes
(5) Increase Ha B B
• This will increase ∆Hs B B
• This alternative will cost money and or energy
(6) Reduce the manufacturer’s coefficient of variation

• This will require more expensive emitters and raise the system cost
(7) Increase the number of emitters per tree (Np) B B
• This will reduce the value of νs B B
(8) If Ns > 1, increase Ta per station

B B B B
• Try operating two or more stations simultaneously
• Now go back to Step (b) and repeat the process.
(g) Compute the manifold inlet head,
Hm = Hl + khf + 0.5∆Em (460)
where, k = 0.75 for a single size of manifold pipe

k = 0.63 for two pipe sizes
k = 0.50 for three or more sizes
• For non-critical manifolds, or where ∆Hm < (∆Hm)a, decrease Qs (or just B B B B B B B B
design using another sizing method) in the Economic Pipe Selection

Table to dissipate excess head
• For non-rectangular subunits, adjust F using a shape factor:
Fs = 0.38S1.25
f + 0.62 (461)
where Sf = Qlc/Qla; Qlc is the lateral discharge at the end of the manifold
B B B B B B B B
and Qla is the average lateral discharge along the manifold. Then,
B B
⎛ JL ⎞
hf = Fs F ⎜ ⎟
⎝ 100 ⎠ (462)
IX. Manifold Pipe Sizing by the “HGL” Method
• This is the “Hydraulic Grade Line” method

• Same as the semi-graphical method, but performed numerically
(a) Uphill Side of the Manifold
• Get the smallest allowable pipe diameter and use only the one diameter
for this part of the manifold
(b) Downhill Side of the Manifold
Largest Pipe Size, D1 B B
• First, determine the minimum pipe diameter for the first pipe in the
downhill side of the manifold, which of course will be the largest of the
pipe sizes that will be used
• This can be accomplished by finding the inside pipe diameter, D, that will
give a friction loss curve tangent to the ground slope
• To do this, it is necessary to: (1) have the slope of the friction loss curve
equal to So; and, (2) have the H values equal at this location (make them
B B
just touch at a point)

• These two requirements can be satisfied by applying two equations,
whereby the two unknowns will be Q and D1 B B
• Assume that Ql is constant along the manifold…

B B
• See the following figure, based on the length of the downstream part of
the manifold, xd B B
• Some manifolds will only have a downhill part – others will have both
uphill and downhill parts
H flow direction
(∆Hm)a
hf
e
rv
D1 = ???
cu
ss
lo
i on
fr ict
So ∆Em
1
0
xd
• For the above figure, where the right side is the mainline location and the
left side is the downstream closed end of the manifold, the friction loss
curve is defined as:
JFL
H = ( ∆Hm )a + ∆Em − hf + (463)
100
where, using the Hazen-Williams equation,
1.852
⎛Q⎞
J = K⎜ ⎟ D−4.87 for 0 ≤ Q ≤ Qm (464)
⎝C⎠
1 1 0.852
F= + + (465)
2.852 2N 6N2
⎛ x ⎞⎛ Q ⎞
N = ⎜ d ⎟⎜ ⎟ for N > 0 (466)
⎝ Sl ⎠ ⎝ Qm ⎠
where N is the number of outlets (laterals) from the location of “Q” in the
manifold to the closed end
⎛ Q ⎞
L = xd ⎜ ⎟ (467)
⎝ Qm ⎠
For Q in lps and D in cm, K = 16.42(10)6 P P
• The total head loss in the downhill side of the manifold is:
1.852
J F x ⎛Q ⎞
hf = hf hf d = 0.01K ⎜ m ⎟ D−4.87Fhf x d (468)
100 ⎝ C ⎠
where Fhf is defined as F above, except with N = xd/Sl.

B B B B B B
• The slope of the friction loss curve is:
dH 1 ⎛ dJ dF dL ⎞
= ⎜ FL + JL + JF ⎟ (469)
dQ 100 ⎝ dQ dQ dQ ⎠
where,
dJ 1.852KQ0.852
= (470)
dQ C1.852D4.87
dF xd ⎛ 1 0.852 ⎞
=− ⎜ + ⎟ (471)
dQ SlQmN2 ⎜⎝ 2 3N ⎟⎠
dL x
= d (472)
dQ Qm
• Note that dH/dQ ≠ J

• The ground surface (assuming a constant slope, So) is defined by: B B
⎛ Q ⎞
H = SoL = So x d ⎜ ⎟ (473)
⎝ Qm ⎠
and,
dH So x d
= (474)
dQ Qm
• Combine the two equations defining H (this makes the friction loss curve
just touch the ground surface):
⎛ Q ⎞ JFL
So x d ⎜ ⎟ = ( ∆Hm )a + ∆Em − hf + (475)
⎝ Qm ⎠ 100
• Solve the above equation for the inside diameter, D:
−0.205
⎡ 1.852 ⎛ So x dQ ⎞⎤
⎢ 100C ⎜ − ( ∆H )
m a − ∆ Em ⎟⎥
D= ⎢ ⎝ Qm ⎠⎥
( )
(476)
⎢ K Q 1.852 1.852
FL − Qm Fhf x d ⎥
⎢ ⎥
⎣⎢ ⎦⎥
• Set the slope of the friction loss curve equal to Soxd/Qm,

B B B B B B
So x d 1 ⎛ dJ dF dL ⎞
= ⎜ FL + JL + JF ⎟ (477)
Qm 100 ⎝ dQ dQ dQ ⎠
• Combine the above two equations so that the only unknown is Q (note: D
appears in the J & dJ/dQ terms of the above equation)
• Solve for Q by iteration; the pipe inside diameter, D, will be known as part
of the solution for Q
• The calculated value of D is the minimum inside pipe diameter, so find the
nearest available pipe size that is larger than or equal to D:
D1 ≥ D & minimize(D1 − D) (478)
Slope of the Tangent Line B
• Now calculate the equation of the line through the origin and tangent to
the friction loss curve for D1 B B
• Let St be the slope of the tangent line

B B
⎛ Q ⎞
H = StL = St x d ⎜ ⎟ (479)
⎝ Qm ⎠
then,
⎛ Q ⎞ JFL
St x d ⎜ ⎟ = ( ∆Hm )a + ∆El − hf + (480)
⎝ Qm ⎠ 100
• Set the slope of the friction loss curve equal to Stxd/Qm, B B B B B B
St x d 1 ⎛ dJ dF dL ⎞
= ⎜ FL + JL + JF ⎟ (481)
• Combine the above two equations to eliminate St, and B B
solve for Q (which is different than the Q in Eq. 476)

• Calculate the slope, St, directly B B
Smaller (Downstream) Pipe Sizes B
• Then take the next smaller pipe size, D2, and make its B B
friction loss curve tangent to the same line (slope = St); B B
JFL
H = H0 + (482)
100
where H0 is a vertical offset to make the friction loss curve tangent to the
B B
St line, emanating from the origin

B B
• Equating heads and solving for H0, B B
⎛ Q ⎞ JFL
H0 = St x d ⎜ ⎟− (483)
⎝ Qm ⎠ 100
• Again, set the slope of the friction loss curve equal to St, B B
St x d 1 ⎛ dJ dF dL ⎞
= ⎜ FL + JL + JF ⎟ (484)
• Solve the above equation for Q, then solve directly for H0 B B
• Now you have the equation for the next friction loss curve
• Determine the intersection with the D1 friction loss curve to set the lengthB B
for size D1; this is done by equating the H values for the respective
B B
equations and solving for Q at the intersection:
1.852
Hbig − Hsmall +
FLK ⎛ Q ⎞
⎜ ⎟
100 ⎝ C ⎠
(Dbig
−4.87 −4.87
− Dsmall )=0 (485)
where, for the first pipe size (D1): B B
Hbig = ( ∆Hm )a + ∆El − hf (486)
and for the second pipe size (D2): B B
Hsmall = H0 (487)
and F & L are as defined in Eqs. 437 to 439.
• Then, the length of pipe D1 is equal to:

B B
⎛ Q ⎞
LD1 = x d ⎜ 1 − ⎟ (488)
⎝ Qm ⎠
• Continue this process until you have three or four pipe sizes, or until you
get to a pipe size that has D < ½D1 B B
Comments about the HGL Method B
• The above equation development could also be done using the Darcy-
Weisbach equation
• Specify a minimum length for each pipe size in the manifold
so that the design is not something ridiculous (i.e. don’t just
blindly perform calculations, but look at what you have)
• For example, the minimum allowable pipe length might be
something like 5Sl
B B
• Note that the friction loss curves must be shifted vertically

upward to provide the correct average (or minimum, if pressure regulators
are used) pressure head in the manifold; this shifting process determines
the required manifold inlet pressure head, Hm
B B
• Below is a screen shot from a computer program that uses the HGL
method for manifold pipe sizing
Lecture 24
Hydraulic Design of Mainline & Supply Line
I. Introduction
• Chapter 24 of the textbook contains a good summary and discussion of the

design process for trickle irrigation systems
• Keller & Bliesner divide the trickle irrigation design process into three phases:
1. Phase 1: Planning factors, gross application depth,

preliminary system capacity, filtration requirements, etc.
2. Phase 2: Hydraulic design of laterals, manifolds, and

subunit layout.
3. Phase 3: Hydraulic design of main and supply lines,

control head, and pumping plant.
II. Factors Affecting the EU
• Solomon (1985) studied the various factors affecting emission

uniformity and ranked them according to importance using model
studies and field surveys. This is his ranking:
1. Plugging of emitters and other system components

2. Number of emitters per plant, Np
B B
3. Emitter coefficient of variation, ν

4. Emitter exponent, x
5. Emitter discharge variation with temperature
6. Pressure head variations in subunits
7. Coefficient of variation on pressure regulators
8. Friction loss from manifolds to laterals
9. Number of pipe sizes in manifolds
• In general, plugging (or the lack thereof) has the greatest

influence on EU, except in cases where the system design is
very poor
• Plugging cannot be prevented by design only -- it is up to the
management to continuously maintain the system as necessary
• The above ranking implies that a poorly designed system might
perform better (in terms of application uniformity) than a well
designed system, provided that the system operation and
maintenance is given due attention and effort
III. Mainline, Supply, and Control Head Design
• After designing laterals and manifolds, the mainline, supply line

(if necessary), control head, and pump must be
designed/selected. This involves calculating the TDH at the
pump.
• Maximum system capacity would have already been determined
in the preceding design steps, based on the number of subunits,
their sizes, emitter discharge and spacing, and number of
emitters per plant.
• Subunits should be designed to have nearly the same discharge
so that the pump is not wasting energy in lower capacity
subunits; but this is not always possible.
• Mainline design is the same as in sprinkle systems, and various
pipe selection methods are available (including the economic
pipe selection method).
• The control head includes the filters, sand separators, chemical
injection equipment, flow rate and or volumetric meters, timers,
pressure gauges, and/or other hardware. Most of this equipment
is located in the same place in a trickle irrigation system.
However, other filters, screens, and gauges may be installed at
downstream locations in some cases.
• Sand media and other filters may have a combined head loss of
10-20 psi when “dirty”, and 3-10 psi when clean. However, if too
much flow is forced through the filters (i.e. not enough filter
capacity) the head loss can be higher, even when clean.
• Flow meters and chemical injection equipment may have1-5 psi
loss, and valves can have up to several psi loss (even when fully
open).
• The pressure changes from the control head to the subunits are
due to a combination of friction loss and elevation change: if the
subunit inlet is at a higher elevation than the control head, both
friction and elevation change contribute to a higher required
pressure at the control head, otherwise they tend to cancel out
(partially or completely).
• Mainline friction losses include minor (local) losses at bends,
through valves, and through screens.
• Losses at reductions in pipe diameter (in series) are usually
small, unless the reduction is sharp and abrupt. Diameter
transitions for PVC are often smooth.
• A “critical” subunit can be defined by calculating the combined
friction loss and elevation change to each subunit, and taking the
highest value plus the required inlet pressure head to the
manifold. The textbook calls this (Hm + Hfe)c.
B B B B B B
• The critical subunit will define the “worst case” for which the
pumping unit should be designed. It may then be possible to
reduce pipe sizes in other subunits if they will have excess head
available.
IV. Trickle System Mainline Design
• Trickle mainlines can be sized using the same approach as is

used for sprinkle irrigation systems.
• Manifolds can be considered as stationary sprinkler laterals.
• You can use the EPST approach, where Ot depends (in part) on B B
the number of stations, Ns B B
Example calculation:
• Size sections A-B, B-C and C-D for a trickle irrigation mainline
having three manifolds
• First, decide on the number of stations, Ns B B
This could be 1, 3 or 6:
1. All laterals at the same time, or

2. Only laterals on one manifold at a time, or
3. Only laterals on one half of a manifold (on one side
of the mainline) at a time
• Set up a table to see the effect of Ns on the flow rates and Ot per B B B B
station (also see the figure below):
Number of Stations
1 2 3
Qs (lps) B B 30 30 30
Ot/station (hrs)
B B 1000 330 165
Ot, CD B B 1000 330 330
QCD B B 10 30 30
Ot, BC B B 1000 660 660
QBC B B 20 30 30
Ot, AB B B 1000 1000 1000
QAB B B 30 30 30
• Ot, CD, for example, is the number of hours that water is flowing in
B B
section CD per season

• This is the Ot that would be used in creating an economic pipe
B B
sizing table to determine the pipe size of section CD

• Ot varies with the number of stations and with the location of the
B B
mainline segment in the system

• Therefore, the EPST must be developed several times (this is
easy in a spreadsheet, as you have seen)
• Notice that the product of Q and Ot is a constant, as this is

B B
proportional to the irrigation water requirement, which is the

same for each station and location in the field
• Pipe diameters are selected using QAB, QBC, and QCD as the
B B B B B B
break-even flow rates in the EPST; Qs remains constant

B B
The TDH is the sum of:
1. static lift
2. well losses (if applicable)
3. supply line losses (elevation and friction)
4. control head pressure losses
5. losses to the critical subunit plus inlet pressure, (Hm + B B
Hfe)c
B B B B
6. screen and valve losses (if applicable) at subunit inlets
• As in many hydraulic designs for pressurized pipe systems, it is

often recommendable to add a “safety factor” to the losses,
because losses are not precisely known, and they will probably
increase with time
• A common safety factor is to add 10% to the friction losses for
the calculation of TDH
• A separate safety factor can be added, in some cases, to the
TDH for compensation for emitter plugging and degradation
• See sample calculations of TDH in Chapter 24
VI. We Live in an Imperfect World
• In many trickle systems, the ground slope

changes significantly between subunits, or
within subunits
• This complicates the system design and may
require more pressure regulation and larger
pipe sizes than would otherwise be necessary
• In many locations and countries the available

pipes, fittings, emitters, filters, and other
hardware are very limited
• Therefore, innovation, resourcefulness, and
improvisation may be very important
• If hardware is very limited, it may be best to consider another
type of irrigation system
• The system capacity for a micro-irrigation system can have a

safety factor added on to account for the possibility of:
1. slow, partial clogging of emitters and laterals

2. changes in crop type
3. inaccurate estimations of peak crop ET
4. more system “down time” than originally anticipated
5. various other factors
• It is desirable to minimize the hardware cost of an irrigation

system, but the cost of having an insufficient system capacity
may be many times higher than the marginal cost of larger pipes,
filters and valves
• When in doubt, it may be a good idea to increase the calculated
system capacity by 10 or 20%
• See Chapter 25 of the textbook for a thoughtful discussion of

factors in selecting a sprinkler or trickle system, or for selecting
another type of irrigation system

Sprinkle & Trickle Irrigation Lecture Notes PDF

Uploaded by

Copyright:

Available Formats

Sprinkle & Trickle Irrigation Lecture Notes PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sprinkle & Trickle Irrigation Lecture Notes PDF

Uploaded by

Copyright:

Available Formats

Sprinkle & Trickle Irrigation

Biological and Irrigation Engineering Department

This material has been duplicated by special permission of the copyright

© Copyright 1990 to 2004

Sprinkle & Trickle Irrigation Lectures Page 3 Merkley & Allen

Note: Equations are numbered consecutively in these lecture

Sprinkle & Trickle Irrigation Lectures Page 5 Merkley & Allen

1 ppm ≈ 1 mg/liter (usually)

PVC = Polyvinyl chloride

Sprinkle & Trickle Irrigation Lectures Page 7 Merkley & Allen

• Design of sprinkle and trickle systems – perhaps the most comprehensive

II. Textbook and Other Materials

• Textbook by Keller and Bliesner

III. Homework and Design Project

• Work must be organized and neat

IV. Tests, Quizzes, and Grading Policy

• Maybe some quizzes (these will not be announced)

• It is often necessary to convert units in design calculations

VI. Irrigation Systems

• On-farm level (field)

Sprinkle & Trickle Irrigation Lectures Page 9 Merkley & Allen

Type U.S. Area World Area

VIII. Sprinkler Systems

1. effective use of small continuous streams of water

1. initial cost can be high (compared to surface irrigation systems) at $500 to

IX. Slides of Sprinkler Systems

[these will be shown in class]

Merkley & Allen Page 10 Sprinkle & Trickle Irrigation Lectures

I. Sprinkler System Categories

• Two broad categories: set and continuous-move

II. Set Systems:

• very common type of sprinkler system

• similar to hand-move, but pipes are more rigidly connected

• very common in western U.S.

Sprinkle & Trickle Irrigation Lectures Page 11 Merkley & Allen

• 5/8-inch (16 mm) or larger nozzles

Other Set Sprinklers

Fixed (Solid-Set) Systems

• enough laterals to cover entire field at same time

Merkley & Allen Page 12 Sprinkle & Trickle Irrigation Lectures

III. Continuous-Move Systems

• could be big gun or boom on platform with wheels

• costs about $40,000 for 400 m of lateral

IV. LEPA Systems

• Low Energy Precision Application (LEPA) is a concept developed in the mid

Merkley & Allen Page 14 Sprinkle & Trickle Irrigation Lectures

II. Irrigation Interval

• The maximum irrigation frequency is:

where fx is the maximum interval (frequency) in days; and Ud is the average

• The range of fx values for agricultural crops is usually:

0.25 < fx < 80 days (3)

Sprinkle & Trickle Irrigation Lectures Page 15 Merkley & Allen

III. Peak Use Period

• Irrigation system design is usually for the most demanding conditions:

IV. Leaching Requirement

where LR is the leaching requirement; ECw is the EC of the irrigation water

• Equation 4 is taken from FAO Irrigation and Drainage Paper 29

Sprinkle & Trickle Irrigation Lectures Page 17 Merkley & Allen

See Eq. 5.3 from the textbook regarding nonuniformity losses.