UNIVERSITY OF CALIFORNIA • COLLEGE OF AGRICULTURE 
 
 AGRICULTURAL EXPERIMENT STATION 
 
 BERKELEY 4, CALIFORNIA 
 
 CIRCULAR 362 
 
 June, 1945 
 
 FARM IRRIGATION STRUCTURES 
 
 C. N. JOHNSTON* 
 
 The subject of farm irrigation structures includes irrigation ditches or 
 canals, irrigation piping systems, and the necessary controlling devices for 
 applying the transmitted water. Since design and installation procedure for 
 the canals have been carefully considered in hydraulics texts (1, 2, 8) 2 it need 
 only be summarized here. Many controlling devices, especially those used in 
 ditches, are not new, but have been developed in various districts of the West. 
 As a result, colloquialism has often applied several names to the same device. 
 Moreover, since a device may be viewed in different lights, the names were 
 not always synonymous. For example, one structure used to hold back a 
 portion of the flow in a ditch, keeping the water levels higher upstream, may 
 be called a check by an observer on the upstream side. The same person, 
 viewing the same structure downstream, might call it a drop because the 
 downstream water levels were below those of the upstream. Because of this 
 confusion, irrigation structures will be described here by their design details 
 and not particularly by name. 
 
 The discussion considers these structures in the following order: (1) open 
 channels such as canals or ditches; (2) ditch or canal structures and supple- 
 mental equipment; (3) pipe lines; (4) control devices in pipe lines. 
 
 Canals and ditches are the simplest and oldest means of transporting water 
 in considerable volume. A discussion of the laws that govern flow in them will 
 be useful later when considering flow in pipe lines. The second subject, ditch 
 or canal structures and supplemental equipment, can logically be taken up 
 after the ditches. Since these structures comprise a rather wide variety, exam- 
 ples of economical design in the range of structural materials will be presented 
 so an operator can select a unit according to his needs. Pipe lines behave quite 
 like ditches when water is being carried through them. Their points of su- 
 periority over the ditch, together with certain shortcomings, will be shown by 
 comparison to aid the operator in deciding which carrier system to install. The 
 last subject, control devices in pipe lines, covers the aids needed to make a 
 pipe line work efficiently. 
 
 1 Assistant Professor of Irrigation and Associate Irrigation Engineer in the Experiment 
 Station. 
 
 2 Italic numbers in parentheses refer to "References for Further Reading," at the end 
 of this circular. 
 
 [1] 
 
2 California Experiment Station Circular 362 
 
 CANALS AND DITCHES 
 
 Canals and ditches are one method of transmitting water from source of 
 supply to area of use. To transmit water they must have a slope or fall toward 
 the destination. The greater the slope, the higher the velocity. If such channels 
 are lined with nonscouring materials, they might be run down the side of a 
 mountain. Under these conditions, the velocities would approach those of a 
 free-falling body in the air. The flow speeds would be extremely unmanageable ; 
 and the water would swirl and tumble upon itself fiercely. In fact, ditch grades 
 must be restricted so that the motion will be gentle and not swirling. Calm, 
 
 TABLE 1 
 Maximum Mean Velocities Safe Against Erosion 
 
 Material 
 
 Very light pure sand of quicksand character . 
 
 Very light loose sand 
 
 Coarse sand or light sandy soil 
 
 Average sandy soil 
 
 Sandy loam 
 
 Average loam, alluvial soil, volcanic ash soil . 
 
 Firm loam, clay loam 
 
 Stiff clay soil, ordinary gravel soil 
 
 Coarse gravel, cobbles, shingles , 
 
 Mean velocity, 
 
 in 
 feet per second 
 
 0.75 to 1.00 
 1.00 to 1.50 
 1.50 to 2.00 
 2.00 to 2.50 
 2.50 to 2.75 
 2.75 to 3.00 
 3.00 to 3.75 
 4.00 to 5.00 
 5.00 to 6.00 
 
 steady flow is more important in unlined ditches, where the swirling action 
 cuts the bottom and sides (wetted surfaces), destroying the shape and even- 
 tually the grade. Even the lined channels, however, are cut by solids moving 
 in the stream; and this cutting becomes more rapid as the velocities are in- 
 creased. Primarily because destructive erosion does result from high Velocity 
 in canals and ditches, the grades are generally kept low (table 1). The quan- 
 tity of water carried by a ditch is directly proportional to the average velocity. 
 The necessity of keeping grades down demands, therefore, larger ditches than 
 if higher velocities were permissible. Size, in turn, increases the cost and 
 accentuates the advisability of accurate design. 
 
 The design of these channels is relatively simple, provided available data 
 are utilized (1, / h 5). Grade or slope, as already noted, influences velocity of 
 flow. Another influence is smoothness: the water "drags" on a rough channel 
 and glides more easily on a slick, smooth one. This slowing effect occurs only 
 along the line of contact between water and channel face, or on the "wetted 
 perimeter." A given volume, passing through a ditch at a fixed rate, will have 
 a given depth. If the width of the ditch is doubled and the grade is kept the 
 same as before and the condition of the lining is the same as in the original 
 ditch, certain changes in the flow occur. Though the same amount of water is 
 passing, the velocity and the depth of flow have changed. The depth is not 
 half the former depth, nor is the velocity the same as before, though the slope 
 
Farm Irrigation Structures 3 
 
 is the same. This change in velocities and depths results from the drag on the 
 water caused by the wider ditch. In other words, the length of the surface 
 (ditch width) of the conduit in contact with the water at any given cross 
 section has been increased ; and that change has altered the velocity and the 
 depth. This surface of contact measured at a stream cross section is called the 
 wetted perimeter and is given the mathematical symbol P. If, in this same 
 widened ditch, sufficient water is admitted to double the depth of flow, the 
 velocity will be about doubled. Here, again, is a new situation — a third pos- 
 sible velocity with constant slope. 
 
 w 
 
 Initial dirch sech'on 
 
 2W 
 
 Dihch wif"h double widl"h 
 
 Dirch wirh double widrh and double depfh 
 Fig. 1. — Ditch cross sections illustrating the effect of varying bottom width and depth. 
 
 Figure 1 illustrates the three conditions listed. The upper ditch in this 
 figure is the initial ditch section, with the water-surface level shown. In the 
 center detail, the ditch has been doubled in bottom width, or dimension w 
 has become 2w; and the same amount of water flowing requires a little more 
 than half the depth of the initial ditch section. The ditch detailed in the lower 
 third of figure 1 illustrates the same widened ditch (width equals 2w) but with 
 water twice as deep as in the one just above. As can be seen, doubling the 
 depth of water in the widened ditch has not greatly increased the length of 
 wetted cross section or wetted perimeter in the ditch. As a result, a much- 
 increased cross-section area of water is moving over only a slightly increased 
 wetted perimeter. In other words, the drag has not gone up much, but the 
 area of flow has more than doubled, while the slope of the ditch bottom re- 
 mained constant. Tests of ditches similar to those in figure 1 would demon- 
 
4 
 
 California Experiment Station Circular 362 
 
 strate that the average velocity in the ditch shown in the lower detail would 
 be more than twice the velocity found for the ditch detailed in the drawing 
 above it. The situation may be stated thus: the drag effect in the two ditches 
 is similar, approximating equality; but the cross section of flow has so in- 
 creased in the lower one that a greater body of water moves along in it at a 
 distance from this surface of drag. Being farther away, this larger body of 
 
 Side slope I hoi 
 E»D 
 
 Side slope \j fe I 
 E=«iD 
 
 Side slope 2 to i 
 E= 2D 
 
 Side slope 3 hoi 
 E=3D 
 
 Fig. 2. — Typical ditch sections. 
 
 water moves more freely and faster than is possible where the depth of flow 
 is shallow, as in the middle detail of the figure. Such are the conditions in 
 actual ditches. In studies made on ditch flow, such investigators as Ganguillet 
 and Kutter (6) have found that the cross-section area (represented by the 
 symbol A) of the stream divided by the wetted perimeter, P, gives a value, R, 
 that can be used to interpret the average value of velocity from the slope for 
 
 any given channel. The formula would be — = R. R is called the hydraulic 
 
 radius, and is a mathematical term applied to flow in all channels and pipes. 
 When the average or mean velocity is known, the volume of water moving 
 = A X V; or area in square feet, times velocity in feet per second, equals the 
 
Farm Irrigation Structures 5 
 
 quantity in cubic feet per second — sometimes represented as Q. The drag of 
 the wetted surface is another variable. Every obstruction (brush, grass, roots, 
 stones, sand or gravel on the bottom or in the lining material of a channel) 
 resists the motion of the flowing stream. Even the smoothest surfaces create 
 some drag through adhesion of the water itself. Any obstruction tends to re- 
 duce the velocity along the surface of contact. The term used to identify the 
 obstructing effect of the surface is its "roughness," symbolized for mathe- 
 matical calculations by the letter n. The value n becomes part of an equation 
 that expresses the flow relation for any given ditch. 
 
 Unfortunately, most ditches and canals are not simple square-cornered 
 channels whose cross-section area is the product of width and depth. As a rule 
 they have sloped sides and a flat bottom (fig. 2). Both bottom width and 
 
 TABLE 2 
 Ditch Cross-Section Relations Developed from Figure 2 
 
 Side slope ratio, 
 EtoD 
 
 Value E 
 
 Length F 
 
 Wetted perimeter, 
 P = F + B + F 
 
 A, Area, in 
 square feet 
 
 1 to 1 
 
 D 
 
 2D 
 3D 
 
 D\/2 = 1.414 X D 
 Dy/sJE = 1.803 X D 
 D^/l = 2.236 X D 
 Z^VlO = 3.162 X D 
 
 B + 2.828 X D 
 B + 3.606 X D 
 B + 4.472 X D 
 B + 6.324 X D 
 
 (B + D) X D 
 
 \y 2 to\ 
 
 (B + iy 2 D) X D 
 
 2 to 1 
 
 {B + 2D) X D 
 
 3 to 1 
 
 (B + 3D) X D 
 
 
 slope of sides vary in different ditches. The solutions, however, for cross- 
 section area and wetted perimeter are relatively simple even for these some- 
 what complex shapes. Lined ditches and canals can have side slopes as great as 
 45 degrees (1 to 1) or slightly steeper. Unlined earthen channels cannot retain 
 their shape if steep sides are constructed : the wetted sides will slump by their 
 own weight till the bank becomes stable. This slumping fills the channel so that 
 the flow capacity decreases. To eliminate this deterioration, the constructed 
 banks are given the flatter slopes initially. Clays can stand up with 13^ to 1 
 slopes (fig. 2) ; but sandy materials require as much as 3 to 1 slopes if traversed 
 by ditches, where dimension E (fig. 2) is three times dimension D. Other soil 
 types require ditch bank slopes between these extremes. 
 
 Regardless of the slope used in the ditch sides, the cross-section area of the 
 stream in square feet is approximately equal to half the width of the bottom 
 plus half the width of the stream surface multiplied by the depth. In the upper 
 left detail of figure 2, dimension E equals dimension D, because the slope is 
 1 to 1 ; so the surface of the stream = E + B + E = D+B + D=B + 2D. 
 Now if the width of the bottom is added to the width of the surface, the result 
 is B + 2D + B = 2B + 2D. If these terms are divided by 2, the result is 
 
 27? -4- 2D 
 
 = B + D, the average width. Multiply this by the depth to get the 
 
 A 
 
 area, A = (B + D)D. If the slope were V/2 to 1 instead of 1 to 1, as in 
 figure 2 (upper right hand), then E would equal l\^D and the area A would 
 
6 California Experiment Station Circular 362 
 
 be (B + 1}/2D)D by the same reasoning as above. Similarly, if the side slopes 
 were 3 to 1, then E = W and A = (B + SD)D (see table 2). 
 
 The wetted perimeter is the bottom width, B, plus the two sides F in figure 
 2, of the stream. Since the length of the sloping sides is a number involving 
 the sums of squares of numbers and their square roots, the easiest way to 
 handle them is to use a tabulation showing the values as in table 2. 
 
 As a typical application, assume a bottom width of 3 feet, a side slope of 
 2 to 1, and a depth of 2 feet. From line 3 of table 2, the wetted perimeter for a 
 2-to-l side slope = B + 4.472 XD = 3 + (4.472 X 2) = 3 + 8.944 = 11.944 
 = P. Area for this assumed condition is A = (B + 2D)D = (3 + 2 X 2) X 2 
 
 A 14 
 
 = 7X2= 14 square feet. The hydraulic radius, R, = — = = 1.172 
 
 feet. P n -^ 
 
 One cubic foot per second is equivalent to the following volumes : 
 
 1 cubic foot per second = 1,728 cubic inches per second 
 = 60 cubic feet per minute 
 = 3,600 cubic feet per hour 
 
 = 448.83 gallons per minute (approximately 450 gallons) 
 = 50 southern California miners 1 inches 
 = 40 California statute miners' inches 
 = 0.992 acre-inch per hour (approximately 1 acre-inch) 
 = 1.984 acre-feet per day (approximately 2 acre-feet) 
 
 With these data available, a series of charts could be drawn to express each 
 equality. It is desirable for users of water to know the units. One of the com- 
 monest conversions, however, is from gallons per minute (g.p.m.) to cubic feet 
 per second (c.f.s.). If 1 c.f.s. = 448.83 g.p.m., then 2 c.f.s. = 897.66 g.p.m., 
 and so forth. Figure 3 portrays this relation. The dotted lines indicate the 
 procedure in using the chart. Starting at the scale for gallons per minute, on 
 the left, move horizontally to the sloped line and from that intersection, ver- 
 tically, down to the bottom of the chart, reading the cubic feet per second at 
 that intersection. The chart operates as well if the cubic feet per second are 
 known and the gallons per minute are to be determined ; the procedure in that 
 case is reversed. For general checking, the value 1 c.f.s. = 450 g.p.m. is close 
 enough. 
 
 Included, for convenience, are three graphs (figs. 4, 5, 6) from which one can 
 obtain the expected velocity of flow for any grade and for the three roughness 
 values, n = 0.015, 0.025, and 0.035. These graphs, drawn from the proved 
 data (13) of many field and laboratory experiments, are suitable for most 
 farm-ditch designs if a safe value of n is used. Values of n may range from 
 0.010 to 0.060. A safe value is larger than will probably be found; the three 
 values given in the chart are safe for use with the ditch type described for each. 
 
 A ditch must slope toward the point of delivery. This term, obvious in 
 meaning, is used synonymously with fall or grade to describe the change of 
 elevation of the ditch bottom or the stream surface for a given distance. For 
 example, the slope of a ditch 3,000 feet long may be 0.5 foot per 1,000 feet. 
 
Farm Irrigation Structures 
 
 4.0 
 
 Fig. 3. — Conversion chart — gallons per minute to cubic feet per second. 
 For explanation see text, page 6. 
 
 SLOPE - 
 
 IN FFET PER THOUSAND 
 
 30 40 60 80 100 
 SO 70 
 
 Fig. 4. — Value of n, 0.015, for ditches having very smooth beds — for example, wood-lined 
 channels using unplaned boards, not in perfect order because of inferior workmanship or 
 age ; slightly roughened concrete ; ordinary brickwork ; smooth stonework ; foul and slightly 
 corroded iron. These conditions might be called very good. (Diagram reproduced from 
 reading-reference 13, based on Kutter's formula.) 
 
California Experiment Station Circular 362 
 
 SLOPE — 
 
 IN FEET PER THOUSAND 
 
 Fig. 5. — Value of n, 0.025, for average ditches, reasonably clean but with fairly rough 
 beds — f or example, fairly new well-formed channels lined with natural sediment or 
 composed of loose coarse gravel. These conditions are about average. (Diagram reproduced 
 from reading-reference 13, based on Kutter's formula.) 
 
 SLOPE ► 0.1 0.2 0.3 0.4 0.6 0.8 1 2 3 4 6 8 10 20 30 40 60 80 100 
 
 IN FEET PER THOUSAND .12.15 25 0.5 0.7 1.2 1.5 2.5 5 7 9 12 15 25 50 70 
 
 0.15 
 0.12 
 •0 10 
 
 i Velocity of 
 Uniform Motion of Water in 
 Open Channels , from 
 
 v = c ^TsT 
 
 where R = Hydraulic Radius ; 
 S = tlie Slope; and 
 C = Kutter's Co-efficient. 
 
 Fig. e. — Value of n, 0.035, for weed-filled ditches with rough beds — for example, channels 
 in bad order, with stones and trash strewn on the bottom and sides, or about one-third full 
 of vegetation. These are very poor conditions. (Diagram reproduced from reading-reference 
 13, based on Kutter's formula.) 
 
Farm Irrigation Structures 9 
 
 In that case, the fall in the complete ditch is 1.5 feet; and the fall per 1,000 
 feet is 0.5 foot. The grade is 0.5 foot per 1,000 feet or 0.05 foot per 100 feet 
 or 0.0005 foot per 1 foot. The fall, grade, or slope of a ditch may also be ex- 
 pressed as so many feet per mile. For the ditch discussed above, the fall would 
 be 0.5 X 5.280 = 2.64 feet per mile. Sometimes when the unit distance is 
 understood— for example, per 1,000 feet or per mile — the slope, grade, or fall 
 is said to be simply 2.64 feet or 0.5 foot, which, for 2.64 feet per mile, would 
 be equivalent to 0.5 foot per 1,000 feet. The following discussion will identify 
 the unit of length in discussing these three synonymous terms. Another term, 
 friction loss, is used interchangeably with fall, drop, or grade. Friction loss 
 results largely from the drag or roughness characteristics of the wetted surface 
 of the conduit ; it equals the fall or drop in the water surface between any two 
 
 Fig. 7. — Ditch section with depth, bottom width, and side slope known. 
 See problem 1, page 9. 
 
 points in the stream. To illustrate the use of these ditch-velocity charts 
 (figs. 4, 5, 6), several typical problems are given below: 
 
 Problem 1. — A concrete-lined canal (fig. 7) in fair condition has a slope of 
 0.4 foot per 1,000 feet. What is its velocity and capacity? For this type of 
 canal, a safe value of n = 0.015. The cross-section area, 
 
 A water surface width -f- bottom width . „ .„ 
 A = X depth (from 
 
 2 table 2) 
 
 (2 4- 4) + 2 
 = ^ ^ ; -r X 2 = 8 square feet. 
 
 A A 
 
 The hydraulic radius = — = (from table 2) 
 
 P B + (2.828 XD) 
 
 A 8 =1.045 = *. 
 
 P P 2 + (1.414X2) 7.656 
 
 From the chart on figure 4, with a hydraulic radius of 1.045 and a slope of 0.4 
 foot per 1,000 feet, the mean or average velocity should be 2.0=1= feet per 
 second. 
 
10 
 
 California Experiment Station Circular 362 
 
 With a cross-section area of 8 square feet and a velocity of 2.0 feet per 
 second, the quantity of water in motion would be 8 X 2.0 = 16.0 cubic feet 
 per second. 
 
 Problem 2. — Reverse problem 1 : determine what depth of water will flow in 
 a 2-foot wide ditch bottom with 1 to 1 sides having concrete lining in fair 
 condition, and with a slope of 0.4 foot per 1,000 feet when 16.0 cubic feet is 
 being carried. Roughness, n, = 0.015. Assume that water 3 feet deep is re- 
 quired, and that problem 1 is not available. 
 
 A = (B + D)D = (2 + 3) X 3 = 15 sq. ft. (from table 2) 
 
 P = B + (2.828 XD)=2+ (2.828 X 3) = 10.484 sq. ft. (from table 2) 
 
 15.0 
 
 P 10.484 
 
 1.43 ft. 
 
 Fig. 8. — Ditch section with depth and bottom width unknown, but with 
 side slope known. See problem 3, page 10. 
 
 We must now check to see whether correct depth has been assumed. For 
 R = 1.43, slope S = 0.4, and the coefficient n = 0.015 for concrete lining of 
 this condition, we find from the chart in figure 4 that the average velocity 
 = 2.5 feet per second and the quantity handled = 2.5 X A (area) = 2.5 X15.0 
 = 37.5 cubic feet per second, which is too much. Evidently, too great a depth 
 has been assumed : both the area and the velocity are excessive, judging by the 
 solution to problem 1. Lacking problem 1 to go by, as would be the case in 
 normal design, another and smaller assumed depth would be taken, and 
 another computation would be made. A third or fourth computation might be 
 needed to determine the exact depth, which would prove to be 2 feet as in 
 problem 1. 
 
 Problem 3. — What size unlined clean earth-bank ditch would be necessary 
 to carry 2 c.f.s. with a slope of 0.2 foot per 1,000 feet? The value of n for 
 friction coefficient is 0.025 for a clean earth-bank ditch. The slopes of a dirt- 
 side ditch should be less steep than in a fined one because if the slope is too 
 great, the earth will slide in on becoming wet. A slope of 1.5 to 1 is satisfactory 
 and can be assumed (fig. 8). Assume a bottom width of 1 foot and a depth of 
 1.6 feet. 
 
Farm Irrigation Structures 11 
 
 A = (B + 1%D)D = (1 + 2.4) X 1.6 = 5.44 square feet (from table 2) 
 P = B + 3.6062) = 1 + (3.606 X 1.6) = 6.77 (from table 2) 
 
 P 6.77 
 
 For R = 0.804, slope S = 0.2 foot per 1,000 feet, and n = 0.025, velocity 
 = 0.65 (see chart, figure 5). The flow = A XV = 5.44 X 0.65 = 3.54 c.f.s., 
 which is too large. In design, one can stop here, because evidently a ditch with 
 the assumed dimensions will have over one and one-half times the desired 
 capacity — an adequate safety factor. To get closer to the true flow depth, we 
 might make another calculation. For a depth of 1.3 feet, 
 
 A = (B + 1}^D)D = (1 + 1.95) X 1.3 = 3.835, or 3.84 (from table 2) 
 
 P = B + 3.606Z) = 1 + (3.606 X 1}£) = 1 '+ 4.69 = 5.686 or 5.69 
 
 (from table 2) 
 
 P 5.69 
 
 Under these assumptions, then, velocity V = 0.58 (from chart, fig. 5). The 
 flow = 3.84 X 0.58 = 2.22, which is about 11 per cent over capacity and close 
 enough on the safe side. 
 
 Problem 4- — Suppose we desire to run 1,350 g.p.m. a distance of 800 feet 
 between two points having elevations of 452.64 and 452.40 feet, respectively. 
 Since the water supply is contaminated with weed and grass seed, the earth- 
 bank ditch will be quickly overgrown, and the flow impeded. How large a 
 ditch is necessary? We know that 1,350 g.p.m. = 3 c.f.s. (fig. 3). The value of 
 friction coefficient n for these weedy ditch conditions = 0.035. (Assume worst 
 conditions, which will develop soon in this ditch.) The slope or grade of ditch 
 = 452.64 feet - 452.40 feet = 0.24 foot per 800 feet or 0.3 foot per 1,000 
 feet. Assume a lj^-foot-wide bottom to ditch, 1 J^-to-1 side slopes, and a depth 
 of flow of 1.5 feet. (This guess is made from the solution to problem 3, where a 
 ditch with a 1 -foot-wide bottom, flowing 1.3 feet deep, carried 2.22 c.f.s.) 
 
 A = (B + \y 2 D)D = (1.5 + 2.25) X 1.5 = 5.62 sq.ft. (from table 2) 
 P = B + 3.606D = 1.5 + (3.606 X 1,5) = 1.5 + 5.41 = 6.91 
 
 P 6.91 
 
 For n = 0.035, R = 0.81, and slope 0.3 foot per 1,000 feet, the average ve- 
 locity should be 0.55 foot per second (from chart, fig. 6). 
 
 The quantity would be then 0.55 X 5.62 = 3.1 c.f.s.; and the depth, 1.5 
 feet, would appear to have been a very lucky guess, together with the assump- 
 
12 
 
 California Experiment Station Circular 362 
 
 tion that the bottom width is 1.5 feet. Obviously, no ditch should be made 
 just deep enough to handle the computed depth of water. Additional depth 
 not only gives a safety factor for settlement, but adds bank height above the 
 water level to prevent spilling over or erosion at the sides. Stock or other 
 trampling will then not be immediately destructive. 
 
 These examples and their solutions show the typical calculations necessary 
 in designing ditches. By using the three charts covering friction coefficients 
 and the simple formulas given above, one can solve many water-distribution 
 problems on the farm. For a large job involving many ditches and considerable 
 expense, one should obtain some hydraulics text (1, 4, 5) to supplement the 
 
 EU4SZ.6A 
 
 s El. 452.40 
 D B 
 
 Fi g> 9. — Typical field with ditch diagonally across it showing elevations in feet. 
 
 material available herewith or obtain the services of an engineer. Data avail- 
 able in such publications include a wide range of values for n, the friction 
 coefficient; a truly economical design can be completed by their use. This does 
 not mean that the three values for n (0.015, 0.025, and 0.035) are not subject 
 to economic application to field conditions. They provide a rough check of 
 ditch sizes required for the three general conditions — lined, unlined, but clean, 
 and unlined and weedy, or foul. Normally, such a rough check suffices for 
 farm use : the unlined ditches, which are most common, are built approximately 
 to dimension. For this reason the results of calculations must be on the safe 
 
 side. 
 
 DITCH CONSTRUCTION 
 
 One of the foregoing design problems might have referred to the field shown 
 in figure 9. Here water is to run from point A to point B, a distance (as in 
 problem 4) of 800 feet with the water level at A, 452.64 feet, and at B, 452.4 
 feet. A ditch has been designed to carry 1,350 g.p.m. or 3 c.f.s. from A to B, 
 utilizing the available slope and even allowing for the weedy conditions that 
 will develop soon after the ditch flow starts. Construction is begun. Midway 
 between A and B (fig. 10) is a 100-foot-long streak of clean gravel. Because, 
 
Farm Irrigation Structures 1 3 
 
 obviously, such a streak will cause excessive seepage losses, the ditch must be 
 lined in that zone. A new design problem for this 100-foot length of ditch 
 arises and is solved as in the examples. It might be possible to go around the 
 gravel strip, passing through C or D en route to B. Since, however, much addi- 
 tional distance would result and the slope would be reduced proportionally, 
 requiring a larger ditch than in the original design, lining the 100-foot strip 
 of the gravel zone might be the cheapest. Without checking the economics of 
 the various routings, we may conclude that success depends upon careful 
 preliminary studies, both by using surveying instruments and by inspecting 
 the physical aspects of the proposed route. The ditch must be located on a 
 path suitable from several aspects. First, it must be convenient for irrigation : 
 that is, its water surface must be above the land. In general, therefore, it 
 should follow a ridge or start from a high point. Second, it must have fall or 
 drop in the direction of flow. If a uniform grade cannot be kept, we must 
 
 I surface' J " ' J ~| ' 
 
 '"Swale" 1 ' Gravel 5 one "' ^ Q \\om of dihch^ 
 
 Fig. 10. — Profile of the ditch line across a field, shown in figure 9. 
 
 change the ditch dimensions by solving satisfactorily a problem for each grade 
 used and for the flow desired. Third, the ditch should not interfere with culti- 
 vation operations any more than is necessary. This requirement may add 
 additional length to a ditch. Figure 9 shows a possible bad location. If the 
 ditch runs from A to B, the field is cut in two, forcing cultivation equipment 
 to go a long way around unless bridges are supplied. Since the example, A-B, 
 is a comparatively short ditch, 800 feet, the distances might not be too great 
 if the equipment had to go around; but if A and B were much farther apart, 
 bridges might be necessary. Each operator must decide how far convenience 
 in operation should control the final layout. 
 
 Additional considerations can be drawn from ditch A-B, which will have 
 both lined and unlined sections. An assumed profile along the path of A-B 
 will illustrate many matters that must receive attention. In reality not all the 
 procedures suggested would be warranted on such a short ditch; if all were 
 used, they would complicate the construction unnecessarily. This statement 
 will be clearer later on. The reader can then decide which method suits his 
 conditions and can be followed out with available equipment. As a profile 
 shows (fig. 10), the land surface is irregular. The ditch line is straight on the 
 bottom and at the crest of the banks. The relative position of ditch section 
 and present earth levels are shown as sections taken at points A, E, F, and G 
 in figure 11. This figure will be used to illustrate the various construction 
 procedures. Starting with section A at the left-hand end of the ditch, the sketch 
 
14 
 
 California Experiment Station Circular 362 
 
 indicates that more dirt will be cut to make the ditch than is necessary in fill to 
 make the bank at this point. This condition is fortunate : any excess dirt can 
 be used at section E, where a fill must be made to bring the grade up to the 
 ditch line. If the ground surface were regular all the way from A to B and the 
 
 ^j}£^ 
 
 _FIH_ 
 Surface of SoiM 
 
 DiFch SecHon at A 
 
 I surface ' 
 
 Dihch SecHon ah F (no scale) 
 
 Dihch SecHon a I" G 
 
 Fig. 11. — Details of cut, fill, and lining in the ditch across a field, showing 
 construction details of locations designated in figure 10. 
 
 ditch were cut into the soil as at section A, the excess cut would become a 
 problem. A solution might be to raise the bottom and the bank tops so the cut 
 equalled the fill. Thus the excavated material placed on the adjoining uncut 
 soil banks would make the fill necessary to form the ditch to the dimensions 
 required. Cuts and fills balance up the cross-section area if the actual cut 
 equals the cross-section area of settled fill or is slightly greater (not over 10 
 per cent). The material removed is loose and will settle; it should therefore be 
 
Farm Irrigation Structures 15 
 
 piled higher on the banks than the final design requires. Unless such allowance 
 is made for settlement, the banks may not be high enough to provide a safe 
 freeboard. At the top they have a flat surface, whose width should not be less 
 than 6 inches. This width gives protection against weathering and wave action. 
 Tamping any fill will help consolidate it and reduce the hazard of leakage till 
 settlement further consolidates it. One should not attempt to pack wet soil: 
 the water has caused it to swell, and mechanical packing has little effect on 
 this condition. Such soil shrinks as it dries, and the effort of packing is wasted 
 because cracks appear and leakage is almost certain. Completely dry soil is 
 equally poor packing material; a moist soil not wet enough to be plastic is 
 
 Fig. 12. — Fresno scraper in use making ditch banks. 
 
 best. Since the textures vary, no general rule is safe; one should experiment 
 with each soil to determine the best condition for working. In general, a good 
 condition prevails when plowing does not glaze the soil at the surfaces of 
 contact with the plow. In the detail of section E, figure 11, the land surface is 
 below the bottom of the completed ditch. Under these conditions the ditch 
 must be built on a fill, which should be compacted as much as possible during 
 construction operations. This section of the ditch at E resembles that for a 
 ditch built by raising two parallel banks on the surface of the ground. The 
 necessary dirt must be brought to the site from the adjoining surface — perhaps 
 by a fresno scraper. Soil can be skimmed from the surface or off high points 
 adjoining the ditch and dumped to form the two banks. In this procedure the 
 teams or tractor, as well as the fresno scraper (fig. 12), cross and recross the ac- 
 cumulating fill, packing it down as work progresses. Final shaping is sometimes 
 given the ditch by hand shoveling. In E (fig. 11), a rather large volume of ma- 
 terial will be needed for fill. Some of this might come from site F, where a ridge 
 rises above the water surface in the proposed ditch. Taking off this surface 
 
16 
 
 California Experiment Station CmcuiiAR 362 
 
 soil would permit irrigation from the ditch at that point and would supply the 
 soil deficiencies at E as well. If the material is used from F, the gravel zone, the 
 gravel must be mixed and buried through the bottom of the fill, and watertight 
 soil must make up the foot or so of thickness adjoining the ditch sides and 
 
 Fig. 13. — Upper view, disk ridger suitable for making a ditch bank. 
 Lower view, four-blade furrowing implement for making several small 
 ditches at once. 
 
 bottom. Such a fill must be packed as well as is feasible during its building. It 
 should be settled by allowing water to stand in it as soon as convenient and 
 before the water is permitted to flow through. After settlement, the filled area 
 may need reshaping. A fill is a precarious location for a ditch and should be 
 avoided if possible. Whenever the nearby soil is not suitable for holding water, 
 clay should be brought in if possible, and a 6- or 8-inch layer of it should be 
 
Farm Irrigation Structures 17 
 
 tamped in place in a filled section. Since the best soil on any farm lies in the 
 first foot depth, avoid wholesale stripping of this layer to make fills. 
 
 Detail F, figure 11, shows the lined section in this ditch. The occasion for the 
 lining is the gravel zone in figure 10 and in problem 4. A use for material 
 removed at F has been suggested in the foregoing as part of the fill at E. If no 
 large volume of fill is needed, the excavated material can be spread over a 
 considerable area. In constructing the ditch at F, the normal procedure would 
 be rough-shaping with disks or plowing and rough-shaping with the V-board 
 shaper (figs. 13 and 14). Finish-trimming would be done carefully with shovels, 
 to furnish a smooth surface on which the lining will be spread. The finished 
 earth is cut 2 to 3 inches below the finished face of the lining and acts as the 
 
 Fig. 14. — Metallic V-board ridger useful in forming diteh banks. 
 
 outside form for it (fig. 15). The lining, whether asphaltic or plain concrete, 
 is troweled or bladed in place and smoothed as well as possible. Reinforcing 
 may or may not be used. Smaller ditch linings are generally made without 
 reinforcing. Expansion joints (fig. 16) are sometimes provided at 10- or 15-foot 
 intervals on concrete linings. Three or 4 strips of tar paper in the joint between 
 sections of concrete will act as expansion points. Concrete can be put on by 
 means of a jet or "gunited" to the V-shaped ditch banks (fig. 17), forming a 
 dense, satisfactory lining. This procedure is followed for large ditches, and 
 steel reinforcing is generally used with it to bind the concrete. Troweling 
 improves "Gunited" cement linings. Ditches are sometimes lined with con- 
 crete poured in place against forms. 
 
 Detail G, figure 11, discloses a situation somewhat like that in section A. 
 There is, however, less cut than fill ; some soil must be carried in to make the 
 banks of the fill high enough and wide enough. The disks on the plow, to- 
 gether with the V-board shaper, can be used to cut the rough ditch. Then the 
 necessary additional soil must be scraped in by a fresno. When that soil is in 
 place and compacted, the final hand-trimming can be done. 
 
 All hand-trimming follows grades set by reference to surveyed stakes so the 
 final grade is true. Frames of light wood, having the correct dimensions of the 
 ditch, can be moved along during trimming to insure accurate slopes and level 
 
18 
 
 California Experiment Station Circular 362 
 
 Fig. 15. — Hand-trimming a ditch to the dimensions of a template before 
 pouring the concrete lining. 
 
 b!i i 
 
 Fig. 16.- 
 
 -Concrete-lined ditch with expansion joints (showing as 
 black joint lines). 
 
 bottoms. Figure 15 details such a frame, suitable for use in shaping the un- 
 lined sections of the ditch designed in problem 4. Many earthen- wall ditches 
 are finished by the rough shaping processes alone. Such ditches cannot be as 
 clean and true as hand-trimmed ones. 
 
 If the ditch A -B (fig. 9) must be bridged for a road crossing, a logical spot 
 
Farm Irrigation Structures 
 
 19 
 
 would be at F (fig. 10), where the ditch is lined and lies entirely below the 
 normal soil level. A typical bridge is detailed on page 36. 
 
 The filled section over which the ditch is carried at E (fig. 10) might be 
 
 Fig. 17. — Placing gunite cement on stucco ware reinforcing, wdth the trimmed 
 earth serving as the outside form. 
 
 Fig. 18. — Typical small wooden flume across a depression. 
 
 replaced with a flume (fig. 18). This resembles a lined ditch of half-round, 
 half-rectangular, or possibly triangular section made of wood or metal and 
 resting on framing that maintains the grade line of the ditch. Computations 
 for a short flume at this point could be safely based on a value of n = 0.015, 
 as for lined concrete ditches. A long flume should be designed more accurately, 
 
20 California Experiment Station Circular 362 
 
 for economy; a textbook on hydraulics (1, 5, 7) will provide a better n value. 
 Normally there would be a smaller value of n than 0.015 for a smooth metal or 
 wood flume. If so, a smaller flume could do the work than would be indicated 
 using 0.015. For a short flume the economy might not be worth the trouble 
 of looking up the text. 
 
 The preceding discussion of design and construction of ditches has purposely 
 omitted the structures generally present. These devices cannot be neglected 
 in design; they sometimes act as partial dams, backing up the water and 
 changing the slope of its surface in the ditch a considerable distance upstream. 
 This change gives rise to a new set of conditions to control the flow. The 
 gradient is lowered, or the slope is less; and the depth of the flowing stream is 
 increased. The wetted perimeter and the cross-section area have changed; and 
 the hydraulic radius has also shifted in value. If such structures have a moder- 
 ate effect on water levels in the ditch, no trouble may be experienced in a ditch 
 designed to operate without them. The freeboard from the water surface to 
 the top of the banks must be adequate — a strong argument for providing such 
 freeboard in constructing a ditch. 
 
 As was mentioned earlier, the control devices in ditches are not new, but the 
 name applied to the same structure varies in different localities. The general 
 design details of these various controls will be described hereafter on the basis 
 of what they are supposed to accomplish, and some of the names will be given. 
 
 The ditch needs protection from one thing — namely, the hazard of washing 
 out. This problem arises during normal operations and also from abuse or 
 carelessness; and it is present with either lined or unlined channels. The lined 
 ones can withstand higher rates of flow than the unlined, but neither can with- 
 stand overflow of the banks very long without experiencing a serious washout. 
 Both types must be protected by taking special precautions at points where a 
 side delivery is made to a field or into another ditch. Often such side outlets 
 involve a sudden change in grade between channel water surface and that of 
 the outlet. This available fall for the water sharply increases the velocity and 
 the swirling erosive action of the stream. To avoid damaging the banks and 
 to protect critical outlets, one must build permanent structures along the 
 sides and across ditches. 
 
 SUPPLEMENTAL STRUCTURES FOR DITCHES AND CANALS 
 
 Permanence in a structure on a ditch depends upon the uses and abuses to 
 which it is subjected as well as upon the strength and design details of the 
 unit. Design assumes that some abuse must be met and absorbed and that the 
 construction is heavy enough for safety. The trampling of animals can destroy 
 the ditch bank, permitting a washout that will wreck the finest structure. To 
 pry stuck gate slides will rack and loosen the frames of wooden structures, 
 eventually causing leaks and washouts. High-velocity flows are ruinous. Fail- 
 ure to maintain the structures will result in serious deterioration. All these are 
 abuses. The amount of protection to be specified in design, both for mechanical 
 strength and for prevention of washouts, depends upon the soil characteristics 
 
Farm Irrigation Structures 
 
 21 
 
 near the structure. If the soil is loose or sandy, washing will be more likely; the 
 structure must have more strength and greater washout protection than in 
 soils having greater natural stability. Gate openings smaller than are required 
 may fail to pass all the water. When this happens, the water backs up in the 
 ditch, causing overflows and washouts upstream, or at the structure, or in both 
 
 Top.opening gofe wi. h h removable section cover 
 
 1 N. 
 
 "** 
 
 
 
 
 i \ 
 i \ 
 
 Opening 
 
 
 
 
 y i 
 
 i w 
 
 
 
 Center- opening gale wil"n unit" slide cover 
 
 Bottom-opening gafe with swinging cover 
 Fig. 19. — Three simple ditch structures, showing different types of opening. 
 
 places. Structures made too lightly will warp and sag, permitting leaks and 
 eventual washouts. Such problems must be solved. 
 
 Normally a ditch structure is a bulkhead containing an aperture that can 
 be opened and shut at will or through which the flow can be somehow con- 
 trolled. This aperture may be located at the top, the center, or the bottom of 
 the face of the structure. The movable cover or gate may slide as a unit, or be 
 removable in sections, or swing in an arc across the opening (fig. 19). The 
 simple bulkhead becomes more useful if supplementary framing is added to 
 protect the ditch further, or to make the structure more flexible in its use, or for 
 
22 
 
 California Experiment Station Circular 362 
 
 both reasons. The name of the structure will change as different framing is 
 added: a new name will identify some special capacity of the unit as con- 
 structed. The simple bulkhead could be used unsupplemented for every control 
 operation if the natural weakness of the ditch, the washout hazard, did not 
 necessitate more protection. For example, a simple bulkhead with an aperture 
 at the top could successfully divert water into a flume or pipe while holding the 
 
 Ga\e boards removed 
 
 Fig. 20. — Typical installation of two single-wall wood gates in a ditch. 
 
 level in the ditch fairly high. If this same structure had to keep the water levels 
 up in the first ditch and if at the same time water poured over the crest of the 
 opening and down to an adjoining field or into a ditch,* trouble would result. 
 The overpouring stream would increase in velocity in falling, and the energy 
 thus developed would be dissipated where the stream landed. With the simple 
 bulkhead the stream would fall on earth; and the earth would be washed 
 away, forming a hole into which the structure itself would eventually fall. To 
 prevent this potential failure, an apron of impervious material (wood, con- 
 crete, metal) could be built below the structure at, or a few inches below, the 
 normal ditch bottom. The simple structure holding the water levels in the 
 ditch might be called a "check gate" : it halts the flow in the ditch till the water 
 
Farm Irrigation Structures 
 
 23 
 
 levels are raised a predetermined amount. It might be called a "turnout gate" 
 or a "field gate," while acting at times as a "check." Given the apron, with the 
 ability to stand up in use while the water dropped down on the apron, it could 
 be called a "drop gate" or a "check drop"; the latter name shows its dual 
 capacity. Where such a gate is the start of a lateral ditch, the term "head gate" 
 might also be applied because the structure is, in fact, the head gate of that 
 lateral. If the gate cover in any structure is raised partially, permitting the 
 water to flow under the cover, the flow is "undershot," and the gate might be 
 described as an undershot gate. As the foregoing illustrates, the name indicates 
 
 Fig. 21. — Prefabricated single-wall concrete gates ready for installati 
 
 the function; but there is really only one basic structure to which supple- 
 mental framing is added to give the required protection under special con- 
 ditions. 
 
 As these illustrations show, one may discuss the ditch structure by starting 
 with the simple bulkhead type, adding to that type till the most complicated 
 structure is evolved. On this basis one can explain the uses and limitations of 
 each stage in the evolution of the structure. 
 
 Single-Wall Structure. — Figure 20 illustrates two typical single-wall struc- 
 tures located near each other on a ditch. (Figure 21 shows a concrete single- 
 wall gate.) The one labelled a (fig. 20) spans the ditch, whereas b lies across 
 an opening in the bank; a has a full opening, whereas b opens near the bottom. 
 Each could replace the other. The gate a has the following uses: (1) A stop, 
 to prevent flow in the ditch beyond a. (2) A check to back the water up a little 
 so the flow through b will be satisfactory. (3) A check drop, to back water up 
 in the ditch so some flow will pass through b while the excess passes over the 
 removable gate boards to the downstream side of the ditch. Unless the drop 
 
24 
 
 California Experiment Station Circular 362 
 
 in this case is very limited (an inch or two), damaging erosion will result. 
 (4) An undershot check, with the same purpose as in (3), but with excess water 
 flowing under the gate boards on a instead of over. Possible erosion might 
 result if water rises very high on the upstream side of a. (5) Unobstructed flow 
 of the water when the gate is wide open. 3 
 
 Sh-uchure 
 
 Dihcn bonk 
 
 Sheef" of convos, burlap ormehal 
 fastened on down-slream edge. 
 
 Fig. 22. — Three typical apron installations below single-wall gates. 
 
 The gate shown at b has the following uses: (1) A stop, to prevent flow if 
 none is desired. (2) A turnout to release water to an area of use. (3) An under- 
 shot check permitting undershot release of water along the bottom of the gate 
 boards while a head of water stands above the opening of release. (This use, 
 however, is limited.) 
 
 As may b e noted in figure 20, both a and b extend into the bank a consider- 
 
 3 The opening a or 6 must be large enough to permit the required flow to pass 
 through quietly; otherwise erosion will result downstream. The tabulation for permissible 
 velocities in ditches for various types of soil (table 1) gives the limiting velocities for 
 the several soil types. Thus, if the soil type will accommodate 1 foot per second velocity 
 and if we desire to pass 900 g.p.m. or 2 c.f.s. through b, the size of opening b should be not 
 less than 2 square feet — or, for safety, 2}4 to 3 square feet. 
 
Farm Irrigation Structures 
 
 25 
 
 able distance along both sides and at the bottom of the structure. This arrange- 
 ment is necessary both for support and as a seal to permit the earth to close off 
 flow from one side of the wall to the other. Because of the simple, light wall 
 
 ShuHer 
 
 * 
 
 1 
 
 
 IT 
 
 
 i 
 
 < C-2 5- 
 
 H 
 
 FRONT VIEW 
 
 REAR VIEW 
 
 RECOMMENDED SIZES 
 
 C.F.S. 
 
 A 
 
 B 
 
 C 
 
 D 
 
 E 
 
 F 
 
 3 
 
 7'0" 
 
 2' 6" 
 
 3'cr 
 
 r6" 
 
 2'0" 
 
 3'0 M 
 
 6 
 
 9'0" 
 
 3'0" 
 
 3'0" 
 
 2'0" 
 
 3'0" 
 
 3*6 M 
 
 9 
 
 10' 0" 
 
 3*0" 
 
 4'0" 
 
 20" 
 
 3'0" 
 
 3' 6" 
 
 12 
 
 i rcr 
 
 3'6- 
 
 S'O" 
 
 2'0" 
 
 3'0" 
 
 36" 
 
 16 
 
 iro" 
 
 4'0" 
 
 5'0- 
 
 2*6" 
 
 3'0 H 
 
 4*0" 
 
 Fig. 23. — Single-wall gate with side walls only for protection of banks. 
 The table shows dimensions required for various flow capacities. 
 
 construction, these gates are limited to rather narrow ditches (4 to 6 feet of 
 unsupported space is maximum) . They are not strong enough for long spans, 
 but would bow under load, jamming the gate boards fast, and would probably 
 be ruptured by the resulting pressure. More complex designs gain strength 
 that permits their use in large ditches. 
 
 There are many shortcomings of a and b: (1) as already noted, they are 
 
26 
 
 California Experiment Station Circular 362 
 
 structurally weak; (2) the wing walls and bottom of the structure, though 
 imbedded in the ditch body at least V/2 times the height of the outlet hole, 
 are subject to seepage or leakage, which brings disaster; (3) spilling over of 
 water from one side of either structure to the other side will cause serious 
 erosion; and (4) downstream surfaces are subject to erosion. 
 
 Single-Wall Structure Plus Apron on Downstream Side. — The addition of an 
 
 Apron depressed 
 4" to 6" below 
 soil surface 
 
 Fig. 24. — Two simple gates with apron : upper has apron flush with ground ; 
 lower has apron depressed below the soil surface. 
 
 apron is the simplest improvement that can be made to the single-wall struc- 
 ture. The apron (fig. 22) increases the safety of the unit. The installation can 
 be safely used as a drop for short periods or for small flow for indefinite periods. 
 The overpour, as it lands on the apron, rushes both downstream and at right 
 angles toward the ditch banks. As a result, the banks close to the apron 
 are subject to the swirling action of those side-thrusting parts of the stream. 
 This is a weakness and renders the apron by itself only partially effective. 
 Aprons can be made of almost any convenient material, ranging from burlap 
 sacks to concrete. The addition of side walls to the apron is the obvious next 
 step. (Figure 23 shows details of structure with side wall only added.) 
 
Farm Irrigation Structures 
 
 27 
 
 Single-Wall Structure Plus Apron with Side Walls on Downstream Side. — 
 Figure 24 illustrates variations in the use of the apron under the nomenclature 
 c and d. Although they are actually the same apron, one (d) is depressed 4 to 
 6 inches below the soil level at the outlet to make a pond just below the gate 
 opening that can help to absorb the energy if water is dropped over the top of 
 the gate boards, thus helping to prevent erosion. The side walls on either cord 
 assist in protecting the bank of the ditch from the splash and eddy of the 
 
 jff&m&num: 
 
 LONGITUDINAL SECTION 
 
 Fig. 25. — Gopher-type gate. Single wall plus a boxed-in apron. 
 
 water. Of the two aprons, d is the more effective. As with the simple wall, the 
 limitation of these structures is the frailty of the wall itself and the likelihood 
 of leakage. The apron framing does add some strength; but if leakage starts, 
 this added strength is of little value. The side walls on the apron make that 
 addition a short-length flume. 
 
 Gopher-Type Turnout or Field Gate. — This special adaptation of the single- 
 wall structure is used where delivery must be made in the field at a short dis- 
 tance from the ditch and with the least possible turbulence — a feature desirable 
 where sandy soil prevails and where washing may be severe. Details are given 
 in figure 25. The gate is a single-wall structure with a completely boxed apron 
 extending through the ditch bank discharging at, or slightly below, the surface 
 
28 California Experiment Station Circular 362 
 
 of the field being served. Its chief uses are as follows: (1) a stop when no de- 
 livery to the field is desired; (2) an undershot check if a controlled part of 
 the ditch flow is to be taken from the ditch; (3) a device that gives a quiet 
 welling flow upward onto the field with little agitation; (4) a structure that 
 does not cut the main ditch bank by wing walls ; and (5) a protected inlet due 
 to extension of apron a short way into the ditch. 
 
 The gopher-type gate has been utilized only where special precautions are 
 desirable to prevent washing. Considerable savings in soil otherwise lost by 
 erosion would result, however, if it had more frequent use. One objection to it 
 is the maintenance, in the field, of an open hole into which stock or farm 
 
 Fig. 26. — Simple form of gopher-type gate having no wing wall. 
 
 equipment may fall. In addition, the shallow soil cover (if any) makes the 
 apron subject to damage if crossed by heavy farm implements or large animals. 
 Figure 26 shows a simpler form, extending through the ditch bank only. 
 Adaptations to concrete (fig. 27) permit stronger construction, with the outlet 
 in the field closed by a disk-type valve. 
 
 Double-Wall Structure with Connecting Flume. — To remedy the shortcomings 
 of the simple wall-type gate, the double- wall gate is used with a flume joining 
 the two walls. This construction adds much mechanical strength and almost 
 eliminates leakage. Figure 28 shows a typical gate. In it the two simple walls 
 tied together by the flume and braced by the earthen fill between their wings 
 can be made firm and watertight even in very loose soil. Where there is still 
 hazard of leakage, the length of the flume is increased, widening the distance 
 between the walls. This structure may serve as (1) a stop or a check; (2) a 
 drop, if spill is into flume; (3) an undershot or overshot delivery gate, deliver- 
 ing preferably into a flume; (4) a turnout or field gate replacing b in figure 19 
 in general, but especially in loose soil; (5) a head gate; (6) a gate across large 
 
Farm Irrigation Structures 
 
 29 
 
 L ONGI TUOINAL SEC TION 
 
 TABLCi OF DIMENSIONS 
 
 d 
 
 D 
 
 d 1 D 
 
 0" 
 
 24 
 
 16 | 24' 
 
 IO 
 
 24' 
 
 IS' 
 
 ■30' 
 
 lZ" 
 
 2*' 
 
 20' 
 
 JO' 
 
 1-4-' 
 
 24' 
 
 24' 36' 
 
 MOTES OA/ DIMENSIONS 
 H - Stondard /teionrs, 4"0~,*3 0" ond 6 0' 
 h - From 6" to 12' 
 L> - Usuol/y S'O", rim, mum 26 " 
 Z* - As required Usually ?'6", /O'O" or /?'€>" 
 E'ltot -tube rneasurt men-fa ore mode /r> 
 pipe upstream from s fond pipe . 
 
 Fig. 27. — Application of concrete pipe to gopher-type gate. 
 
 Fig. 28. — Double-wall structure with gate in connecting flume. The 
 dimensions show typical construction proportions. 
 
30 
 
 California Experiment Station Circular 362 
 
 spans where the simple gate would fail; and (7) a device into which double 
 gates can be put, one behind the other, with earth between them to assure 
 tightness. 
 
 Evidently, therefore, this type has almost universal application. It is espe- 
 cially useful for unlined ditches. Variations have the wing walls set at an angle 
 other than 90 degrees to the flume line (fig. 29) or wing walls of differing length. 
 The angle-set wing walls create a throat at the start and end of the flume, 
 reducing turbulence and erosion. The varying length in these walls adapts the 
 structure to each situation without waste. 
 
 \-' r -"^SiU.-l- , ^»*t ' * *. 
 
 Fig. 29. — Double- Avail gate. The throat in this gate is shown reduced 
 to the minimum. 
 
 All gates, regardless of their use, are variations of those already described. 
 Not all the modifications can be mentioned ; but with the description and illus- 
 trations given, anyone can devise a gate to satisfy special requirements. Since, 
 however, some of those detailed will meet almost any condition, individual 
 planning will usually be needless. One phase in the placing of any structure 
 has been referred to without details — namely, the tamping of earth about the 
 walls to form a water seal and, at the same time, to support them. Where a 
 flume or an apron accompanies a single-wall or multi-wall structure, the bot- 
 tom cutoff wall must be placed first, and the excavated earth tamped about it. 
 This tightly packed earth must be carefully graded and trimmed to receive 
 the remaining structure. Tamped earth must have the damp-soil character- 
 istics already described as necessary for ditch banks. Because small leaks 
 through soil become large ones in time, careless refilling of excavations may 
 result in destruction by washout when the structure is in use. Obviously, 
 careless procedure here can ruin a good design. Special attention must be 
 given newly installed structures to make sure they are securely sealed in the 
 earth banks. 
 
 Besides the true structure, there are several devices used as gates as well as 
 
Farm Irrigation Structures 
 
 31 
 
 special adaptations of gates and other contrivances for controlling water in 
 and delivery from ditches. Most of these units have a limited application. 
 A serviceable gate can be made from a section of large-diameter sewer tile 
 
 Fig. 30. — Sewer tile used as a gate on the side of a ditch 
 
 Fig. 31. — Concrete pipe with gate cast at one end. 
 
 or from a large concrete pipe, a removable wood disk being used as a seal when 
 no flow is desired. Such a gate (fig. 30) may be installed where ditch banks are 
 relatively impervious, so that seepage is not a problem along the outside sur- 
 face of the pipe. Big pipes (see also fig. 31) pass large flows of water for flooding 
 operations; but for furrows several small streams are better. A series of small 
 galvanized-iron pipes or rectangular or other milled-bore wooden tubes with 
 
32 
 
 California Experiment Station Circular 362 
 
 TABLE 3 
 
 Flow through Redwood Furrow Tubes of Different 
 
 Dimensions as Shown in Figure 32* 
 
 Shape of tube 
 
 Length 
 
 Diameter 
 
 of 
 
 cross 
 
 section 
 
 Depth 
 of water 
 over inlet 
 
 of tube 
 
 Flow 
 
 Circular 
 
 Circular 
 
 Circular 
 
 Circular 
 
 Circular 
 
 Circular 
 
 Circular 
 
 Circular 
 
 Circular 
 
 Circular 
 
 Circular 
 
 Circular 
 
 Flattened ellipse. 
 Flattened ellipse. 
 Flattened ellipse. 
 Flattened ellipse. 
 
 feet 
 
 2 
 
 2 
 
 inches 
 
 iVs 
 
 iVs 
 
 v/% 
 
 1X2 
 1X2 
 1X2 
 1X2 
 
 feet 
 
 0.64 
 
 .206 
 .617 
 .227 
 598 
 .184 
 .590 
 .194 
 .578 
 .156 
 .573 
 .175 
 .517 
 .135 
 .526 
 0.146 
 
 gallons per 
 minute 
 
 5.22 
 
 3.50 
 
 4.22 
 
 2.75 
 
 9.17 
 
 5.62 
 
 7.78 
 
 5.22 
 12.30 
 
 7.62 
 10.21 
 
 6.24 
 19.69 
 10.90 
 18.08 
 10.07 
 
 Discharge of tube passes freely into air (not submerged), and the tube is horizontal. 
 
 Fig. 32, — Wooden furrow tubes in use. The openings are usually 
 made elliptical or round, 
 
Farm Irrigation Structures 
 
 33 
 
 proper movable gates can serve a group of furrows all at one time. Such small 
 furrow feeders or furrow tubes are shown in figure 32. As used, these tubes 
 pass through the ditch bank — a weakness, since each separate tube may permit 
 leakage and washout. Table 3 gives the capacities of several shapes and sizes 
 of furrow tubes. One can substitute for the tubes a series of small light-weight 
 siphons, 1 to 2 inches in diameter. These are filled by submerging in the 
 ditch. Then one end only is withdrawn, sealed by the palm of the hand; and, 
 when this outside end is lower than the water in the ditch, the hand is removed. 
 Flow commences at once ; and a man can control 20 or 30 siphons easily, either 
 as furrow feeders or for flooding operations. He picks them up when the irri- 
 gation is complete and moves them to the next location, where the cycle is 
 
 Fig. 33. — Six-inch galvanized iron siphon in operation, showing flow from 
 ditch (right) to field (left). 
 
 repeated. Occasionally, larger 6- to 8-inch siphons (fig. 33) are used; but they 
 are extremely cumbersome. The advantage of the siphon is that ditch banks 
 are unbroken, yet the water is available at any time at any area of use. In 
 addition, one set of siphons will serve for all the irrigation. A labor saving may 
 by made possible by the use of siphons in some locations. 
 
 Often it is desirable to use a ditch in sections, filling new parts lower 
 down as the irrigation progresses. A series of permanent structures for this 
 purpose would be costly ; but a portable dam or stop in the form of a canvas on 
 a frame or metal panel, called a tappoon, that can be driven into the earth and 
 across the ditch, can be used repeatedly to control the ditch flow. These devices 
 (fig. 34) are applicable to earth ditches. To permit the same control on lined 
 systems, slots for such gates could be cast into the faces of the channels at 
 any desired intervals. One set of gate boards could then be used at any point 
 where slots are provided. Under some conditions, permanent structures serving 
 as tappoons are placed in earthen ditches where the grade conditions either in 
 the ditch or on the land necessitate their use. Two such applications are de- 
 tailed in figure 35. 
 
 Since an open ditch is an obstruction, a passageway for farm machinery 
 
34 
 
 California Experiment Station Circular 362 
 
 must be provided. Two procedures are possible. The first is to build a bridge 
 over the ditch; the second to run the ditch under the road. This latter method 
 is common where ditches cross highways and the road cannot be graded up 
 over the ditch banks to permit the use of a bridge. Figure 36 illustrates a 
 typical bridge. Most ditches are not too wide to permit use of a simple flat 
 span, as illustrated. If a center pier, or piers, is added, two or more flat spans 
 
 Canvas 
 
 Metal plate with 
 sUffeners attached 
 for driving info dihch 
 
 F\g. 34. — Two types of tappoon shown in use. 
 
 of this type may be used. Larger, more complex permanent bridges can be 
 built; their construction will probably be more economical if a competent 
 designer is given the assignment. The bridge shown in figure 36 can be 
 adapted to most farm requirements. In many locations the road grade must 
 be raised from some distance away toward the bridge to keep the bottom of 
 the span above the high- water level in the ditch. The ends of the bridge should 
 rest on a solid footing or foundation to protect the banks, which when wet 
 might not support the ends of the bridge. 
 
 The operation of running the ditch under a road is that of transferring the 
 flow into a depressed pipe. Such a pipe is called an inverted siphon (see fig. 37 
 
Farm Irrigation Structures 
 
 35 
 
 for details). It need not necessarily be round, but may be rectangular or have 
 any other easily constructed cross section. The size necessary may be calcu- 
 lated as in the pipe-line problems given later in this paper. 
 
 One should design the inlet and outlet of a siphon so that the water will 
 enter and leave with a minimum of turbulence. A tapered inlet throat and a 
 similar outlet with the large end of the taper toward the ditch will provide 
 
 Ground surface 
 
 Spill line or ponding wafer 
 surface _ 
 
 jC Ditch bank 
 
 r — -* 
 
 6 ale w'ifh apron 
 
 Qoitom oi ditch 
 
 Ground surface 
 
 Pitch bank 
 
 5pill line and wafer 
 surface grade line- 
 
 Gate mth apron -7 g 
 
 Fig. 35. — The upper drawing shows use of a simple-wall structure as check or stop gate 
 to pond water behind it in order to raise the water level in the ditch above adjoining land 
 to be watered. Its use is similar to that of a tappoon with a drop overflow; the lower draw- 
 ing shows the simple structure used as a drop gate along a ditch whose slope is the maximum 
 permissible for the soil type. The ground surface slopes more sharply than the ditch j so the 
 drop must be used to lower the ditch line, keeping it closer to the ground-surface elevations. 
 
 desirable conditions. Figure 37 shows these throats. Their design is not critical: 
 any shape that produces a gradual change in dimensions from the ditch section 
 to the pipe size will work. Economy in construction limits the length of the 
 throats. As shown in the figure, the pipe itself should be completely sub- 
 merged at the small end of the throat ; otherwise the flow of water to the siphon 
 will be limited to the amount that can enter the partly filled pipe inlet, and 
 
36 
 
 California Experiment Station Circular 362 
 
 most of the value of the throat will be lost. A covering of at least 18 inches of 
 earth is desirable to protect the top of the siphon from damage by heavy 
 traffic. Such structures are sometimes called culverts. 
 
 A 
 
 PLAN 
 V/idf-Ji as requird 
 
 r - - * i 
 
 A 
 
 M 
 
 N 
 
 « » H 
 
 M 
 
 H 
 
 « 
 
 Ms H 
 
 N 
 
 N 
 
 H 
 
 M 
 
 ^ 
 
 I"xl2" 
 1 
 
 Boltom 
 
 Plate -* 
 
 of ditch 
 
 112") 
 
 1 12" 
 
 siTf 
 
 
 - 4'x 
 
 6" Sfrir 
 
 gers 
 
 1 
 
 SECTION A-A 
 
 4x6" Oregon pine s hungers 
 
 l"x IE" Wood plafe. 
 
 IE x 12" Wood or concrete 
 mud sill. 
 
 END VIEW 
 Fig. 36. — Details of simple wooden bridge to span an irrigation ditch. Eedwood 
 is most suitable for wood mud sills. 
 
 Structures Used as Meters. — Certain devices may be adapted to use as meters 
 for indicating the flow. Where water is being sold, this feature is especially 
 useful. Suitable meters are those in which changes do not occur during use. 
 As a rule, therefore, they are confined to installations having permanent con- 
 
Farm Irrigation Structures 
 
 37 
 
 struction like that afforded by concrete and steel. Permanence is necessary so 
 that the meter will remain consistent — a condition not found in wooden struc- 
 tures where the boards warp, come loose, or are torn off. One can test any 
 structure to obtain a rating curve that will be useful for a short period. 
 Since, however, a test is difficult to make, it is not worth while unless the results 
 
 LONGITUDINAL SECTION 
 
 Fig. 37. — Inverted siphon used to pass ditch flow under road or similar obstacle. 
 
 Fig. 38. — Single-wall structure with thin metallic edges on upstream face of 
 opening, for accurate measurement of water. 
 
 are good for the life of the structure. A metal strip may be put on the edges of a 
 wooden structure (fig. 38) to make a full-contracted weir opening, and many 
 of the installations already described may be fitted to permit their use as 
 meters. To serve thus, they should follow a standard design, duplicating a 
 tested device for which tables of flow are available. California Agricultural 
 Experiment Station Bulletin 588 (8) gives performance tables and describes 
 many forms of measuring devices used in ditches. Construction details must 
 
38 
 
 California Experiment Station Circular 362 
 
 -»— Any gahe shape 
 
 (Circular or sou arc) 
 (wif+iour sharp edges) 
 
 View upsrreom 
 
 Submerged discharge 
 Flov 
 
 Side view 
 
 _, Mehal plo te v»iHi sharp 
 Bulkhead — P ,^9 e VP shre f r 
 
 II ^ Th ' 5 P'° fe n ° h f° rf ° f 
 
 ^^ design for fables 4 or 5) 
 
 Fig. 39.— Gate for metering water without use of sharp-edged facing. 
 
 Hole for aerating nappe 
 Metal plate on weir crest 
 
 pfH JR- 
 
 ▼ V SECTION A-A SECTION B"B 
 
 Fig. 40. — Gate in flume used as metering unit. Direct flow through flume to gate pre- 
 vents water from curving inward in passing through the gate; this eliminates "end con- 
 tractions." 
 
 be carefully reproduced so that behavior may duplicate that of the standard 
 devices. In contrast, the average field gate may be roughly put together so 
 that no two supposedly identical structures have exactly the same character- 
 istics. Nevertheless, installations of similar general shape will behave roughly 
 alike. Figures 39 and 40, with accompanying tables 4 and 5, give the approxi- 
 
Farm Irrigation Structures 
 
 39 
 
 TABLE 4 
 
 Approximate Flow through an Orifice Gate, as in Figure 39, without Sharp 
 
 Entrance Edges but with Some Contraction 
 
 Area, in 
 
 
 Cubic feet per second for various indicated heads, in feet (h in fig 
 
 ure 39)* 
 
 
 square feett 
 
 0.2 ft. 
 
 0.4 ft. 
 
 0.6 ft. 
 
 0.8 ft. 
 
 1.0 ft. 
 
 1.2 ft. 
 
 1.4 ft. 
 
 1.6 ft. 
 
 1.8 ft. 
 
 2.0 ft. 
 
 0.2 
 
 0.59 
 
 1.18 
 
 1.76 
 
 2.35 
 
 2.94 
 
 4.41 
 
 5.88 
 
 8.83 
 
 11.77 
 
 14.71 
 
 17.65 
 
 20.60 
 
 23.54 
 
 26.50 
 
 29.42 
 
 32.35 
 
 35.30 
 
 38.25 
 
 0.83 
 
 1.66 
 
 2.50 
 
 3.33 
 
 4.16 
 
 6.24 
 
 8.32 
 
 12.48 
 
 16.64 
 
 20.80 
 
 24.97 
 
 29.15 
 
 33.29 
 
 37.45 
 
 41.61 
 
 45.80 
 
 49.97 
 
 54.12 
 
 1.02 
 
 2.04 
 
 3 06 
 
 4.08 
 
 5.10 
 
 7.65 
 
 10.20 
 
 15.30 
 
 20.40 
 
 25.50 
 
 30.60 
 
 35.70 
 
 40.80 
 
 45.90 
 
 51.00 
 
 56.10 
 
 56.12 
 
 56.63 
 
 1.18 
 
 2.36 
 
 3.54 
 
 4.72 
 
 5.90 
 
 8.84 
 
 11.79 
 
 17.68 
 
 23.58 
 
 29.47 
 
 35.37 
 
 41.25 
 
 47.16 
 
 53.05 
 
 58.95 
 
 64.81 
 
 70.72 
 
 76.61 
 
 1.32 
 
 2.63 
 
 3.95 
 
 5 26 
 
 6.58 
 
 9.87 
 
 13.16 
 
 19.74 
 
 26.32 
 
 32.90 
 
 39.48 
 
 46.05 
 
 52.64 
 
 59.20 
 
 65.80 
 
 72.38 
 
 78.96 
 
 85.54 
 
 1.44 
 
 2.88 
 4.32 
 5.76 
 7.20 
 10.80 
 14.40 
 21.60 
 28.80 
 36.00 
 43.20 
 50.40 
 57.60 
 64.80 
 72.00 
 79.20 
 86.40 
 93.60 
 
 1.56 
 3.11 
 
 4.67 
 6.23 
 7.78 
 11.68 
 15.57 
 23.35 
 31.14 
 38.92 
 46.71 
 54.50 
 62.28 
 70.50 
 77.85 
 85.63 
 93.52 
 101.30 
 
 1.66 
 
 3.33 
 
 4.99 
 
 6.66 
 
 8.32 
 
 12.48 
 
 16.64 
 
 24.96 
 
 33.28 
 
 41.60 
 
 49.92 
 
 58.22 
 
 66.56 
 
 74.85 
 
 83.20 
 
 91.52 
 
 99.84 
 
 108.16 
 
 1.77 
 
 3.53 
 
 5.30 
 
 7.06 
 
 8.83 
 
 13.24 
 
 17.66 
 
 26.49 
 
 35.32 
 
 44.15 
 
 52.98 
 
 61.80 
 
 70.64 
 
 79.18 
 
 88.30 
 
 97.13 
 
 105.96 
 
 114.79 
 
 1.86 
 
 0.4 
 
 3.72 
 
 0.6 
 
 5.58 
 
 0.8 
 
 7.44 
 
 1.0 
 
 9 30 
 
 1.5 
 
 2.0 
 
 13.95 
 18.60 
 
 3.0 
 
 27.90 
 
 4.0 
 
 5.0 
 
 37.20 
 46.50 
 
 6.0 
 
 55.80 
 
 7.0 
 
 65.10 
 
 8.0 
 
 9.0 
 
 10.0 
 
 74.40 
 83.68 
 93.00 
 
 11.0 
 
 102.30 
 
 12.0 
 
 111.60 
 
 13.0... 
 
 120.90 
 
 
 
 * Quantity in cubic feet per second = Ca v' 2gh where C = 0.82, where g — 32.2, the acceleration of gravity, and 
 h = the head given at the top of the table above. 
 
 t To determine area of gates having circular cross section, square the diameter, in feet, and multiply by 
 0.7854. 
 
 TABLE 5 
 
 Approximate Flow per Foot of Width, through Gate with No End Contractions, 
 
 as in Figure 40* 
 
 Head in inches 
 (approximate) 
 
 Head 
 
 Cubic feet per second per foot of width under the given weir heights 
 (A in figure 40) 
 
 0.5 ft. 
 
 0.75 ft, 
 
 1.0 ft. 
 
 1.5 ft. 
 
 2.0 ft. 
 
 3.0 ft. 
 
 4.0 ft. 
 
 m- 
 
 4% 
 
 6... 
 
 m. 
 
 10% 
 12... 
 
 13%. 
 
 U%. 
 15... 
 16% 
 18... 
 
 0.10 
 0.20 
 0.30 
 0.40 
 0.50 
 0.60 
 0.70 
 0.80 
 0.90 
 1.00 
 1.10 
 1.20 
 1.30 
 1.40 
 1.50 
 
 0.111 
 
 0.312 
 
 0.583 
 
 0.916 
 
 1.31 
 
 1.76 
 
 2.26 
 
 2.82 
 
 3.43 
 
 4.11 
 
 0.110 
 0.307 
 0.569 
 
 0.888 
 
 1.26 
 
 1.68 
 
 2.14 
 
 2.66 
 
 3.22 
 
 3.82 
 
 4.48 
 
 5.17 
 
 0.109 
 
 0.305 
 
 0.562 
 
 0.873 
 
 1.23 
 
 1.64 
 
 2.08 
 
 2.58 
 
 3.11 
 
 3.68 
 
 4.30 
 
 4.95 
 
 5.64 
 
 6.48 
 
 7.15 
 
 0.109 
 
 0.302 
 
 0.555 
 
 0.858 
 
 1.21 
 
 1.60 
 
 2.03 
 
 2.49 
 
 3.00 
 
 3.54 
 
 4.12 
 
 4.72 
 
 5.36 
 
 6.04 
 
 6.75 
 
 0.108 
 
 0.300 
 
 0.552 
 
 0.851 
 
 1.20 
 
 1.58 
 
 2.00 
 
 2.45 
 
 2.95 
 
 3.47 
 
 4.02 
 
 4.61 
 
 5.23 
 
 5.87 
 
 6.56 
 
 0.108 
 
 0.299 
 
 0.548 
 
 0.844 
 
 1.18 
 
 1.56 
 
 1.97 
 
 2.41 
 
 2.89 
 
 3.40 
 
 3.93 
 
 4.50 
 
 5.09 
 
 5.71 
 
 6.36 
 
 0.108 
 
 0.299 
 
 0.545 
 
 0.840 
 
 1.18 
 
 1.55 
 
 1.95 
 
 2.39 
 
 2.86 
 
 3.36 
 
 3.89 
 
 4.44 
 
 5.02 
 
 5.62 
 
 6.26 
 
 Data from table 14, bulletin 588 (8). 
 
40 
 
 California Experiment Station Circular 362 
 
 mate flow through two rather typical forms of field gates. Using such tables, 
 the designer can approximate the required size of opening needed to pass a 
 given flow when a head loss is available. To be safe, he will build the opening 
 a little larger than the tables indicate, for his design might not perform quite so 
 well as the one upon which the tables were built. Sharp edges — that is, metallic 
 faces with beveled edges downstream (fig. 39) — actually restrict the flow 
 through such gates; and if these facings are present the openings must be 
 made at least 1.33 times as large as in the tables. In any field structure not built 
 strictly to some standard design details may vary in behavior 10 to 20 per cent 
 
 Fig. 41. — Automatic flow-control gate installed in canal. 
 
 from the figures given in these tables. Sharp edges of standard shape supply 
 precise metering data. Complete details on such structures have been pre- 
 viously reported (8) . 
 
 Automatic Gates. — The capacity to meter the flow to an area may simplify 
 the operation of a conducting ditch that may be capable of handling a certain 
 maximum safely. If conditions in the supply ditch are constant, one can set 
 the gate for that flow and leave it. But since constant conditions are never 
 encountered on some ditches, attempts have been made to put in automatic 
 gates (used as head gates as a rule) that will vary their opening as the supply- 
 ditch water levels vary. An effective automatic structure of this kind will save 
 the operator much grief. It will serve also — though less effectively — as a meter. 
 Figure 41 illustrates a typical automatic gate installation — float-operated lever 
 that moves the gate board, the float being located in a sump, or a well, con- 
 nected by a pipe to the main ditch. The float could be connected to the supply 
 ditch or to the secondary ditch, whichever is intended to receive the control. 
 Further refinements include electrical contacts on the float system, with an 
 electric motor drive on the gate. These latter installations can be made to 
 operate almost perfectly with widely fluctuating flow levels in the supply ditch, 
 keeping constant the flow in the ditch being supplied. Most automatic gates 
 without supplementary electrical equipment cannot be completely effective 
 
Farm Irrigation Structures 
 
 41 
 
 during severe fluctuations in the supply-ditch water levels. Almost any slide 
 gate can be made automatic according to the general principles pictured in 
 figure 41. 
 
 Materials Used in Structures. — Various materials have been described and 
 pictured in the figures in the preceding discussion of ditch or canal structures. 
 Occasionally, one or the other material has been noted as best for a special use. 
 Concrete and steel are, as might be expected, more permanent than wood. In 
 localities where ditch grades shift and fields wash regardless of the care taken 
 to prevent these changes, the concrete structure may be too permanent. Little 
 can be done with a permanent structure when soil levels have receded or when 
 
 TABLE 6 
 Diameters, Areas, Weights, and Cost of Concrete Pipe Installed* 
 
 Diameter of pipe 
 
 Normal 
 
 length of 
 
 each section 
 
 Cross-section areas 
 
 Weight 
 
 Approximate 
 cost, installed 
 and backfilled, 
 
 per foot 
 (24-inch cover) 
 
 inches 
 6 
 
 inches 
 
 24 
 
 30 
 30 
 36 
 36 
 36 
 36 
 36 
 
 square 
 inches 
 28.28 
 50.24 
 78.50 
 113.04 
 153.86 
 201.00 
 254.34 
 314.00 
 
 square 
 
 feet 
 0.1964 
 0.3489 
 0.5451 
 0.7854 
 1.0684 
 1.3947 
 1.7661 
 2.1804 
 
 pounds 
 per foot 
 
 21 
 
 28 
 
 38 
 
 48 
 
 71 
 
 91 
 108 
 128 
 
 dollars 
 0.25 to 0.42 
 
 8 
 
 0.27 to 0.42 
 
 10 
 
 0.30 to 0.45 
 
 12 
 
 0.35 to 0.53 
 
 14 
 
 0.51 to 0.65 
 
 16 
 
 0.75 to 0.79 
 
 18 
 
 0.95 to 1.04 
 
 20 
 
 1.25 to 1.35 
 
 
 
 * Pots, hydrants, stands, gates, and gate installations are all extra. 
 
 the unusual has occurred and levels have risen. A wood structure, on the con- 
 trary, can be remodeled somewhat if conditions change. These statements are 
 not made to belittle concrete or steel structures, but to stress the fact that 
 permanent structures are best where conditions remain constant. 
 
 UNDERGROUND PIPE DISTRIBUTION SYSTEM 
 
 Irrigation water is distributed successfully by underground systems using 
 either steel or concrete pipe — more often the latter. Pipe offers certain advan- 
 tages: fields unobstructed by ditches; a capacity to disregard grade within 
 reasonable limits; elimination of open ditches to prevent the spread of weed 
 seeds through water transportation; practical elimination of seepage and 
 evaporation losses; and permanence. Its chief disadvantage is the initial cost, 
 which is higher than for ditches of equal capacity. Unobstructed fields simplify 
 cultivation and the movement of equipment, and save valuable land areas for 
 productive use. Within reasonable limits, a pipe line can go straight over 
 ridges and across swales, whereas a ditch must be run around such obstruc- 
 tions, unless expensive structures are built to avoid such detours. The elimina- 
 tion of seepage into porous soils from unlined ditches may pay for the more 
 expensive underground pipe system. Evaporation losses from ditches, although 
 
42 California Experiment Station Circular 362 
 
 usually not great, may well be considered. The pipe system eliminates this 
 loss. Most unlined ditches either must be remade often, under normal cultiva- 
 tion practice, or cleaned and reshaped occasionally. This work is eliminated 
 by the subsurface pipe systems. 
 
 Most concrete irrigation pipe is bought from manufacturers who produce 
 and stock large quantities of the standard sizes. Thus the consumer benefits 
 by mass production, and the cost of forms is spread over a large volume of the 
 product. Special equipment is required to produce even a few hundred feet of 
 a given size of pipe in 30-inch lengths — the usual dimension; and only rarefy 
 
 Fig. 42. Making concrete pipe in place in the field. Backfilling follows 
 removal of outside forms immediately. 
 
 can the individual afford the time to make his own. Since the process has been 
 described elsewhere (,9) and can be observed in local yards throughout Cali- 
 fornia, it will not be discussed here. Table 6 lists the various stock sizes of this 
 pipe, its weight, and its approximate cost. 
 
 Ditching machines are usually employed to excavate the trench for the pipe. 
 Then the trench is trimmed by hand to smooth out irregularities and to put 
 the bottom on the exact grade, which may be a straight line, or a series of 
 straight lines. The top of the pipe is kept safely below any anticipated cultiva- 
 tion operations (18 to 30 inches). An additional purpose in placing the pipe 
 low is to reduce temperature and moisture changes, thereby reducing expan- 
 sion and contraction. Pipe joints are connected by filling the recessed end of 
 one pipe with a cement mortar and pushing the tongue end of the second pipe 
 into the first, squeezing the mortar in the recess partly out of the joint. In 
 
Farm Irrigation Structures 43 
 
 small pipes the excess mortar inside is wiped off smooth with a wet sack or 
 brush, and the outside extrusion of mortar may be supplemented with more 
 to make a ring or band around the joint. In large pipes, 24 inches or greater, 
 a man often works inside, wiping and cleaning these joints; the outside of the 
 joint is then banded as with smaller-diameter pipe. This work is often done 
 as much as 24 hours after the joints have been made, to give settlement cracks 
 a chance to form and the man the opportunity to seal them. This practice, to a 
 great extent, eliminates leaks. Concrete pipe may be constructed on the job, 
 using the ditch bottom as a form and using removable inside and top forms 
 (fig. 42). Such a pipe is poured continuously and has no joints. In the larger 
 pipes (30 inches or greater) so made, reinforcing may be incorporated as the 
 construction progresses. The prefabricated pipe discussed above is the most 
 popular. Since most concrete lines are laid by experienced crews from pipe 
 yards, the finished job is usually satisfactory. The tabulation below shows 
 the approximate cost of concrete pipe cast in the field as a continuous unit 
 (sometimes called monolithic pipe, meaning all one piece). The smaller sizes 
 are not made in the field. 
 
 Size, inches Cost per foot, dollars 
 
 24 1.00 to 1.13 
 
 30 1.25 to 1.75 
 
 36 1.80 to 2.25 
 
 42 3.00 to 3.50 
 
 Flow through a pipe meets resistance; the rougher the pipe, the greater that 
 resistance. The surface of concrete pipe resembles that of a concrete-lined 
 ditch; and the same friction coefficient n = 0.015 can be applied to both. As 
 a rule, steel pipe is smoother than concrete and would have a smaller value of 
 n, or would pass more water for the same cross section. The pipe and the ditch 
 are dissimilar in that the pipe is usually full, whereas the ditch has varying 
 depth with varying flow. The hydraulic radius, R, in a pipe flowing full does 
 not change, in contrast to the shifting value of R for open channels. The 
 hydraulic radius of a full pipe equals 
 
 „ _ A area irr 2 _ r 
 
 P perimeter 2wr 2 
 
 Because the hydraulic radius is unvarying for any pipe size as long as the pipe 
 is full of water, this R value need not be checked each time, and the flow- 
 characteristic curves (figs. 43 and 44) can be simplified as compared with 
 open-channel curve sheets (figs. 4, 5, and 6). Figure 43 gives the friction loss 
 in ten sizes of steel pipe for a range of flows and is based, as noted in the figure, 
 on Williams and Hazen's formula (10) for their coefficient of friction, C = 100. 
 This is a conservative value for the friction that would correspond to about 
 n = 0.013 in ditch design. Figure 44 gives the friction loss for the same ten 
 sizes of concrete pipe, but supplies two curves for each pipe size from data pub- 
 lished by Scobey (1, 11). The heavy or left-hand curve for each pipe size is 
 
44 
 
 California Experiment Station Circular 362 
 
 safe for the normal dry-mix pipe where the joints are not too well finished in 
 the field. The right-hand or dash-line is satisfactory for wet-mix pipe, which is 
 naturally smoother inside, and assumes that the field joints are well finished. 
 The solid lines are safe for general design work and will be used in the following 
 discussion. To use either chart (fig. 43 or 44) with the flow given, start at 3 c.f .s. 
 on the bottom of figure 43, for example, and travel vertically to, say, the 
 
 10. 
 9. 
 
 
 
 
 
 
 t 
 
 
 
 a 
 
 10 12 14- 
 
 L —h-h 
 
 16 
 
 -A 
 
 -4- 
 
 K> 22 24 
 7-f-J-- 
 
 
 
 
 
 
 
 ■ / 
 
 
 v 
 
 / 
 
 —i—h 
 
 tH 
 
 r- 
 
 ^t-l- 
 
 
 
 
 
 
 
 / 
 
 
 / 
 
 i r 
 
 / / 
 
 / 
 
 / 
 
 _J T _ 
 
 
 
 
 
 
 
 / 
 
 
 / 
 
 / 
 
 
 
 
 Tl - 
 
 
 
 
 
 
 
 / 
 
 / 
 
 
 
 ' / / 
 
 
 i / 
 
 r~r 
 
 
 
 
 
 
 
 / 
 
 / 
 
 
 / / 
 
 / / 
 
 V 
 
 / 
 
 / 
 
 
 
 
 
 
 
 / / 
 
 / 
 
 / 
 
 T 
 
 f / / 
 
 / 
 
 7 
 
 
 
 
 
 
 
 / 
 
 ' / 
 
 / 
 
 / 
 
 // 
 
 //// 
 
 7 
 
 / 
 
 
 .9 
 
 
 
 
 
 -/- 
 
 
 4— 
 
 -A 
 
 L 
 
 L'—L-L 
 
 
 
 
 Oi -8 
 
 
 
 
 
 — L- 
 
 -f- -j 
 
 
 t- 
 
 7 7 
 
 --/■ / / ; 
 
 J / / / 
 
 
 
 
 if 6 
 
 
 
 
 
 / 
 
 / / 
 
 / 
 
 
 / ' 
 
 / / / / 
 
 
 
 
 
 
 
 
 
 1 
 
 i / 
 
 
 / 
 
 / 
 
 / / / / 
 
 
 
 
 £ .4 
 
 
 
 
 / 
 
 
 f / 
 
 / / 
 
 
 / 
 
 / / 
 
 / / / 
 
 
 
 
 8, 
 
 
 
 
 / 
 
 
 
 1 
 / 
 
 / 
 
 
 / // 
 
 
 
 
 
 
 
 / 
 
 
 r 
 
 / 
 
 j / 
 
 / 
 
 / 
 
 / 
 
 / / 
 
 
 
 
 Cy 
 
 
 
 / 
 
 / 
 
 / 
 
 / / j 
 / / / 
 
 / 
 
 / 
 
 //, 
 
 / 
 
 / 
 
 
 
 
 °-° 9 
 
 
 
 
 hA 
 
 t 
 
 
 /—* 
 
 -v 
 
 -/-/■ 
 
 
 
 
 
 c .0B 
 
 
 
 
 i — 
 
 7 
 
 
 —t 
 
 ~r 
 
 T7 
 
 
 
 
 
 
 
 
 J 
 
 
 r 
 
 1 / / 
 
 
 ! 
 
 ^t^ 
 
 
 
 
 
 
 
 
 
 
 / 
 
 / / / 
 
 / / 
 
 / 
 
 / 
 
 
 
 
 
 U. 
 
 
 
 
 
 / / 
 
 / / / 
 
 / 
 
 / 
 
 / 
 
 
 
 
 
 
 
 / t 
 
 f 
 
 / 
 
 f 
 
 / / / 
 
 // 
 
 / 
 
 
 
 
 
 
 
 
 / / 
 I / 
 
 / 
 
 / 
 
 / / 
 
 / / / / 
 / / / 
 
 // 
 
 / 
 
 
 
 
 
 
 .01 
 
 / 
 
 / 
 
 / / 
 
 / 
 
 / 
 
 7 
 
 >'//, 
 
 / 
 
 
 
 
 
 
 
 .04 .05 .06X17.08.09.1 
 
 y. 5. 6. 7. 8.9.10. 
 
 Cubic Feet Per Second 
 
 30. 40. 50. 60.70.80.90.100. 
 
 Fig. 43. — Flow of water in steel pipe, based on Williams and Hazen's C = 100. The 
 pipe size given is nominal inside diameter. (See reading-reference 8.) 
 
 12-inch pipe size. The friction loss or head loss or hydraulic gradient necessary 
 for the steel pipe is 0.7 foot per 100 feet or 0.007 foot per foot; and on figure 44, 
 following the same procedure, the head loss in the dry-mix 12-inch concrete 
 pipe for 3 c.f .s. flow is 0.68 foot per 100 feet = 0.0068 foot per foot. 
 
 The fact that a pipe line can disregard grade changes within reasonable 
 limits is very important too. If an open ditch is run down into a swale, there is 
 no way to bring the water up out of that swale again except by pumping. A 
 pipe being closed does not lose the potential elevation by descent into the 
 swale, but uses the water pressure on the high side to force the water up and 
 out of the low side of the swale. 
 
 Air trapped in the water tends to collect at high points on a pipe line; one 
 must put small-diameter vents at such points, raising them high enough above 
 
Farm Irrigation Structures 
 
 45 
 
 the surface of the ground so that they will not overflow. Without the vents, 
 the collected air acts as a blockade, reducing the flow. 
 
 Figure 45 illustrates the conditions along a typical pipe line of uniform size 
 throughout. The pump at the left discharges into a box, A ; and the pipe proper 
 takes off near the bottom of that box. The pipe can be assumed to run parallel 
 with the ground surface at a distance 30 inches below the surface. It climbs 
 
 
 
 
 
 
 
 
 
 / 
 
 
 f 
 
 zp. -i 
 
 / / > // 
 
 I'M 1 
 
 n 
 
 f^^7 
 
 
 
 
 
 
 
 
 
 
 
 J 
 
 tZ 1 
 
 ' p 
 
 / ' V / 
 
 fj 
 
 -*7Z 
 
 
 
 
 
 
 
 
 
 ft 
 
 
 
 n 
 
 p // 
 
 f> h t 
 
 J 
 
 -LL 
 
 
 
 
 
 
 
 
 ft 
 
 
 fj 
 
 
 p i 
 
 \ A > 
 
 v 
 
 ut 
 
 
 
 
 
 
 
 
 
 ft 
 
 J 
 
 i 
 
 It 
 
 t; // i 
 
 f P\ / 
 
 i / 
 
 u __ 
 
 
 
 
 
 
 
 
 
 //' 
 
 ft 
 
 i 
 
 // 
 
 a // // 
 
 //I// 
 
 i> 
 
 / 
 
 
 
 
 
 
 
 
 f<a 
 
 T 
 
 
 A' 
 f / 
 
 1 1 if / 
 
 &/&& 
 
 $ 
 
 
 
 
 
 
 
 
 
 
 p 
 
 // 
 
 i 
 
 ii 
 
 
 J f J/J f 
 
 
 # 
 
 i 
 
 
 V 
 
 
 
 
 
 
 
 1 
 
 If 
 
 // 
 /i 
 
 u 
 
 i 
 
 i 1 '. 
 
 / 
 
 v-'M 
 
 /f f i /// 
 
 '// 
 
 'ii 
 
 // 
 l 
 
 
 
 "■* q 
 
 
 
 
 
 
 
 Pt 
 
 / 1 
 
 
 ! 
 
 £ 
 
 // // /// / 
 
 ft 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 I 
 
 _^^ 
 
 14 i i/r / 
 
 i / 
 
 
 
 
 
 
 
 
 
 
 
 7 
 
 
 
 i ; 
 
 ^Zt-ji 
 
 f/fff/i 
 
 / 
 
 
 
 
 ? - 6 
 
 
 
 
 
 
 / 
 
 
 
 i 
 
 i 
 
 
 i ' fff U 
 
 
 
 
 
 *.!, 
 
 
 
 
 
 
 // 
 
 
 n ti- 
 
 , 
 
 t 
 
 / 
 
 ffff/ff/ 
 
 
 
 
 
 O 
 O .4 
 
 
 
 
 
 ( 
 
 '/ 
 
 
 ff // 
 
 / 
 
 / 
 
 in 
 
 ///ffp 
 
 
 
 
 
 V. 
 
 
 
 
 
 / 
 
 
 / 
 
 ' // 
 
 / 
 
 // 
 
 ///< 
 
 'P//// 
 
 
 
 
 
 a. ; 
 
 1> 2 
 
 
 
 
 
 i 
 
 'I 
 
 
 
 /i / 
 
 V! II 
 
 '// 
 
 
 in 
 
 P V/ 
 f (// 
 
 
 
 
 
 C 
 
 o 
 
 
 / 
 
 f — 
 
 
 / 
 
 L. / 
 
 - -x-t-r — t~I 
 
 ' /> 
 if 
 
 'A 
 
 
 Hi/ 
 
 /// 
 
 // 
 
 
 
 
 
 §.09 
 
 
 
 ft 
 
 
 
 
 P 
 
 ---/-!■ 4-t- 
 
 // 
 
 P 
 
 vp-t 
 
 4 
 
 
 
 
 
 
 
 
 II 
 
 
 1 i 
 
 
 ', 
 
 f 
 
 
 i 
 
 tij-Q 
 
 
 
 
 
 
 
 
 
 // 
 
 
 // 
 
 
 
 ff / f 
 
 i / 
 
 7 
 
 p 
 
 
 
 
 
 
 .05 
 
 
 
 ft 
 
 / 
 
 / 
 
 
 fj 
 
 /> ff I 
 
 ; P 
 
 f 
 
 * < 
 
 
 
 
 
 
 .04 
 
 
 
 // 
 
 /' 
 
 
 / 
 
 
 > P 
 
 1 ll 1 1 
 
 / 1 1 
 
 ' f 
 
 ff/ 
 
 
 
 
 
 
 .03 
 
 
 
 
 1 
 
 
 // 
 
 // 
 
 1 / P / 
 
 1 ! II I 
 
 / p 
 
 / 
 
 
 
 
 
 
 
 
 .02 
 
 
 
 '/ 1 
 
 I ' 1 
 
 7 
 
 // 
 
 
 Ii , 
 
 P ffh 
 11 //// 
 
 1 I ' I / ) 
 
 ti 
 
 l /\ 
 
 i 
 
 
 
 
 
 
 .01 
 
 
 // 
 
 f fl 
 
 A / 
 ii 
 ii 
 /i 
 
 ' 
 
 1 / 
 
 / 
 
 / 
 
 \ 
 
 / 
 
 ff 
 
 ' 'l Ii 
 ' I 1 
 
 'Pi 
 
 w 
 
 V 
 
 
 
 
 
 
 
 .04 .05 jD6.07j08fl9J .2 .3 .4 .3 .6.7.8.9 1. 2. 3. 4. 5. 6.7.8.9.10. 20. 30. 40. 50. 60.70. 80. 9Q.IOO. 
 
 Flow in Cubic Feet Per 5econd 
 
 Fig. 44. — Flow of water in concrete pipe, based on F. C. Scobey's data covering dry-mix 
 (heavy line) and wet-mix (dashed line) pipes. (See reading reference 11.) 
 
 from A to B, where there is a ridge crest, then drops toward C and D. At C is 
 located another box, or stand, on the line; and at D is a swale with the right- 
 hand side higher than the left hand. From the high side of the swale the pipe 
 again descends to the box, E, where the flow is being discharged. When the 
 pump is running, water stands in boxes A, C, and E; and a straight line will 
 pass through these surfaces, which would correspond to the surface of the water 
 in an open ditch of uniform cross section and construction, handling the pump 
 discharge between the pump and E. This line is also the hydraulic gradient of 
 that size pipe line for the flow carried. Obviously, a ditch constructed on this 
 same line across the piece of land shown would encounter many difficulties. 
 It must be built up very high at A and across D and again presumably at E. 
 The pipe has the same hydraulic gradient a ditch could use, but disregards the 
 
46 
 
 California Experiment Station Circular 362 
 
 changing grade of the ground. This grade or hydraulic gradient depends on 
 the friction coefficient of the pipe. The height of the hydraulic grade line above 
 the pipe at any point is the static water pressure in the line at that point in 
 feet of water (2.31 feet of water = 1 pound per square inch). If the pipe is 
 rough, the slope from water surface at A to water surface at E will be larger 
 than if the pipe is smooth. The ditch behaves likewise. If the pipe distance from 
 A to E is 1,500 feet and the elevation difference between the two water surfaces 
 is 3 feet, the hydraulic gradient is 2 feet per 1,000 feet = 0.2 foot per 100 feet; 
 and the flow passing is the maximum which will pass through the pipe with that 
 loss in head or with that hydraulic gradient. If the pump discharge is increased 
 and the water level at E stays constant, the pipe must carry more water, and 
 this will involve more friction. A steeper gradient will then be necessary. To 
 provide it the water at A must rise higher in the box. The water at C also rises 
 
 Pump 
 
 A B 
 
 Water level 
 
 — m=A=z. _* vdra u I i c 
 
 Soil surface 
 
 W ater level 
 
 Fig. 45. — Typical pipe-line installation. 
 
 to keep the line between water surfaces at A and E straight. The new gradient 
 might have a slope of 3 feet per 1,000 feet = 0.3 foot per 100 feet, and the flow 
 through the pipe would be the maximum for this gradient. Another use of a 
 figure plotting the hydraulic gradient is to be sure the pipe line will run full. 
 Below the hydraulic grade line there is water pressure in the pipe equal to the 
 vertical distance from the pipe to the line. If the top of the pipe goes above 
 this hydraulic gradient, it goes above the line of zero pressure, and the pipe 
 runs partly full or not at all, unless a sufficient partial vacuum exists along 
 that section of the line. Trouble will result; the condition should be avoided. 
 In addition, any stand along a pipe line must be higher than the hydraulic 
 gradient or it will overflow. Because the designer can draw the hydraulic 
 gradient before constructing the line, he can avoid locating any section of his 
 pipe above that critical line and can dimension all stands with safety. Figures 
 43 and 44 illustrate the friction-loss (feet loss per 100 feet of line) capacity 
 characteristics of steel and concrete. These curves, being based on average 
 pipe conditions, are safe for general application to farm problems. As in ditch 
 or canal jobs, where a considerable installation is proposed, more accurate and 
 economical design might be secured from more specialized data (1,11). 
 
 Some typical problems will illustrate the use of these charts (figs. 43 and 44) 
 and may help to clarify the discussion of pipe-line behavior. 
 
 Problem 1. — What size concrete pipe line will be required to carry 1,500 
 g.p.m. between points A and B (fig. 46) located 1,540 feet apart? Elevation 
 of ground surface at A = 56.00 feet; at B, 64.46 feet. Water stands 10 feet 
 
Farm Irrigation Structures 
 
 47 
 
 above the ground surface at standpipe at A, and water discharges at level of 
 ground surface at point B. Water-surface elevation at A = 56 + 10 = 66 feet, 
 and at B = 64.46 feet, or a difference in elevation of 1.54 feet. The length of 
 pipe line = 1,540 feet, and the drop in 1,540 feet = 1.54 feet. The gradient 
 = 1.54 ^ 1,540 - 1 foot -h 1,000 = 0.1 foot per 100 feet. 
 
 This problem resolves itself to the question of what size line is required to 
 handle 1,500 g.p.m. when the available friction loss can be 0.1 foot per 100 
 feet = hydraulic gradient. Referring to figure 44, the left-hand side gives fric- 
 tion loss per 100 feet of line. Then follow horizontally on the 0.1 foot per 100- 
 foot line to the right till the vertical line from 1,500 g.p.m. (3.33 c.f.s.) is inter- 
 
 Suppty- 
 
 — Elev. woter-66.0 ,. . 
 
 H ydrou lie grodien r — 
 
 El. soil- 64.46 
 
 B 
 
 Surface soil El. 56.0 
 
 Fig. 46. — Detail of pipe line for piping problem 1. 
 
 T 
 _L 
 
 SuppK^ 
 
 El. water -67' 
 El. soil-59' 
 
 B 
 El. wafer -62.' — . 
 
 jCI 
 
 El. soil -57' 
 
 12' 
 
 P«pe line 2100' long 
 
 Use 
 
 Fig. 47. — Detail of pipe line for piping problem 2. 
 
 sec ted (see fig. 3). The line for the 18-inch pipe passes diagonally through the 
 intersection. An 18-inch pipe will be required. This problem illustrates several 
 interesting points. First, the hydraulic gradient, which is the slope of pressure 
 loss along the pipe, disregards soil-surface levels and concerns itself only with 
 free- water surface levels. Second, the standpipe at A must be built somewhat 
 higher than 10 feet above the ground surface there, or it would be subject 
 to frequent overflows. Incidentally, a standpipe 10 feet or higher is not 
 always satisfactory if valves must be operated from the top by means of a 
 ladder. In the present case, the land required the excess height on this stand- 
 pipe to permit the desired flow through an 18-inch pipe. 
 
 Problem 2. — It has become desirable to increase the flow through a 12-inch 
 concrete line (fig. 47) as much as possible by raising a certain low stand at A 
 to a clear working height of 8 feet above the surface of the ground. The next 
 stand at B, downstream, will spill over its top 5.2 feet above the ground surface 
 if too much is passed beyond it; this is assuming that B is the limiting stand, 
 incapable of being raised for some reason. The height of ground at A =59 feet ; 
 at B, 57 feet. The distance between A and B = 2,100 feet. Assume a maximum 
 safe rise of water at box B = 5 feet above the soil surface. (This leaves only 
 
48 California Experiment Station Circular 362 
 
 0.2 foot freeboard; except in this extreme case, this is less than should be 
 allowed.) Here is the solution: 
 
 Elevation water (maximum) at A = 67 feet 
 Elevation water 0.2 foot under top stand B = 62 feet 
 Elevation difference, A to B, = 5 feet 
 
 5 
 
 Hydraulic gradient = — — = 2.38 feet per 1,000 feet = 0.238 foot per 100 feet. 
 2, 100 
 
 From figure 44, left-hand side, giving friction loss per 100 feet of line, follow 
 value 0.238 foot per 100 feet to right to intersection with 12-inch pipe size 
 diagonal line; then follow vertically from this intersection to the bottom fine 
 of the figure where capacity of the line is given in cubic feet per second. Under 
 these conditions, a 12-inch line will handle 1.78 or 1.8 c.f.s., approximately, 
 or 800 g.p.m. (from fig. 3) . 
 
 B 
 
 El. water-62' 
 
 -Jftftjufc QroW . C 
 
 ~^^-,_ EL wafer-40' 
 
 Dil-ch-f 
 Fig. 48. — Detail of pipe line for piping problem 3. 
 
 In this problem we note that, on a long pipe line, the flow that can be accom- 
 modated in any section depends often on the flow through some other section 
 of the pipe. In other words, if something downstream from B in this problem 
 did not cause the water to back up at B, there would be a steeper gradient 
 from A to B. This fact is illustrated in problem 3, which follows. 
 
 Problem S. — All the water entering B from A leaves B through a 10-inch 
 steel pipe discharging into a ditch below its water surface, which is at eleva- 
 tion 40 (see fig. 48, an extension of fig. 47). How long is the 10-inch pipe? 
 
 The flow into B = 1.8 c.f.s. This same flow must go on or spill over the top 
 of the stand at B; and so 0.2 foot freeboard was kept in the hope of preventing 
 this spill. The flow from B to the ditch = 1.8 c.f.s. The gross drop equals eleva- 
 tion of water surface at B (62 feet) less water-surface elevation in ditch (40 
 feet), or 22 feet. Using figure 43 for flow through steel pipe and starting from 
 the bottom line of capacity at the point corresponding to 1.78 c.f.s., follow 
 a vertical to the 10-inch pipe size diagonal line; then follow a horizontal line 
 to the left side of the figure to read the corresponding friction loss in feet per 
 100-feet of pipe. The necessary loss to pass this flow through the 10-inch pipe 
 is 0.66 foot per 100 feet. The loss necessary per 100 feet of line = 0.66 foot 
 equals the hydraulic gradient, and the gross loss in this 10-inch line = 22 feet. 
 
 22 
 
 The line, therefore, must be hundred feet long to accumulate this loss, 
 
 0.66 
 
 or about 3,300 feet of 10-inch line. 
 
Farm Irrigation Structures 
 
 49 
 
 Problem 4. — How high will the water stand at point A (see fig. 49) if the 
 14-inch concrete line leading from A must pass over a ridge before descending 
 on the other side under the following conditions? Elevation top of inside of 
 pipe at A = 57 feet ; elevation top of inside of pipe at B (top of ridge) = 62 feet; 
 pipe distance A to B = 1,350 feet; and flow from A to B = 840 g.p.m. = 1.87 
 c.f.s. (from fig. 3). In addition, there is an air vent at B which would prevent 
 the line from acting as a siphon, and would prevent overflowing. 
 
 Though this problem reverses the previous ones, the reasoning remains 
 similar. If the pipe is full to the top at B and is standing at A to keep the 
 section A to B filled, then the level of the water at A and that at the top of the 
 pipe at B will be the same, or 62 feet. The hydraulic gradient is zero, there 
 being no flow. As soon as water starts to move from A to and past B, friction 
 results, and a hydraulic gradient is necessary. If 1.87 c.f.s. is to pass from A 
 
 El. 57! 
 
 Z\. water surface ? 
 
 B 
 
 7550' ° f 
 
 Fig. 49. — Detail of pipe line for piping problem 4. 
 
 to B, the elevation of the water at A must be built up above the level of stand- 
 ing water (elevation, 62 feet) far enough to give the required hydraulic gra- 
 dient. Using figure 44 for concrete pipe for the flow 1.87 c.f.s., on the bottom 
 line of the figure, move vertically to the solid diagonal 14-inch pipe line; and 
 from that intersection proceed on a horizontal line to the left-hand side of the 
 figure. Here the friction loss per 100 feet of line is read as about 0.12 foot. Since 
 the distance from A to B is 1,350 feet, the total friction for that distance 
 
 would be 
 
 0.12 
 100 
 
 X 1,350 = 0.0012 X 1,350 = 1.62 feet. The water must rise 
 
 about 1.6 feet above the 62-foot elevation in stand A to put the required 
 amount through the 14-inch line to B. If the line from B onward to the right 
 is also 14 inches and slopes at the rate of 0.12 foot per 100 feet or faster, this 
 flow will pass along it freely, with the grade meeting or exceeding the needed 
 hydraulic gradient. In other words, under those conditions the line from B 
 onward descends as fast as the flow requires, or faster. If, on the contrary, the 
 grade from B on is less than the friction loss resulting from passage of this 
 volume of water, the flow will cease to pass on freely by B and will react to the 
 resistance caused by friction. This reaction will be observed at A, the first 
 open stand behind the resistance. If the needed additional gradient totals 5 
 feet from B on to the point of escape, then the water in the stand at A will 
 
50 California Experiment Station Circular 362 
 
 rise 5 feet more, supplying the necessary head for the steeper gradient. If 
 stand A is not high enough to permit an additional 5-foot rise of water in it, 
 then it will spill the excess it cannot deliver; and the line will settle down to 
 the passage of what the available hydraulic gradient at the elevation of spill 
 permits — a smaller volume than the 1.87 c.f.s. it can handle from A to B. A 
 pipe with occasional vents on it is like a chain, no stronger than its weakest 
 link : the line as a whole can handle no more flow than the poorest section will 
 carry. For this reason, design and not chance governs the accurate selection 
 of pipe sizes, lengths of run, and gradients to accomplish the construction of a 
 balanced distribution system. Since a closed pipe line without open standpipes 
 acts as a unit, excess grade at one section can be made available at some point 
 below. As the net result, such a line behaves like any short line between open 
 stands; that is, the true hydraulic gradient from inlet- to outlet-water sur- 
 faces governs the flow. The pipe should not pass over that gradient at any 
 point; otherwise the flow will suffer because the pipe tends to run partly full. 
 
 A buried pipe line is difficult to clean. If possible, the water it carries should 
 not contain solid matter (sand, silt, vegetation). This foreign material reduces 
 the net cross-section area in the pipe at the point of a concentration. Because 
 of this reduction in area, the velocity must be greater to accommodate the 
 same flow; and the result is an increased friction loss. A system that functioned 
 perfectly before trash accumulated may later become unoperable. Once fouled, 
 the pipe line can be returned to its original operation performance only when 
 the solids have been sluiced out. This cleaning process may be difficult and 
 expensive; precautions to eliminate foreign material in advance may save 
 much expense at a later date. Incidentally, certain chemical fertilizers injure 
 concrete pipe if transmitted through it to the fields \ they should be handled 
 with care (14)- 
 
 Control Devices on Pipe Lines. — The control devices used on pipe lines have 
 two major purposes: protection of the line itself, and effective delivery of the 
 water. Certain devices satisfy both these requirements; others have only a 
 single usefulness. Standard concrete irrigation pipe is not reinforced, so it 
 cannot be used where high internal pressure is possible. Its weakness comes 
 chiefly from its lack of reinforcement. To protect the usual concrete pipe 
 systems, open standpipes are commonly used. When these are used, no water 
 pressure greater than the spill height of these stands can be reached. In addi- 
 tion, air trapped in the line can escape at such stands. Trapped air can dam 
 up a pipe if it cannot escape, as on top of the ridge illustrated at B in figure 49 
 of problem 4. If a standpipe were located at B in this figure, or if one of the 
 small commercial air-release valves as suggested (see fig. 50) were put there, 
 the air could escape, thus eliminating the hazard, and also possibly preventing 
 siphoning — a feature which might be favorable at that point on occasions. A 
 pipe like the one in figure 41 might run partly full at B if air were trapped as 
 a large bubble at the highest point of the pipe on the crest of the ridge. The 
 possible flow in the pipe line would then be cut down; and, if the 1.87 c.f.s. 
 
Farm Irrigation Structures 
 
 51 
 
 were supplied at stand A, the design would seem to fail. Bleeding of the trapped 
 air would eliminate this block and enable the design conditions to be met. 
 
 If a pump discharges directly into a concrete line without any air relief 
 nearby, the advancing stream may encounter an air pocket of the type just 
 
 ■4— Relief valve 
 
 Boll Floods when water 
 enters and seals vents 
 in spherical top. 
 
 Iron pipe riser to put valve 
 few inches to foorormore 
 above ground surface. 
 
 Cement and sand potch 
 
 Concrete pipe 
 
 Hole picked into pipe 
 
 Fig. 50. — Small ball float-relief valve for venting air on pipe lines. 
 
 described and, being unable to pass quickly, may cause a sudden and severe 
 shock. A sharp bend in the pipe can effect a similar shock condition. Because, 
 with this "water hammer," the stresses are many times greater than normal, 
 and pipe or joints often fail when a pump is connected directly to a concrete 
 line. A high standpipe at the point of connection with the pump will help to 
 eliminate water hammer. As an additional precaution, some sort of slip-joint 
 connection may be inserted in the discharge pipe between the pump and the 
 
,52 
 
 California Experiment Station Circular 362 
 
 standpipe. Otherwise expansion and contraction movements may cause 
 broken standpipe, or shifting of the pump, or both. Where the pump lifts 
 considerable amounts of sand, the safest practice is to pass the pump dis- 
 charge through a large, closed concrete or metal box where the velocity will be 
 much reduced and the sand will settle out before entering the pipe line. Such 
 a box is called a sand trap. Without a sand trap, under these conditions, the 
 pipe system may become clogged and, as a result, useless. 
 
 Small-diameter stands (24 inches or less) have value as air-relief points and 
 limited value as surge chambers. Small-diameter, light-weight pipe (2 to 4 
 
 inches) can be used as air vents only. 
 Larger stands (over 24-inch diameter 
 or rectangular sections) will serve sev- 
 eral purposes. They relieve trapped air 
 and, being larger, absorb surges more 
 easily. If a block occurs down stream, 
 the water rises more slowly in them, 
 and the necessary hydraulic gradient is 
 arrived at gradually. In addition, larger 
 standpipes provide sufficient diameter 
 to admit several lead-off or take-off 
 pipes if needed. When several take- 
 off pipes are installed at a stand, valves 
 are desirable at the inlet of each ; they 
 are generally located inside the stand- 
 pipe (fig. 51), with valves at the stand 
 for every take-off line. All lines can be 
 used, if need be, while one is being 
 worked upon. Also, the operator need 
 not fill all other take-off lines when using 
 water through any one of them. Large 
 standpipes thus equipped with valves 
 and with connected lateral pipes become control points, division points, or diver- 
 sion boxes, on the system, regulating the delivery. It may be desired to carry 
 water down two or three laterals from such a regulating stand. One of these 
 laterals may be short or sharply sloped, whereas the others are long and lack 
 a good grade. If the short or sharply sloped one had no gate, it might take 
 almost all the supply, the other two getting only a fraction of the flow. With a 
 gate before this free-flowing outlet, and on the other two as well, the operator 
 can restrict the flow to the ones having easiest get-away conditions and can 
 balance the flow to all, or arrange any desired distribution between the outlets 
 in use and still keep the stand from spilling over. Figure 51 also shows typical 
 gates for use at stands. A special adaptation of the open stand is the closed 
 stand or closed division box, which, as the name implies, is a roofed inter- 
 section point for several pipe lines. The control valve stems are run through 
 
 
 **mmm*» 
 
 1 *p 
 
 iS^** ^ : , J8 
 
 Wmwm lift 
 
 1 ' 
 
 ^# "ll 
 
 
 ^ sL 
 
 
 
 mBam 4111 
 
 
 Bl ^- 
 
 r x * 
 
 ' M ; 
 
 
 
 ^«ijfiS9B^ 
 
 
 
 
 
 \ 
 
 j$^' 
 
 
 ^:^ 
 
 
 
 
 
 Fig. 51.— Cut-away view of two small 
 gate valves installed in a section of con- 
 crete pipe to be used as a standpipe. 
 
Farm Irrigation Structures 
 
 53 
 
 stuffing boxes in the roof (fig. 52). Their chief application is at points where 
 division boxes are a necessity; but because of the hydraulic gradient con- 
 ditions, water would back up too high in an open stand to permit easy 
 operation of the valves (over 10 feet above the ground surface). Such a closed 
 stand might have a small (4 to 6 inches) sheet-iron stack at its roof center to 
 vent air and so might be partially effective in absorbing surges. Such stacks 
 can be guyed and may stand 5 to 10 feet or higher than normal water levels 
 at the stand. Figure 52 illustrates a closed stand with stack and control valve. 
 
 Another special stand (fig. 53) is the di- 
 vided type, in which a vertical wall across 
 the middle makes the structure a two- 
 compartment unit. The water supply en- 
 ters on one side of such a stand, and one 
 or more pipes take off from each compart- 
 ment. For greatest flexibility all take-off 
 pipes have valves ; and a valve is also put 
 near the bottom of the dividing wall over 
 a port joining the two sides of the wall. 
 Regardless of variable use by the take- 
 offs on the other side of the wall, those 
 on the supply side can be assured of a 
 constant height of water in the stand as 
 long as the dividing wall-gate valve is 
 closed. Such a steady height of water as- 
 sures steady flow in those supply-side 
 take-offs, and, if they supply ditches, 
 there is little danger of breaks in the 
 ditches under these conditions. Without 
 the wall, variable use by other lines would 
 create variable supply to the ditch lines 
 and probably frequent over-supply, with 
 broken ditch banks resulting. These two 
 special stands — the closed and the divided — are chiefly delivery structures. 
 
 Concrete pipe lines carrying water down steep grades, as in hilly orchard 
 localities, must have divided stands at frequent intervals (fig. 54), or a sub- 
 stitute for these stands available in the form of a float-operated valve which 
 gives complete control of the water. These stands or the float valve actually 
 break the pipe into short lengths so that each unit can have but 3, 6, or 8 feet 
 of water pressure on it. Water can be drawn off from the upstream disk-type 
 valve (fig. 54, detail 6). Water can be used at both top and bottom of such a 
 line at any time or at any intermediate point. If the divided overflow stands 
 were not located on such a steeply sloped line, the pressure at the bottom of 
 the line would be excessive. Water, when introduced, would plunge to the 
 bottom and must stop at the end. In all probability, the resultant shock would 
 
 Fig. 52. — Metal stack or riser on low 
 concrete stand. Stuffing box of valve 
 stem shown in front of stack. 
 
54 
 
 California Experiment Station Circular 362 
 
 Fig. 53. — Special stand or box with dividing walls. Note use of pipe sections 
 adjoining rectangular field-cast main box. 
 
 Overflow stand for steep grades 
 
 Concrete pad 
 
 Fig. 54. — Overflow stand (a) and pot hydrant (b) for hillside pipe-line system. 
 
Farm Irrigation Structures 
 
 55 
 
 burst the pipe. If it did not, the pressure resulting from the water column 
 would make it spurt out of any open valve located near the bottom of the 
 line. If water was being taken out at the bottom of continuous pipe, the 
 hydraulic gradient in the pipe might drop below the surface of the soil near 
 
 Soil surface 
 
 Source of supply 
 
 Poinl-ofuse W 
 
 Fig. 55. — Hydraulic gradient resulting from use of water at bottom 
 of steeply sloped pipe system. 
 
 
 
 
 
 ^^lltJf 
 
 
 
 ,™ 
 
 
 . jf, :am 
 
 fp.f:i;: »;:»;:;:?.:,; 
 
 ill 
 
 "lp||S|i|||iiii 
 
 alliiiff 
 
 m 
 
 S |. 
 
 ;-"l s 
 
 
 
 
 
 
 
 
 
 %§*§•;• jf* ■• ■■ 
 
 '"'\ 
 
 
 
 
 jppl 1 ^ 
 
 
 
 
 
 
 
 : ,-^J.%r-*nSh^m:^- " 
 
 
 
 
 
 
 
 
 Fig. 56. — Pot hydrants supplying furrows on flat land. 
 
 the upper end and water could not be used at that point under those condi- 
 tions (fig. 55). This recalls the requirement, previously noted, that the 
 hydraulic grade line must lie above the soil surface or water will not flow onto 
 the soil from a pipe. 
 
 Detail b of figure 54 (see also figure 56) illustrates a multiple outlet or 
 hydrant used on hillside irrigation of orchards. These structures are equally 
 
56 
 
 California Experiment Station Circular 362 
 
 useful on flatter land for irrigating row crops or furrows (fig. 56). They simu- 
 late the furrow tubes used in ditches and are, in fact, identical in their use. The 
 group of short furrow gates is congregated around the short section of pipe 
 called a pot, and set over the valve that covers the connection to the supply 
 
 Removable valve disc 
 
 Cemenh and 
 sand collar 
 
 Fig. 57. — Typical riser and "alfalfa" valve for supplying a ditch or for flooding. 
 
 pipe; the furrows are curved about the structure to catch the spill from the 
 individual tubes. Sometimes distributing arms made of 2-inch or larger light- 
 weight pipe are attached to the pots and distribution to the furrows is made 
 through holes in the arms. The flow to furrows can be controlled by manipu- 
 lation of the main valve (usually called an orchard valve) or by use of the 
 slide gates on the separate tubes or by both means. The main control valve 
 must be regulated so that the pot is not overflowed. 
 
 Where irrigation is done by flooding as in orchards and on alfalfa and other 
 
Farm Irrigation Structures 57 
 
 crops, the concrete retaining wall and furrow tubes are not needed and a simple 
 disk valve (usually called an alfalfa valve) flush with, or a few inches below, 
 the surface of the ground, is all that is necessary. It is placed on a short con- 
 necting pipe cemented to the supply pipe (fig. 57). Such outlets can also be 
 the supply source for a ditch or for several ditches (fig. 58). Whatever the 
 application, the flow from the mouth of the opened gate must not be permitted 
 to erode the soil. Often vegetation is grown about the gate to act as a buffer 
 against such erosion. 
 
 Fig. 58. — Special installation of disk valve on riser to feed one or more 
 ditches or to irrigate a pair of adjoining fields. The enclosure prevents erosion. 
 
 As with the ditch structures, there are many types of pipe structures. The 
 basic ones have already been discussed, with certain possible variations. Pre- 
 sumably, using the description supplied, the irrigator can make any further 
 variations necessary for his system. 
 
 The remaining structures are also delivery devices with particular applica- 
 tion to the release of water reasonably near the area of use. They comprise 
 the risers and accompanying valves or gates used for supplying ditches, fields, 
 or furrows. The risers (figs. 57 and 58), as their names imply, are the vertical 
 ducts that come to the soil surface from the buried pipe line. The tops of these 
 ducts, which are usually sections of concrete pipe cemented to a hole on top of 
 the supply line, must have valves in them so that water can be released only 
 at the point or points desired. Since the use of steel lines for irrigation distri- 
 bution is uncommon, no special equipment has been produced to go with such 
 lines for irrigation service only. In general steel is not so long-lived as concrete. 
 
 Sprinkling Systems. — Sprinklers and sprinkling systems comprise a distinct 
 development in irrigation, and a thorough discussion of them is available (12). 
 
 Inverted Siphons. — Inverted siphons are relatively short pipe lines, usually 
 installed to conduct the flow of a ditch under some surface obstruction such as 
 
58 California Experiment Station Circular 362 
 
 a road or railway embankment. One of these applications has been discussed 
 
 in connection with passageways across field ditches. A typical design problem 
 
 for such a siphon is solved as follows : suppose that the length of the concrete 
 
 pipe siphon is 54 feet from inlet to outlet; that the ditch handles 2 c.f.s. ; that 
 
 the water must not stand in the ditch at the entrance to the siphon above an 
 
 elevation of 267.0 feet; and that the water flows in the ditch at the downstream 
 
 side of the siphon at an elevation of 266.7 feet. The available difference in 
 
 head = 267.0 feet — 266.7 feet = 0.3 foot, which can be used as friction loss. 
 
 54 
 The length of siphon, 54 feet, = ttt = 0.54 foot. The available loss per 
 
 3 
 100 feet of pipe = — : — = 0.556 foot per 100 feet of pipe = H. Referring 
 
 to figure 44, from H = 0.556 on the left side of the chart, move to the right 
 to the vertical line from 2 c.f.s. on the bottom of the chart. Since these two 
 lines intersect between the curves for the 10- and 12-inch pipe sizes (although 
 nearer the 10-inch), the 12-inch pipe size would be installed. 
 
 This problem is solved as though an open standpipe were located at each 
 end of the siphon. The elevation of the water surfaces at the inlet and outlet of 
 the siphon will control the flow for any given size of pipe, just as do the eleva- 
 tions of water surface in the stands. Because a siphon is short, as a rule, one 
 can afford to use larger pipe than the design calculations indicate is necessary. 
 Such procedure is wise because of the losses in head at the entrance and exit of 
 the normal siphon. These losses are caused by sudden change in shape from 
 the ditch section to the siphon form; the water does not adjust itself to the 
 change without sacrificing energy in the form of head drop. It is to reduce 
 these losses that throats are provided at the entrance and exit of siphons, as 
 discussed earlier. Even with the throat, one may well use a somewhat larger 
 pipe than the calculations apparently require. 
 
 ACKNOWLEDGMENT 
 
 It is a pleasure to acknowledge the assistance received in preparing this 
 circular as a result of the work by J. E. Christiansen while he was Assistant 
 Irrigation Engineer in the College of Agriculture. Thanks are also due to 
 Edith H. Church of Dry den, N. Y., for permission to reproduce the charts 
 given in figures 4, 5, and 6. 
 
Farm Irrigation Structures 59 
 
 REFERENCES FOR FURTHER READING 
 
 1. King, H. W. 
 
 1939. Handbook of hydraulics for the solution of hydraulic problems. 528 p. 
 McGraw-Hill Book Co., New York, N.Y. 
 
 2. Merriman, Mansfield. 
 
 1916. Treatise on hydraulics. 565 p. John Wiley and Sons, New York, N. Y. 
 
 3. Dodge, Russell A., and Milton J. Thompson. 
 
 1937. Fluid mechanics. 495 p. McGraw-Hill Book Co., New York, N. Y. 
 
 4. United States Department of the Interior, Reclamation Service. 
 
 1917. Hydraulic and excavation tables. 168 p. U. S. Government Printing Office, 
 Washington, D. C. 
 
 5. Scobey, Fred C. 
 
 1915. The flow of water in irrigation channels. U. S. Dept. Agr. Bui. 194: 1-68. 
 
 6. Ganguillet, E., and W. R. Kutter. 
 
 1889. General formula for the uniform flow of water in rivers and other channels. 
 (English translation by R. Herring and J. C. Trautwine.) John Wiley and 
 Sons, New York, N. Y. 
 
 7. Scobey, Fred C. 
 
 1933. The flow of water in flumes. U. S. Dept. Agr. Tech. Bui. 393: 1-99. 
 
 8. Christiansen, J. E. 
 
 1935. Measuring water for irrigation. California Agr. Exp. Sta. Bui. 588: 1-96. 
 
 9. Stanley, F. W., and Samuel Fortier. 
 
 1921. The use of concrete pipe in irrigation. U. S. Dept. Agr. Bui. 906: 1-54. 
 
 10. Williams, Gardner S., and Allen Hazen. 
 
 1920. Hydraulic tables. 115 p. John Wiley and Sons, New York, N. Y. 
 
 11. Scobey, Fred C. 
 
 1920. The flow of water in concrete pipe. U. S. Dept. Agr. Bui. 852: 1-100. 
 
 12. Christiansen, J. E. 
 
 1942. Irrigation by sprinkling. California Agr. Exp. Sta. Bui. 670: 1-124. 
 
 13. Church, Irving P. 
 
 1902. Diagrams of mean velocity of uniform motion of water in open channels; 
 based on the formula of Ganguillet and Kutter. John Wiley and Sons, New 
 York, N. Y. 
 
 14. Pillsbury, Arthur F. 
 
 1941. Corrosive effect of inorganic fertilizers on concrete. Civ. Engin. 11(6): 348-49. 
 
 20m— 7, '45 (3768;