LIQUID METERING SYSTEMS FOR LOW VOLUME SPRAYERS Thesis for the Degree of M. S. MICHIGAN STATE UNIVERSITY STEPHEN DAVID JOHNSON 1972 ‘- m ‘A L R Y Pi'fiChIgan 3mm Riversity a! BINDING av 3" I'IIIAII & SIIIIS' BIIIIK BIIIIIERY INC. L IBRARY BIN 0535 _ml moi..- ‘1: 5,4 I ABSTRACT LIQUID METERING SYSTEMS FOR LOW VOLUME SPRAYERS by Stephen David Johnson Pest control through the application of chemicals has been an integral part of crop production for nearly a century. The most common method of pesticide application for fruits and vegetables is dilute airblast spraying. Recently, however, this method in its present form has drawn fire from ecologists and a concerned public because of its possible detrimental effects to the environment. Alternate methods of pest control have been proposed and also alternate methods of chemical application have been tried. One such alternate method of application is ultra-low volume application. The method offers the advantages of reduced dosage and a reduced number of treatments needed in a growing season. One of the problems associated with this method of pesticide application is the lack of an accurate low-flow metering system. The purpose of this study was to attempt to solve this problem. Several systems were considered and two systems were investigated in detail. A peristaltic pump system which included equipment for desurging the pulsating output was evaluated under laboratory and field conditions. The results of the tests indicated that this system could reliably accomplish the stated research objective. A gear pump system was also evaluated in the laboratory. It demonstrated equal and some- times superior performance in all aspects of testing to that of the peristaltic pump system. Stephen David Johnson Approved :flbmej 3%1’/€Lédé£j§E>—“ Ma ofIProfessor 16 A m Department Chairman LIQUID METERING SYSTEMS FOR LOW VOLUME SPRAYERS By STEPHEN DAVID JOHNSON A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Agricultural Engineering 1972 ACKNOWLEDGMENTS The guidance, time, and interest of the author's major professor, Dr. L. K. Pickett (Agricultural Engineering) is gratefully acknowledged. The time and helpful suggestions of Dr. M. C. Potter (Mechanical Engineering), Dr. D. C. Wiggert (Civil Engineering), and Dr. G. E. Merva (Agricultural Engineering) are greatly appreciated. Dr. B. A. Stout and the Department of Agricultural Engineering are acknowledged for the financial assistance and facilities made available to the author. A special thanks is extended to Dr. A. J. Howitt (Entomology) for his help in defining the research problem and his financial support of the work. ii DISCLAIMER References to company names, products, or trademarks do not constitute condemnation or endorsement of these products, but merely convenient designations for selected equipment and materials. iii TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . . vi LIST OF FIGURES . . . . . . . . . . . . . vii NOMENCLATURE . . . . . . . . . . . . . . x Chapter I. INTRODUCTION . . . . . . . . . . . 1 II. BACKGROUND INFORMATION . . . . . . . . 3 III. OBJECTIVES . . . . . . . . . . . . 12 IV. LITERATURE REVIEW . . . . . . . . . 13 4.1 METERING PUMPS . . . . . . . . . 13 4.2 ROTARY PUMPS . . . . . . . . . 20 4.3 PERISTALTIC PUMPS . . . . . . . . 27 V. CHARACTERISTICS OF SPRAY MATERIALS . . . . 32 5.1 SPECIFIC GRAVITY MEASUREMENT 32 5.2 VISCOSITY MEASUREMENT . . . . . . . 33 5.3 SELECTION OF TEST MATERIALS . . . . . 38 VI. PERISTALTIC PUMP SYSTEM . . . . . . . . 39 6.1 INTRODUCTION . . . . . . . . . . 39 6.2 EFFECT OF PUMP SPEED . . . . . . . 42 6.3 EFFECT OF PRESSURE HEAD . . . . . . 46 6.4 DESURGING THE OUTPUT . . . . . . . 49 6.5 FIELD TESTS . . . . . . . . . . 54 VII. GEAR PUMP SYSTEM . . . . . . . . . . 58 7.1 INTRODUCTION . . . . . . . . . 58 7.2 EFFECT OF PUMP SPEED . . . . . . . 58 7.3 EFFECT OF PRESSURE HEAD . . . . . . 60 VIII. OTHER PUMP SYSTEMS . . . . . . . . . 64 iv Chapter Page IX. DISCUSSION AND CONCLUSIONS . . . . . . 66 REFERENCES . . . . . . . . . . . . . . . 68 APPENDIX . . . . . . . . . . . . . . . 71 A CLASSIFICATIONS OF PUMPS . . . . . . 71 B METERING PUMP MANUFACTURERS . . . . . 72 C CLASSES OF ROTARY PUMPS . . . . . . 74 D TEMPERATURE INFLUENCE ON VISCOSITY . . 75 E OUTPUT CHARACTERISTICS--PERISTALTIC PUMP . 78 F OUTPUT CHARACTERISTICS--GEAR PUMP . . . 80 LIST OF TABLES Table Page 1 Concentration and Acres Covered Per Hour . . . . 6 2 Characteristics of Common Spray Materials . . . 35 3 'Results of Field Testing . . . . . . . . 57 vi Figure 2.1.2 .1 LIST OF FIGURES Concentration versus Acres Sprayed Per Hour Milton Roy Reciprocating Piston with Double Inlet and and Discharge Valves Jaco Mechanically-Accuated Diaphragm Pump Liquid End of a Diaphragm Pump--Mechanical Support Liquid End of a Diaphragm Pump-~Hydraulic Support Peristaltic Pump with Cam-Operated Fingers Revolving Roller Peristaltic Pump External Spur Gear Rotary Pump Three-Lobe Rotary Pump . . . . Internal Gear Rotary Pump with Crescent-Shaped Partition . . Liquid Trapped Between Gear Teeth Three Spur Gear Rotary Pump Specific Gravity Test Apparatus Viscosity Measurement Apparatus Viscometer Calbiration Curve Sigmamotor Pump--Showing Backing Plate and Cam-Operated Fingers . . . . . . . . . . . . . . Output versus Pump Speed Test Setup for the Peristaltic Pump Average Tubing Diameter Perpendicular to Backing Plate Change in Shape of Tubing Cross-section as Tubing is PinChed 0 o o o o o o o o o o o o 0 Output versus Pump Speed Using Cythion--Peristaltic Pump. gutput versus Pump Speed Using Sevinmole 4--Perista1tic ump . . . . . . . . . . . . . . . . Vii Page 16 16 l7 l7 19 19 21 25 25 28 28 34 34 37 41 41 43 43 44 LIST OF FIGURES (CONT.) Figure Page 6.3.1 Output versus Pressure Head Using Cythion--Peristaltic Pump . . . . . . . . . . . . . . . . . 47 6.3.2 Output versus Pressure Head Using Sevinmole 4--Perista1tic Pump . . . . . . . . . . . . . . . . . 48 6.4.1 Pressure Dissipation Test Setup . . I. . . . . . 51 6.4.2 Pressure Dissipation Test Setup Using an Air Chamber . 51 6.4.3 Modified Fuel Line Check Valve . . . . . . . . 53 6.4.4 Ball Check Valve . . . . . . . . . . . . . 53 6.4.5 Complete Peristaltic Pump System . . . . . . . . 55 6.5.1 Peristaltic Pump System Mounted on Sprayall Sprayer . 55 7.1.1 Zenith External Spur Gear Pump . . . . . . . . 59 7.2.1 Output versus Pump Speed Test Setup--Gear Pump . . . 59 7.2.2 Output versus Pump Speed Using Cythion--Gear Pump . . 61 7.2.3 Output versus Pump Speed Using Sevinmole 4--Gear Pump . 61 7.3.1 Output versus Pressure Head Using Cythion--Gear Pump . 62 7.3.2 Output versus Pressure Head Using Sevinmole 4--Gear Pump . 63 8.1.1 Pulsationless Flow Drive . . . . . . . . . . 65 D.l Viscosity versus Temperature At Constant Shear Rate-- Cythion . . . . . . . . . . . . . .7 . . 75 D.2 Viscosity versus Temperature At Constant Shear Rate-- Sevinmole 4 . . . . . . . . . . . . . . 75 D.3 Viscosity versus Temperature At Constant Shear Rate-- Omite 6E . . . . . . . . . . . . . . . 76 D.4 Viscosity versus Temperature At Constant Shear Rate-- Sevin 4LS . . . . . . . . . . . . . . . 76 D.S Viscosity versus Temperature At Constant Shear Rate-- Dioctyl Pthalate (DOP) . . . . . . . . . . . 77 viii Figure D.6 LIST OF FIGURES (CONT.) Viscosity versus Temperature At Constant Shear Rate-- Perthane 4EC Output Output Output Output Output Output Output Output versus versus versus versus versus versus versus versus Pump Pump Pump Pump Pump Pump Pump Pump Speed-~Perthane 4EC Speed--Dioctyl Pthalate (DOP) Speed-~Omite 6E Speed--Sevin 4LS . . . . Speed--Perthane 4EC Speed--Dioctyl Pthalate (DOP) Speed--Omite 6E Speed--Sevin 4LS . . . . . ix Page 77 78 78 79 79 80 80 81 81 cpse fps gph gpm 1bf 1bf/in ml/hr ml/min ml/rev mph psi psig rpm spm °F NOMENCLATURE Centipoise (Viscosity) Feet per second (Speed) Gallons per hour (Output) Gallons per minute (Output) Pounds (Force) Pounds per inch (Spring Constant) Milliliters per hour (Output) Milliliters per minute (Output) Milliliters per revolution (Displacement) Miles per hour (Speed) Pounds per square inch (Pressure) Pounds per square inch gauge (Pressure) Revolutions per minute (Speed) Strokes per minute (Pumping Speed) Degrees Fahrenheit (Temperature) Degrees (Angular Rotation) I. INTRODUCTION Pesticide application and the possible resultant environmental pollution are subjects of discussion in ladies' tea circles, college laboratories, and even in the chambers of state and national govern- ment. At this time there are bills in both the House of Representatives (Obey, 1971) and in the Senate (Nelson, 1971) which if passed would exact stiff regulatory measures on the use and distribution of chemicals including pesticides. Similar legislation has already become law in some provinces of Canada. This national consciousness is due to the crusading of such individuals as Ralph Nader (Nelson, 1969) and Rachel Carson (Westcott, 1966) who,unfortunately,like to dwell only on the negative aspects of pesticide application. However, this negative view that is being instilled in the general public does have the positive effect of forcing much-needed research in the field to be initiated. Many alternatives to pesticide application in its present form have been offered. One of the more promising is a technique called "integrated control". This method calls for the introduction of organisms that are natural predators of the economically harmful insects with a greatly reduced amount of pesticides applied to keep the predator populations in check. Other methods of control that are being considered are microbial disease treatments of insects, insect sterilization and metamorphosis inhibition, sex attractant sprays to draw the insects to one place for more efficient disposal, and developing genetic resistance to insect damage in plants (westcott, 1966). Nevertheless the application of chemical pesticides is still the most widely used means of pest control. According to 1958 statistics, over 92 million acres were sprayed for pest control (USDA 1962). Yet many of the ideas and equipment principles used on today's application equipment are the product of research that was done 25,50 or even 75 years ago. Current research in this area has been directed toward the development of application equipment which reduces substantially the amount of chemicals introduced into the ecosystem while at the same time providing adequate crop protection. This new approach requires a redesign of all the major components of the chemical delivery system including fan and nozzle characteristics and chemical metering of the low flow rates required. II. BACKGROUND INFORMATION Prior to 1868, there was little or no knowledge regarding chemicals for control of plant pests and, therefore, no need for application equipment (Potts, 1958). The Colorado potato beetle was the first pest to be controlled through chemical application. The substance used was called "Paris Green" and was applied using a box with a wire-screen bottom. The box was held over the potato plants and shaken so that some of the dust fell on the leaves. In 1882, copper sulfate and lime, Bordeaux, was applied to grapes in France to prevent grapevine mildew. The first applications were made with a crude whisk broom which was dipped in the Bordeaux solution and shaken at the vines. As soon as the means for controlling these diseases became available, machines for their application began to appear. A more sophisticated revolving brush arrange- ment was invented to apply the Bordeaux solution. Also a number of Sprayers with tanks that were carried on a person's back appeared. In 1883, John Bean invented the first force pump sprayer at Los Gatos, California (Fronk, 1962). From then until about 1900, many models of both the knapsack and force pump sprayer were introduced. In 1911, the pressure regulator was invented and in l9l4,the adjustable spray gun was introduced to the market. From that time until the late 1930's, little progress was made except to increase the capacity of existing machines. About 1937, a citrus grower in Florida toyed with the idea of using an airblast as the carrying force for the spray solution instead of hydraulic pressure. As a result, the method of dilute airblast Spraying was introduced and rapidly became the dominant form of pesticide application for orchard and grove crops. _ 4 - One of the reasons for the rapid acceptance of airblast spraying was that it enabled the operator to spray a large acreage of trees with a limited amount of labor. Since it operates with less pressure than hydraulic Sprayers, there is less wear and tear on valves, nozzles and other parts of the pumping system. When compared with other systems commonly used in orchards and groves, speed Sprayers present the following additional advantages. They provide a very uniform coverage of the plant being sprayed. The formulations used have a very low level of toxicity to humans and are dilute enough so that slight errors in calibration are not critical. Most speed Sprayers provide an adequate supply of air to carry the spray to the upper portions of the tall fruit trees (Brann, 1956). , Dilute airblast spraying does have some disadvantages, however. A tremendous volume of water is required for a single application and a lot of chemical is wasted through runoff which frequently ends up in under- ground water supplies. Airblast spraying equipment is rather large, bulky, and heavy, causing soil compaction and traction problems during wet weather (Courshee, 1960). Several of the problems associated with dilute airblast spraying, namely the amount of water used and the material wasted, prompted the development of concentrate Spraying. In this method, the same amount of chemical is placed in one-half or one quarter or possibly less water while reducing the amount of water sprayed on each tree by the same proportion. This method applies the same amount or approximately the same amount of chemical per acre as the airblast machines while using far less water per acre and, consequently,less total spray time is spent filling the tank. The concentrate sprayer in many instances, is an airblast sprayer that has been calibrated to put out only a fraction of its original volume per acre. - 5 - However, thenehave been some sprayers that are designed primarily for concentrate applications. These machines are usually equipped with a squirrel-cage fan capable of producing between 130 and 140 mph winds with volumes of at least 7000 cfm (Canada Department of Agriculture, 1963). When compared with dilute airblast spraying, concentrate spraying offers several advantages (Brann, 1968). As mentioned earlier, there is a great reduction in the amount of water needed for applications and, as a result of the increased concentration of material, there is an increase in the amount of active chemical ingredient deposited per unit volume applied. With machines made primarily for concentrate spraying there is a reduction in weight of the equipment and therefore not as many problems with traction or soil compaction. There is also an increase in the number of acres that can be Sprayed in a given period of time. Table 1 shows the increase in acres per hour with increasing concentration based on 500 gallons per acre as a dilute application. The savings in time reaches a point above which, however, it is negligible for any further increase in concentra- tion as shown in Figure 2.1.1. Concentrate Spraying also has some disadvantages. One of the most crucial aspects is that slight errors in calibration are greatly magnified. The concentrated chemical solution creates more frothing in the tank while being agitated. Also the nonuniform coverage obtained with this method is not always effective against certain pests, for example mites, certain fungus diseases, etc. (Brann, 1968). The smaller droplets produced by concentrate sprayers lead to problems of drift, evaporation of carrier liquid and failure of some of the smaller droplets to impinge on the plant surface. As a consequence, concentrate applications are limited womH .aamum "mousom m.¢ 0H «mm NH me N.¢ Hm NmH NH on ¢.¢ Ne om NH xm o.¢ mNH we NH Ne m.m omN «N NH xN m.N oom NH NH xH ago: pom opom\Hmo Hump oso How Ho>muu paw :oHumuuaooaoo mouo< oaHu waHmmumm HHHH ou oaHH usom Mom poum>oo mmho< paw GOHumuuooocooun.H MHm :oHumpucmoaou .H.H.N mustm mooH .ccmum “mousom onH Ci) Source: Marton 1963 Figure 4.1.5. Peristaltic Pump With Cam-Operated Fingers. I III” I. H NMWHW IIIIIIIIIIII. Source: The Randolph Company, 1970. Figure 4.1.6. Revolving Roller Peristaltic Pump. - 20 - very small (down to 1 drop per hour) and the fluids do not contact the pump, thus preventing contamination of the pump or the fluid. These pumps are self-priming and require no valves or seals (Marton, 1963). Rotary metering pumps furnish nearly a pulse-free flow under pressures up to 1500 psi and can handle high viscosity fluids. They can have a very high volumetric efficiency. The pumping action is accomplished in a variety of ways including eccentric rotors with sliding vanes, rolling vanes and other impellors, and meshing spur gears. The external gear pump shown in Figure 4.1.7 is one type of rotary pump frequently used for metering. Both peristaltic and rotary pumps Show promise for metering systems for ULV sprayers. The peristaltic pump warrants further inspection because of its high accuracy at low flow rates (Marton,l963) and isolation of the liquid being pumped. Rotary gear pumps present the desired characteristics of high accuracy and pulsationless flow. 4.2. ROTARY PUMPS The general working principle for rotary pumps is that as the rotor turns, it creates cavities which move from suction to discharge forcing the liquid along. A seal between suction and discharge is formed by close running clearances or rolling or Sliding contact (AIChE, 1960). The output of rotary pumps is proportional to speed except for losses due to inlet conditions and normal internal leakage from discharge to suction, called slip, which varies inversely with the viscosity of the fluid and directly with pressure differential. The inverse relationship of slip and viscosity usually does not hold for high speed and high -21- Source: White, 1971. External Spur Gear Rotary Figure 4.1.7. Pump . -22.. viscosity. Under these conditions, there is a noticeable temperature effect due to the high rate of viscous shear in the pump which heats the liquid in the clearance spaces and thereby reduces its viscosity. Slip is a minimum in an external spur gear pump for a peripheral gear Speed of 34 fps and 100 psi at 1500 sec Saybolt viscosity. Slip increases from this point with an increase in viscosity (Pigott, 1944). Inlet conditions also can affect the output of a rotary pump. It is common practice to provide inlet conditions such that the vacuum at the pump inlet does not exceed 5 inches of mercury at normal operating Speeds. Causes of high inlet vacuum include a too long or small inlet line, bends and fittings in the line, large pressure dr0p through a strainer, or lift required from a reservoir which is below the pump. The results of bad inlet conditions include cavitation which decreases the flow rate and a shortening of pump life. As the suction increases, the pump may become noisy and begin to vibrate (Horn, 1968). Theoretically, flow is independent of the pressure differential for rotary positive displacement pumps. In reality, there is a slight increase in slip with increased pressure resulting in decreased flow. Power requirements vary with pressure and speed, or, if both are constant, with the viscosity of the fluid. Rotary pumps are used for pumping a wide variety of materials including chemicals,oils, gasoline, solvents, tar and varnish. They can handle pressures up to 1000 psi with non-lubricating fluids and even higher with lubricants. Rotary pumps can also produce low flows over a wide range of pressures. They have a relatively low initial cost and require little Space. However, the close clearances associated with - 23 - rotary pumps limit somewhat the choice of materials of construction and also the ability to pump liquids containing suspended particles. Also, being positive displacement pumps they must be protected from excessive pressure with a relieving device. A rotary pump is a positive displacement pump which consists of a set of gears, cams, screws, vanes, or other elements in a casing which are actuated by a drive shaft. Rotary pumps are characterized by their lack of inlet or outlet valves, and their close running clearances. Two general classes of rotary pumps are, single rotor and multiple rotor (Hydraulic Institute, 1965). Relationships between types and classes of rotary pumps are shown in Appendix C. There are four basic single rotor pumps (Hydraulic Institute, 1965). They are (l) vane pumps which have a vane or vanes that take the shape of blades, buckets, rollers, or slippers and operate with a cam to draw fluid into and force it from the pump cavity; (2) rotary piston pumps which have reciprocating pistons mounted in a rotor; (3) flexible rotor pumps containing a flexible tube, vane, or liner; and (4) screw pumps. Screw pumps, which can be either single or multi-rotor, displace the fluid axially through the rotor screw threads. Two other types of multiple rotor pumps,lobe (Figure 4.2.1) and circumferential piston, carry the fluid in the spaces between the lobes or pistons from inlet to discharge. Both pumps lack torque transfer contacts between rotor surfaces and, therefore, must be externally timed. The gear pump is one of the most common multiple rotor rotary pumps. Internal gear pumps have one rotor with internally-cut gear teeth that meshes with a rotor having externally-cut teeth. Some - 24 _ internal gear pumps are made with a crescent-shaped partition (Figure 4.2.2) separating inlet and discharge. Other internal gear pumps have gears with teeth of a size, shape, and number so that a separator is not needed. External gear pumps (Figure 4.1.7) have externally-cut gear teeth on all rotors and these teeth may be spur, helical or herringbone in shape. External spur-gear pumps are intended for slow-speed operation up to about 600 rpm for handling most thick viscous liquids which possess lubricating properties. Most spur-gear pumps have a driver and idler gear type arrangement with the driver gear keyed to an external drive shaft. However, sometimes, the rotors are driven by timing gears for several reasons. Rotor drive contact may produce too much wear or a clearance may be desirable to avoid the necessity of lubricants in the pump chamber. Over part of the cycle, contact may be such that torque can only be transmitted in the sense opposite to that required (Hadekel, 1951). The volume delivered by an external gear pump depends on tooth depth, so that the greatest output for a given pump size and speed is produced with the minimum number of teeth (Horn, 1968). Displacement per revolution in a gear pump where the gears mesh perfectly is equal to one-half the volume enclosed between the addendum and dedendum circles of the gear or, for a two-gear pump: 2 2 _ D - (ra - rd )b (4.1) where: D = Displacement per revolution ra = Radius of addendum circle rd = Radius of dedendum circle b = Length of the gear arge Source: AIChE, 1960. Figure 4.2.1. Three-Lobe Rotary Pump. - s Source: Stuart and Heldeman, 1970. Figure 4.2.2. Internal-Gear Rotary Pump With Crescent-Shaped Partition. _ 26 - Volume delivered is equal to displacement per revolution multiplied by the speed of the pump (Wilson, 1950). This, of course, is a simplified and idealized treatment of displacement calculation. An experiment to determine actual displacement per revolution is discussed by Pigott (1944) and a rigorous mathematical calculation of delivery per radian of rotation considering such parameters as number of teeth per gear, gear-tooth pressure angles, etc. is performed by Hadekel (1951). One problem associated with spur—gear pumps, is liquid which is trapped between meshing gears (Figure 4.2.3) and the resulting pressures which produce unbalanced bearing loads. There are several methods used to relieve this pressure. One method is to drill radial holes in the tips and roots of the gear teeth allowing an escape route for the trapped fluid (Kristal, 1953). Small grooves can be cut in the side plates allowing the trapped liquid to pass from tooth to tooth (Pumping Manual, 1962). Another method used for low pressure applications is to design a small amount of backlash into the gears or choose the dimensions of the gears in such a way that the trapped Space is increasing during the entire overlap period (Hadekel, 1951). Another unbalanced pressure bearing load is caused by the basic design of the spur-gear pump. The pressure on the inlet side of the gears is essentially atmospheric while the pressure on the<>utlet Side is whatever the discharge pressure happens to be thus placing an un- balanced load on the bearings. This load is called a differential pressure load and is equal to the outlet pressurezmfltiplied by the projected area of the gear blank (Stuart & Holdeman, 1970). This - 27 - pressure differential affects notonly the bearing loads, but also, as mentioned before, the amount of leakage or slip of the pump. Some pumps have been designed to eliminate this unbalanced load but are seldom used. The three spur-gear pump (Figure 4.2.4) eliminates the bearing load unbalance on the driver gear, and has the added advantage of supplying practically double the outflow for a relatively small increase in Size. 4,3. PERISTALTIC PUMPS Peristaltic motion is defined as "the rhythmic, wavelike motion of the walls of the alimentary canal and certain other hollow organs consist- ing of alternate contractions and dilations of transverse and longitudinal muscles that move the contents of the tube onward" (Guralnik, 1970). Pumps that operate with this type of motion, with no glands or valves, are being used in increasing numbers in medical applications, research and industry. Peristaltic pumps are capable of handling fluids which cannot tolerate atmospheric contamination or, conversely,must not be allowed to corrode pump parts or contaminate the atmosphere. They can handle many types of fluids including slurries, near-solids, liquids, and gases, and are self-priming (Neal, 1965). Since the fluid remains within flexible tubing, there are no cleaning problems and sanitation can be insured. Peristaltic pumps are essentially low-pressure pumps and are not used for continuous heavy-duty pumping. The flexible tubing is subject to repeated stresses and its service life is dependent upon the speed of the pump, the pressure differential, and the nature of the fluid being pumped. _ 28 - Source: White, 1971. Figure 4.2.3. Liquid Trapped Between Gear Teeth. Source: Stuart and Heldeman, 1970. Figure 4.2.4. Three Spur Gear Rotary Pump. _ 29 - The "muscles" that contract the tubing take several shapes, the most common of which is a series of rollers mounted on the periphery of a rotor. The tubing is mounted in a curved cradle so that as the rollers pass over it, the tubing is progressively squeezed, pushing the fluid forward in front of the advancing contraction toward the outlet and simultaneously creating a vacuum in the tubing behind the roller drawing in fluid from the inlet (Figure 4.1.6). Although most peristaltic pumps use this revolving roller scheme, some use a cam-operated sequence of metal fingers which progressively squeeze the tubing (Figure 4.1.5). Another method involves a rotor on an eccentric shaft which squeezes a flexible cylinder liner (Marton, 1963). A new design has been developed for biological and medical research where extremely low flow rates are needed. It has very small elastic tube wound around a delta- shaped rotor which produces kinks in the tube. The kinks travel around as the rotor rotates to force the contents along the tube (Neal, 1965). The design of the cradle and tube-holding devices is an important factor affecting the stresses, and life, of the tubing. Most peristaltic pumps are designed to handle almost any commercially available tubing that is appropriate in size and material properties for a given application. Some manufacturers, however, do provide preshaped tubing for special applications. Some pumps have a fixed-Speed drive and their output is controlled by tubing size or degree of tubing occlusion (Howitt,l970). In others output is varied by changing pump speed through mechanical variable gearing or through an electronic Speed unit. In the most sophisticated pumps, the pump speed, thus the flow rate, can be controlled by a pneumatic or electrical control signal generated by -3o-‘ some process variable (Marton, 1963). Because of the similarity of peristaltic pumping action to body functions, there has been much research (Latham, 1966; Jaffrin and Shapiro, 1971) with the specific objective of modeling the human ureter. The purpose of this modeling was to attempt to discover the mechanism of fluid mixing or motion in the tube which might explain disease transmission from the bladder in a reverse direction to the kidneys. Research has shown that under many conditions of operation, the net time-mean flow in peristalsis is the algebraic difference between a forward time-mean flow in the core of the tube and a backward (reflux) time-mean flow near the periphery (Shapiro et al, 1969). This would seem to be the explanation for the reverse disease transmission, but it has been shown that reflux conditions are not present in a-healthy ureter (Weinberg et a1, 1971). The amount of backward motion of reflux of the fluid can be quite significant. It is affected by the degree of pinch or occlusion of the tube and also by the pressure differential from the inlet to the outlet (Latham,l966). Wfldia constant degree of tube occlusion and constant pump speed, the amount of reflux increases as pressure head increases. If pressure and Speed are held constant and the degree of occlusion is increased, reflux decreases and the output of the pump increases. In peristaltic metering pumps, the tube is completely or almost com- pletely occluded. The pump is essentially positive displacement in nature and outflow is independent of pressure head within certain finite limits (Jaffrin and Shapiro, 1971). The fields of application for peristaltic pumps are continually - 31 - growing. In industry, they are used for handling corrosive and radio- active fluids, cement (both wet and dry), printing ink, and foodstuffs. In the chemical processing industry, they are used for continuous sampling and analysis, for effluent control, and boiler feedwater dosing. In the medical field, peristaltic pumps are used on artificial kidney machines, and for controlled dosing in both surgical application and in research (Neal, 1965). Peristaltic pumps are increasingly taking over many of the duties of other types of pumps because of their simplicity, cleanliness, and versatility. V- CHARACTERISTICS OF SPRAY MATERIALS Before experimenting with a metering system, the characteristics of the materials to be handled should be known. Important character- istics influencing the development of a metering system are density and viscosity of the materials. Also of interest are the Specific rheological characteristics if the metered fluids have non-Newtonian behavior because the performance of the metering system may be affected by resulting changes in apparent viscosity. 5.1. SPECIFIC GRAVITY MEASUREMENT Specific gravity was measured using a beaker of material, a thermometer, and a precision balance (Figure 5.1.1). The procedure was to (1) record the material temperature, (2) weigh the beaker of material and subtract the weight of an empty beaker,and (3) calculate the weight of an equal volume of water at the same temperature. Specific gravity was then obtained using the following formula: W Specific Gravity = Wfig— (5.1) w where: Wm = Weight of volume of material at a fixed temperature WW = Weight of equal volume of water at the same temperature. The results of this test and manufacturers' specifications are given in Table 2. The results show that most of the chemicals tested were Similar in density to water. Thus, density is not expected to seriously affect the flow rate from the metering system. - 32 - - 33 - 5.2. VISCOSITY MEASUREMENT Fluid viscosity influences pump output for a given speed and also power requirements for a given output. A Brookfield Synchro-Lectric viscometer (Figure 5.2.1) was used for the measurement of the viscosities of selected chemicals. This viscometer rotates a spindle in the fluid and measures the torque necessary to overcome the viscous resistance to induced movement. Four Spindle speeds and four Sizes of cylindrical spindles are provided for measuring a wide range of viscosities. Non-Newtonian behavior related to the rate of shear can be Studied simply by changing speeds with a given Spindle size. Cross- like spindles and a special Helipath stand are included with the equip- ment for measuringpapparent viscosity of non-Newtonian fluids for which viscosity at a fixed rate of Shear changes with time. Viscosity measurements were made with a 3-inch diameter beaker of fluid which is greater than the 2.25-inch minimum size required for the viscometer. The procedure followed for each measurement was to use the viscometer factor charts and an estimate of the viscosity to select the spindle to be used for that test. The dial reading was recorded for each of the four spindle speeds. The viscometer was run for several minutes for each test to determine if viscosity changed with time. If a change withtime was observed, the special cross- like spindles and Helipath stand were used for measuring the apparent viscosity. Viscosities were calculated by multiplying the dial reading by the viscometer factor for the Spindle. All substances were evaluated at three different temperatures to observe the effect of temperature on viscosity. The three temperatures chosen were 49°F, 72°F, and 85°F. The samples were brought to 49°F by Figure 5.1.1. Specific Gravity Test Apparatus. Figure 5.2.1. Viscosity Measurement Apparatus. . 5 3 .mo>Humuaomouaou.moHamaeoo HmoHamno :uHS oocoeooamounoo Hmcomnoa swoops» vosHauno aoHumauomcH * .oHanHm>m uo: sowumauomcH s oumuucooaou mHnm :mHaouSSZ « up we moNn um Ne wo.Hreo.H Ho.H uHmHmHsam ovHoHumo< mo ouHao oHummHaovosmm sum oaaouuoxaae ecom :chouSSZIGoz meow up one um mow mo.H mm. oHnmson ovHoHuoomeH mHe SH>om oHummHaoosomm I was oHaouuoxHSH meow um meow e oHos cmHooOSSZIaoz OmNIomH um ooHroem N.H MH.H oHnmson ovHoHuoomcH oH>om 38V Saaaaa mumHmeum cmHaousez s moHN um me Home. Na. oHaoHH ouam uoHuumo IHMUSOHQ mumuuaoocou one :SHSOusoz s moNN um m“« NNo.H Nnm. mHanmHmHsam ooHoHuoomaH oomsuuom cchousoz menu um mn.om moms um NN mHmN.H mNH.H eHsvHH spam ovHoHuuomsH aoHSuao e Homaov Ammaov a» 59330 5.16.5 muHmoomH> NuHmoomH> onHooam onHomam mucoaeoo muonsuommsomz popsmmoz muousuommsamz seasons: sowuHmoaaou aoHuaonHmmmHo HoHuoumz mHmHuoumz mmuem coaeoo mo muHumHuouomSmsouu.N MHmh:u coHumunHHmo HouoeoomH> .N.N.m mustm Ammaov waHmoomH> ommsmemz OS on on om OH H p p N a db ‘I’ q- - q - irmm L.moH .ummH Ammaov HonHSHOm mmomoam Noe zo ammemv HSHmoomH> Hom L d wchD oomam masm msmpo> usauzo Hague ommmm exam 00H 00 I q GI- .N.~.S seamen oeH HeHS\Hsv HDmHDO mZDm qu I‘ 1b 00H 1 d. wch: spasm asam wsmno> usauso Agape ommmm exam oo .QESm oHUHmumHummrrcoHsuzo .H.~.e magmas qr- How .oe Acas\Hev Haaeae ELSA tOOH . qu - 45 - There was quite a substantial difference between the theoretical displacement per revolution (1.964 ml/rev) and the actual displacement per revolution (1 ml/rev to 0.43 ml/rev). The difference can be partially explained by the adverse flow properties of some of the materials eSpecially the friction drag forces caused by the particle content of flowables which restricted inlet capacity at higher pump Speeds. Another factor based on visual observation.was a permanent change in hose shape from round to slightly elliptical after several tests. This permanent deformation would affect hose capacity and, consequently.displacement per revolution. Another possible cause was imprOper pump sealing, i.e. the forward-most pinching finger allowing line pressure into the pumping segment of the hose before the inlet side is completely sealed off, thus causing a small amount of reflux. An assumptiOn made for the theoretical displacement calculation was that the tube sustained pinching force from the fingers only in a direction perpendicular to the tube wall. Any strain or movement in the axial direction was disregarded. In reality, the fingers may also have stretched the tubing longitudinally further reducing the effective diSplacement volume. Furthermore,the deformed shape of the tubing was taken as elliptical for ease of calculation when the actual shape has two parallel sides and two rounded ends (Figure 6.1.4). This shape has the same perimeter as an ellipse but not as much area, thus a further reduction in displacement volume. It was observed that a sudden raising or lowering of the output line made only instantaneous changes in output from the tube and the system quickly stabilized. Observation of the output and flowmeter float showed - 46 - a continual pulsation of the flow at all pumping speeds. At low Speeds this pulsation produced a flow-no flow condition in the output which is not acceptable for a sprayer metering system. This experiment showed that for most Spray chemicals except the flowable formulations,a direct proportionality existed between pump speed and output. The relationship was good enough to preclude the use of calibration charts for obtaining an output rate within 10% of that desired at any speed. Flowable formulationsrequired exact calibration curves because of their nonlinear output versus pump Speed characteristics. Also indicated by this experiment was the need for some form of desurger to correct the flow-no flow conditions created by the pulsating output. 6.3. EFFECT OF PRESSURE HEAD Adjusting the backing plates (Figure 6.1.1) of the pump so that the fingers completely occlude the tube should minimize reflux, thus making the pump positive displacement in nature. In a positive diSplacement pump, the output is dependent upon the speed of the pump, except for normal losses from internal leakage, or slip. In this case, slip will be due to incomplete tube occlusion. To evaluate slip in the pump, a needle valve and pressure gauge were installed in the tubing between the pump and flowmeter. Then, for a constant pump Speed, the pressure was set by adjusting the needle valve. The pressure was varied from 0 to 25 psi in increments of 5 psi for the tests. Figures 6.3.1 and 6.3.2 Show the results of this test for Cythion and Sevinmole 4. Because of pressure fluctuations of up to 1 psi caused by the pulsating output, it was impossible to obtain a steady reading on the pressure gauge. There- fore, an average reading was used. The fluctuations were all within 47 assm UHuHmumHummrrcochzu wch: Umom musmmmum msmpm> usauso .H.m.o oustm Hamav exemmmmm mN oN mH 0H m o .1 I. u a » ..AH 11 cc ulllllllllllk .vme [1 ACHE\HEV ow HDmHDO .Imm m23m 1r ONH wmoH . mmH ooH AEQHV Dmmmm mZDm 48 - gaze oHuHmumHuomurq oHoecH>om wch: seem ouammopm msmuo> usauso .N.m.o oustm Hamav mmommmma MN mN mH mH m 0 ..AH IIIIJ. .? om oe Acae\HeV Irma Haaeso mZDm . oo ow ONH mm ooH HEQHV ommmw mZDm - 49 - :tk psi of the selected evaluation pressures. Output remained fairly constant for both materials at all pressure settings at low pump Speeds. At higher pump Speeds, a small amount of slip was observed at the higher pressures. The largest output loss observed was only 6.8% and occurred at the highest pressure. Some minor difficulties were encountered in maintaining a prescribed pressure reading with the Sevinmole 4 due to the impingement of discrete particles on the surface of the needle valve. An interesting observation made during this test was that the needle valve restriction in the line produced a nearly pulsation-free output for all materials tested. 6,4, DESURGING THE OUTPUT The restriction caused by the needle valve in the previous experiment cauSed the conversion of velocity pulse, AV to a pressure pulse, AP. This pressure pulse took the form of a pressure wave which disturbed the system between the pump and the needle valve at a frequency equal to the pump speed. The pressure wave was absorbed by the elasticity of the tubing, friction forces in the tubing, and, to a small degree, fluid compressibility. The elasticity of the plastic tubing was the most significant factor in the pressure dissipation, as it visibly expanded and contracted with each pulse. To examine this tubing characteristic, another pressure gauge and an additional 50 feet of k-inch Tygon tubing were added to the system between the pump and the needle valve (Figure 6.4.1). Repeating the procedure used in the output versus pressure head investigations and observation of the flowmeter float and pressure gauges indicated that as much as 10 psi of pressure wave could be dissipated by 50 feet of k-inch Tygon tubing. - 50 - Another method for dispersing the pressure wave was attempted by installing an air chamber in the line between the needle valve and the pump (Figure 6.4.2). The air chamber consisted of a vertical 6-inch length of 3/8-inch pipe capped at the top and open to the output line at the bottom. The change in air volume in the chamber for a given pressure was calculated using the relationship. PIV1 = sz2 (6.4a) since the changes are essentially isothermal. The pressure pulse AP produced a change in air volume AV that is found by substituting P = P + AP and V = V - AV into Equation 6.4a yielding 2 1 2 1 v P _ 1 1 Av - v1 —————Pl + AP (6.4b) The between-pulse pressure drop was eliminated as the compressed air expanded to equalize liquid and air pressure. The volume of air in the chamber at atmOSpheric pressure using inside pipe dimensions was 27.4 milliliters. Applying a pressure of 5 psi to the chamber theoretically reduced the volume to 20.5 milliliters. Applying a 1 psi pressure pulse, caused by the pulsation of the pump output, to the chamber which was under a constant 5 psi load produced an instantaneous theoretical air volume change of about 1 milliliter which is sufficient to accommodate the displacement of the pump for one revolution thus allowing an even distribution of the output over the entire pumping cycle. A check of the flowmeter float and the output line indicated the output was pulsation-free. The essentially fixed-orifice flow impediment supplied by the needle valve was unsatisfactory because the pressure increased on the -51.. Figure 6.4.1. Pressure Dissipation Test Setup. Figure 6.4.2. Pressure Dissipation Test Setup Using an Air Chamber. - 52 - system as the flow rate was increased. To alleviate this condition, a pressure-accuated variable orifice (Figure 6.4.3) was placed in the line in place of the needle valve. This device was a fuel check valve which was modified by using a spring of Slightly greater calculated stiffness (K = 3.31bf/in) than the original. The spring chosen was of a length such that in the non-flow state, the spring was compressed enough to require a force of i-lbf acting on the %-inch diameter valve face to‘ initially open the valve. This corresponded to an opening or "cracking" pressure of about 5 psi. The testing procedure followed was to gradually increase the pump Speed, and thereby, the output, and observe the change in pressure as indicated by the pressure gauge. As the flow increased, the increasing pressure exerted more force on the valve spring forcing the valve open further. This system allowed the flow rate to be increased with a relatively small increase in pressure. However, this system still produced unsatisfactory pressures at high flow rates since the flow area in the modified check valve was limited by complete spring compression which required a force of only 0.66 1bf or a pressure of 13.5 psi. A valve consisting of a guided ball and spring (K = 1.42 1bf/in) was installed in place of the original valve (Figure 6.4.4). The testing procedure was the same used in the preceding experiment. Because of the improved flow channel in the ball valve the maximum line pressure was only 5 psi. Testing the output versus pump speed for the representative chemicals yielded essentially the same results as Figures 6.2.1 and 6.2.2. The maximum 5 psi line pressure caused only about a 3 percent output reduction at high pump Speeds. Figure 6.4.5 Shows the final peristaltic pump system configuration. - 53 _ Figure 6.4.3. Modified Fuel Line Check Valve. Figure 6.4.4. Ball Check Valve. - 54 _ Several conclusions were drawn from these tests. The additional length of tubing did dissipate pressure, but to use this method for a sprayer would have required enough extra tubing for at least four lines or about 200 feet of tubing. Also the expansion caused by the pressure wave caused an accelerated deterioration of the tubing. This was due to the exposure of the strained tubing fibers to the xylene base of most of the spray chemicals. Xylene is a mild solvent for Tygon tubing but does not seriously affect the tubing inruiunstrained state. The original variable orifice used did not perform well with Sevinmole 4 at low flow rates. Its con- struction (Figure 6.4.3) was that of a cylinder within a cylinder which created a large contact surface with close clearance. Discrete particles lodged on the surface disrupting normal orifice operation. 6.5. FIELD TESTS The peristaltic metering system was mounted on a Sprayall concentrate Sprayer (Figure 6.5.1) for field testing on experimental plots main- tained by the Michigan State University Departments of Entomology and Plant Pathology. Both registered and experimental chemicals were applied to a variety of fruit crops. Metering system tests were made only when applying a chemical whose effectiveness for a given pest had already been proven. Tests were conducted only on days when spraying conditions were ideal so that the metering system could be evaluated without consideration of factors pertaining to the rest of the delivery system, i.e. the fan and nozzles. The procedure followed was to calibrate the metering system to the prescribed dosage by catching the total output for a one minute interval in a graduated cylinder -55- ' Figure 6.4.5. Complete Peristaltic Pump System. a.” Figure 6.5.1. Peristaltic Pump System Mounted on Sprayall Sprayer. - 56 _ and making appropriate pump speed adjustments. While spraying, the nozzles were visually checked for uniform outflow. Upon completion of the plot, the output rate was checked and compared to the initial output. The standard procedure was followed for gathering the entomological data. A fixed number of leaves ‘was picked from a specific tree before the Spray was applied and the number of harmful or potentially harmful insects present was counted. At 3-day or l-week intervals after application, depending on the chemical applied, an equal number of leaves was picked from the same tree and the insect population tabulated. These consequent insect counts were compared with the pre- Spray counts, Showing the effectiveness or ineffectiveness of the Spray application. The results of the field testing were very encouraging showing excellent uniform protection of cherries against cherry leaf spot, a fungus disease, with a rate of chemical application of only 8 ounces per acre of a pure liquid fungicide formulation. This particular rate reduces to a metered output of 6 m1/min for the given spray conditions. The initial output was 6.2 mllmin and after spraying for 20 minutes an output check showed 6 ml/min or a reduction of only 3.3 percent. Another application for the control of a small insect called psylla on pears required a metered output of 72 m1/min. Initial output was 72 ml/min and the after-spray check was 70 ml/min or a decrease of 2.8 percent. The nozzles showed a uniform outflow during the entire Spray period. Entomological data (Table 3) indicated a substantial reduction in the psylla population. 57 HnoH .waHoaouom mo .uaoa .zuHmuo>H:D oumum cmmHnon "mousom annex: n c as mwwm H U u... NN we mm mm HoH we Homezoo mm mm NHH mm om om .muv mn.o um d osmnuuom MN Ne Hm m NoH oqN .muu m.H om e osmsuuom m S S on S as .ES m om e 2238 C 0 G G Gynyn 0k. 0N\m mH\m Nma woumHH mouse co usam ouo< pom HmHuoumz pom msmahz was mwmm :owumHosuom mo oumm HNmH .q noesouaom "unmaummus mo puma .momoausa >H= HON UonHooe somehow ummHnuHm HHmramuam m :H omussoa mums muomcommHv msH .EJON mo muHmouoa onHouucoo m suHs mm>oon Hausa nuHS wouuHm mummcoamHv umHaooomm LuHs ooSmuam ouo3 .oaoomumHsoocHn s have: voucsoo mums mama»: mHHmmm osH .uoHa some Eouw ooHaemm mums “sum you mo>moH ole :uHs musam oH mo Hmuou < .eoaoHan muoHa woumoHHaourcoa .ouom moo "moonuoz mHmeouma use mumom uuoHuumm nacho Hz .mHHH>coom .apmm mHoson uo>ouH ”soHumooH HNIHN "uaoaHuoaxm wcHuooH chHm mo muHsmomrr.m MHmom wch: pooam seam msmno> usauso .m.N.N ouame wch: womam assm msmum> usauao .N.N.N oustm 2a.: 85% EB eat 95% EB moH mm me mH moH me me mH r p F n n P P p -I P p p p p L b I a q q q 1 q d u q q q q d 1 a . 1 ,o .00 r 00 . acHe\Hev chE\HEV f 3950 .r 33.8 Asam mzzm v 03 .. OCH 1 4.. woeH nu qu 62 .QESm umoourcoHeuxo wch: was: ousmmopm mamum> usauso .H.m.n ousmHm Hamav mmammmmm mN oN mH 0H m o u x x m “ lrmH M on IT mq 1T HcHs\HeV cull-IllIlIlIlf 2 39:5 mZDm A... nrmoH ! 0o 11mm, oNH Hague Qmmmm mZDm 63 - .QESm Memoirs oHoEcH>om wchD one: ousmmoum msmpw> usauso .N.m.m oustm Ln N Hamav mmammmmm mH OH T- oN H on d. d!- .rmH .3 oo ONH Haauv ommmm exam 1 moH u mmH HcHe\Hev Homeao mZDm VIII. OTHER PUMP SYSTEMS Two other pump types were investigated as possible metering systems. A Jaco mechanically-actuated diaphragm pump(Figure 4.1.2) was considered. At the constant speed of 40 spm, the output could be varied from 0 to 63 ml/min. Output was controlled by "lost motion" type stroke adjustments. A Milton Roy "Minipump" reciprocating piston pump(Figure 4.1.1) was also considered. At a constant speed of 96 spm, the output was varied from 0 to 50 ml/min again using the "lost motion" stroke adjustment principle. Both of these pumps exhibited a flow-no flow output condition for all stroke adjustments thus necessitating desurging equipment. An interesting solution to this problem was proposed by Cleary and Bauer(l968). Using the full stroke, a programmed cam was used to operate two pump units ganged together to produce a pulsationless output(Figure 8.1.1). Output quantity was controlled by varying the speed of cam rotation. The programmed cam provided output over 220° of the pumping cycle,compared to the traditional 1803 which allowed overlap of the outputs of the two pumps. Although this system provided ideal output characteristics, it was economically unfeasible. Since two pumps were needed per output line, a minimum of eight pumps would be needed to outfit a sprayer which made the system cost-prohibitive. Use of the desurging equipment developed for the peristaltic pump system and a variable Speed trans- mission with the consequent reduction to four pumps per sprayer was still too costly. -65- ((4101!!! HING 4 «xxx Mame F‘““,”~ v‘ Ira/(r warn- "’7 ' ' 0 T 1 EVA/I Mxmmcu 5| I III; § " (INKIOQ' ’l. .I r]. / ‘ ' I ',,,, ‘ ;:: \ Ill/l. I . \‘7/4 I ___. '.j 'x. 5 ' 9,1,4 "“I I ’ . ' I I r ‘ " 7 A \ / d‘:\§ y rel ~ \:n«fia~ I I ' ’ gr ‘1‘ UJCMKUC § wrwva é . (mo """" ‘ tax/cm (W «we Str . I I WW ., WIMP CM] 7160(4’ Source: Cleary and Bauer, 1968. Figure 8.1.1. Pulsationless Flow Drive. IX. DISCUSSION AND CONCLUSIONS All of the original objectives of the project have been deomonstrated for both the gear pump and peristaltic pump systems except mechanical simplicity and economic compatibility. Maintenance of the peristaltic system included periodic replacement of the small section of chemical- resistant Viton tubing which wassubjected to constant mechanical manipulation by the steel fingers. The pressure-accuated variable orifice should be periodically disassembled and cleaned to insure against deposits which may build up on the ball or its seat. Minor maintenance included lubrication of moving parts. The gear pump required no lubrication except for the drive gear shaft since the liquid it pumped provided sufficient internal gear lubrication for the pressuresused. However, it is imperative that the pump be flushed with a clear liquid before any prolonged idle period since particulate material that might dry on the interior surfaces of the pump could cause pump failure because of the very close clearances. The peristaltic pump system would have a lower initial cost than the gear pump system. A Sigmamotor peristaltic pump of the type evaluated here has a provision for up to Six output lines at once which makes one pump unit sufficient for a Sprayer. The gear pump ($80.00) is less expensive than the peristaltic pump ($120.00) but can only supply one output line. Thus, at least four gear pump units are needed to outfit a sprayer making the initial cost of the gear pump system comparatively high. However, maintenance of the desuring equipment _ 66 - and periodic replacement of the very expensive chemical-resistant Viton tubing ($3.00 per foot) over the lifetime of the peristaltic pump tends to equalize the two systems economically in the long run. The gear pump showed superior performance to the peristaltic pump in the output tests. For a Speed range of 15 to 120 rpm, the gear pump displacement per revolution varied from 1.11 ml/rev to 1.23 ml/rev for both Cythion and Sevinmole 4, while the peristaltic pump for the comparable speed range of 20 to 160 rpm Showed a variation of 0.97 ml/rev to 0.85 m1/rev for Cythion and dropped from 1 ml/rev to 0.43 ml/rev with Sevinmole 4. Pressure head had a slight effect on the output of the peristaltic pump at high speeds and pressure. However the maximum reduction in output was only 6.8 percent. Pressure head effects on the gear pump output were negligible, the maximum observed reduction being 2.0 percent. On the basis of these observations, the gear pump system should be considered seriously as an alternative to the peristaltic pump system. Therefore, a program of field testing for the gear pump system is recommended. Field test of the peristaltic pump system verified the results obtained in laboratory tests and a similar outcome is expected for the gear pump system. REFERENCES Ag Chemicals. 1965. "Low Volume, Application Requires Supporting Data". _Agricultural Chemicals. 20: 115. Sept. AIChE. 1960. Pump Manual. American Institute of Chemical Engineers. New Yerk. Brann, J.L. 1956. "Apparatus for Application of Insecticides". Annual Review of Entomology. 1: 244-260. Brann, J.L. 1968. "Economic Implications of Concentrate Spraying". Proceedings New York State Horticultural Society. 113: 259-62. Canada Department of Agriculture. 1963. "Low Volume Air-Blast Spraying in British Columbia Orchards, 1957-1962". Publication 1191. Carter, G. 1969. "The Application of a Helical Gear Pump to the Handling of Non-Newtonian Fluids". The Chemical Engineer. Jan/Feb. pp. CElZ-CE16. Cleary, J.A. and Bauer, R.D. 1968. "Pulsationless Metering Pumps- Concepts and Achievements". Advances in Instrumentation. Vol. 23. Part II. 68-917. Proceedings of the 23rd Annual ISA Conference. New York. October 28-31. Courshee, R.J. 1960. "Some Aspects of the Application of Insecticides? Annual Review of Entomology. 5: 327-352. Farm Chemicals. 1967. "ULV, Will It Steal the Market?" Farm Chemicals. 130: lO-l4+, July. Fronk, W.D. 1962. "Insecticide Application Equipment." Fundamentals of Applied Entomology. Edited by R. E. Pfadt. The Macmillian Co. New York. pp 191-212. Guralnik, D.B. Editor. 1970. Webster's New Wbrld Dictionary of the American Languagg. Prentiss-Hall, Inc. New Jersey. p 1059. Hadekel, R. 1951. Displacement Pumps and Motors. Pitman & Sons, Ltd. London. 172 pages. Hetz,H.K. 1967. "A Basic Users Guide to Diaphragm Metering Pumps." Instrumentation Technology. Vol. 14. Dec. pp 45-49. Horn, N.W. 1968. "Gear Pumps." Machine Desiganluid Power Reference Issue. pp 10-12. Dec 19. Howitt, A. J. et a1. 1969."New Application Methods for the Control of Pests Attacking Fruit." Research Report 103. Michigan State Ag. Exp. Station Publication. E. Lansing, Mich. pp 9-14. -68 - - 69 _ Howitt, A.J. et a1. 1966. "ULV Ground Sprayers For Pests and Diseases Attacking Fruit." Council on Pesticide Application. Proceedings of Symposium. Oct 19-20. Howitt, A.J. 1970. "Ultra Low Volume-Its Place in Orchard Pest Control." Unpublished Paper. Department of Entomology. Michigan State University, E. Lansing, Michigan. Hydraulic Institute. 1965. Rotary Pump Section. Hydraulic Institute Standards. New York. May. Jaffrin, M.Y. and A. H. Shapiro. 1971. "Peristaltic Pumping." Annual Review of Fluid Mechanics. Vol 3. pp 13-36. Kristal, F.A. and F. A. Annett. 1953. Pumps. McGraw-Hill Book Company, Inc. New York. 373 pages. Latham, T.W. 1966. Fluid Motions in a Peristaltic Pump. S. M. Thesis. M.I.T. Cambridge. Mass. London, A. V. 1965. "Metering Pumps for Chemical and Process Duties." British Chemical Engineering. Vol. 10, N 6. June. pp 400-402+. Marton, F. D. 1963. "Metering Pumps". Instruments and Control Systems. Vol. 36. April. pp 86-88. Michigan State University. Department of Entomology. 1971. Entomology Research Report. Unpublished report. Michigan State University, East Lansing,Michigan. Neal, M. 1965. ”Survey of Peristaltic Pumps, Part 1." Engineering Materials and Design. Vol 8. Nov. pp 815-18. Nelson, C. 1969. "Pesticide Use Has Gotton Out of Hand." Farm Chemicals. 132:44 +. April. Nelson. 1971. A Bill to Amend the Federal Insecticide, Fungicide and Rodenticide Act. (S.660). Senate of the United States. 92nd Congress lst Session. Pending Legislation. February 8. Obey. 1971. A Bill to Amend the Federal Insecticide, Fungicide and Rodenticide Act. (H.R. 4596). House of Representatives of the United States. 92nd Congress. lst Session. Pending Legislation. February 18. Pigott, R. J. S. 1944. "Some Characteristics of Rotary Pumps in Aviation Service." Trans. of A.S.M.E. Vol. 66. August. Potts, 1958. Concentrated Sprpy Equipment, Mixtures and Application Methods. Dorland Books. Caldwell, New Jersey. Pumping Manual. 1962. Prepared by Editors of "Pumping." Trade and Technical Press. Surry, Morden, England. - 7o - Shapiro, A. H. et a1. 1969. "Peristaltic Pumping with Long Wavelengths at Low Reynolds Numbers." Journal of Fluid Mechanics. 37: 799-825. Stuart, R. and J. Heldeman. 1970. "Gear Pumps." Machine Design Fluid Power Reference Issue. Sept. 10. The Randolph Company. 1970. "The Randolph Pump." Catalog No. 36. USDA. 1962. "Extent of Spraying and Dusting on Farms, 1958, with Comparisons." Statistical Bulletin No. 314. Economic Research Service. Farm Economics Division. Weinberg, S.L. et a1. 1971. "An Hydrodynamical Model of Uretral Function." To be published in Proc. Workshpp Hydrodynamic Upper Urinary Tract. Nat. Acad. Sci. wash. D.C. Westcott, C. 1966. "Are Plant Sprays Necessary?" Plants and Gardens. 22: 4-6+. White, J.J. Editor. 1971. "Pump and Valve Selector.” Chemical Engineering. Deskbook Issue. McGraw-Hill. Oct. 11. Wilson, W. E. 1950. Positive Displacement Pumps and Fluid Motors. Pitman Publishing Corporation. New York.250 pages. APPENDIX AoHHsmuvmmv 8mm .ooaH .m.:U.H.< Hmonsom uMHH moo ISSUES! omumoomruon €88.23 Sea. I ASSESS. ammumHuHszlll. o>Humuocwwomv .II oaumcax HmuosmHuom.IIl_ ammum onaHmlll umHHoaeH ommoHo ammumHuHsz _ son potz.III uoHHoaaH coaoHSSm oumum onaHm WIIIIHmwstuucoo III. uoHHoaeH ommoHo Awwwwoflmwmw.llk owmumHuH: uoHHoaaH somoHESm mm: Hausa _llll uoHHoaeH oomoHo AHSOHW Hmewm III. mHZDm uoHHoaaH ammoHamm amoum onaH ammo Son IIII. uoHHoasH coa0.llk Emo.llI mom> I zumuom I. smuom , WWOEoomHamHQ - poumuoao have III_ m>HuHmom onaHuHsz zHHmoHomsooz UrIawmunamHa I onaaHm IIIJ woumnomo oHsHm onaHuHsz homesHm.llLII onumoouaHooM IIJ onaaHm I, SEES: muses. e339 _ soumHm III. ereeHm meeuo< eHmeHm monum momma mommmHo mmHuowoumo mMZDm mo mZOHHHm Hence: awmweamHo suHB onN.H saw oouo ou Nro SOumHa waHumuoumHoom .auoo copes: ammunesHo zuHs ooo.m Sam NHIo ou Hn\Hs mro :OumHa onumoouaHoom .oo MoumHsmaH mamH oe ae\He ewH-o ee e-o eoueamimmmueeoeeeeee seemeeuueem mmm ua\He omH-o as are oHuHeumHuem .oquoeHmeeM :Ha\Ha ooo.oHumoo.o oHuHmumHuom HaoHpoz choHumcuousH oonIN emwieuo on eae\He e-o aweereeem coo.m new omm-o oe m.o-o eoueae weeueuoeaaoem .oo eeeeuez-eHHHm o aHs\Ha oomrNeoo.o oHuHmumHuom .oo mauwummmervum>umm :mNImuo soumwmywcHumoouaHoom onoHB cm>raomamm cHe\muouHH mro UHuHmumHuom .auoo onHucoHom umaHouo ammunmmHo muon voHHHa saws soumua ooo.m new comic ou oxouum\Ha Hro wcHumuou pom wsHumoowaHoom .ooH.w:Huoumz pHsHm mNH maule.e-o on e: emeeeeeum .oo Seeuom e eeaeeam mH mu: eN\Ha H o» cHs\He mro oHuHmumHuom HmoHcmsoozrouuoon oom.m 3mm NHSIo oe .m.m-o eoumae,wmaumooueeemm .>Hmrmwauemmmmwm eomHemm ooqu sMMIOONIo cu eeuo coumHmpHHoaam ANMMuom muonuoum moH>mn ooH emu H-o ammeeeean meauumeeeH mean: meHm mNH emu one-o ea SH-o oresimwaeaHe.meeuom .oo ease neexoeHm ooqu nmewHIo ou Huo mam>ruoHHou .huouom .>HQ SSHuD coo.H new erm.o ou m.Hum.o Meow scam .humuom .auoo xHosam coo.H SSHHe ooe-m ea omH-m eoeeae wmaeeeoeeaoee oeeammwmeseemm eee\He o~-o ea ~-o eoueueimmwueeommaoem samwxeeexoem ANH remIWm-o ea are.e m.o-o eweereeam ooo.mH new Nmm~-o on ee\He H-o eoumae meaeeeoueaumm meaeumeeeH m-H-m oooon emmINHm-o ou mm.o-o eoSeHmeeaeeooeeaeem eHoeueoo your: amusemee oooeoo ImNMIHN.HmIo ou 5¢.oIo coumHm,meumooumHoom .ou usuasuumaH :moHuoa< ooo.oH emu oom-e on e.~H-mo.o eoueae meaumooueqeem meaameeamww,eoum< Amemv Showmoum xmz omcmm mama uauauomwscmz mMMMDHOou\He N¢.N ou NmN.o ummmuusmw .humuom .ou muosooum nuHcoN ooo.m ue\eeeuaH OH-o cu em.o-o emmwemeem nonsense mouse: mNH mum m.He oe m.H .ammmeeeae ooo.¢ new oow ou «.0 soumHa waHumoouaHooM .osH smauoHH w oomHHmz on 8mm oero oHuHmumHuom masm coucm> ooH emu eeH-o or m~-o eemwiweee .Samuom .oo ease HHHeuee ome oxouum\moocso oHIo coumHmrwcHumooumHoom .muoo nosumzruumsoum :Ha\Ha omH Ou mNo.o AoumHmv oHuHmumHuom om rem omu ou eae\He moo.o Heemmeamv eaeHeeeHeem .eeH.eeueaeemHm HmsoHumcuousH wH Anew omv :Ha\muouHH c on o oHuHmumHuom moHom ooaoHom cHawHa NoHro ou m.o:o .INMV msoumHQLNdHuwoouaHoom mmw woumuocow mHHmauouaH %n un\Ha cm on mo.o ooumsuom coumHm poemIonch moccasuumcH swam A3ouom vmoH :onHooua so oo>Hup ns\Ha OONHro owes mums son uamumaoo ooo.om un\Ha Hro moHHaasmv soumHm uonmIonon .muoo opossuumsH mxmam ooH ue\Ha ooo.mNro on ero mSoHHemH .ou mocwHHmm< noumommm oHe\Ha merm.H ou e.HuNo.o oHuHMumHuom mwoo mucoasuumoH e mooooum mH emmIe.~-o ea SH.o-o seesaw mmmweeoeeweem .oo erasure see: e eHeoz oom.H mww,oNHuo ou m.oIo coumHmrwaHumoouaHoom HmoH5o£o ocsumwm saw mdro ou oxouum\mmoup coumHmiwdHumooumHoom .oo unmasuumaH HmcoHumz oon+a new.mmm-o on H.o-o ecumemiweaeeooeeaeee .00 see SOSHHS ooo~wN smwimmmuo ou we.o:o ooumHm,wdHumooumHoom Idwmwdm pamHummoz rewINOm-o oe arHHe om-o eweeemeae eeeaee-ua< ooe.N new ooHro ou u:\Ha ooHro coumHm waHumooumHoQM .oo ucoaaHsvm couumz A.Szoov m xHozmmm< - 74 - .momH.mSSeHumcH vosHuc Illlll. l. UHHsmuomm "mouoom emeHHI 33% onchII II. , HmHucouom850HHo r r] ouom mHaHuHaz m see :2 H H “HWIIIIIIIonoHIJ meaHm ucoomouo oz Hmauouanl. ucoomouo use ooaHucD oaooonuumm omeHH HmoHHom Hmououxm uamm Teasm ammuom 393ml no: HH mHonmHMII umpEQZII. ocm> oHonoH oHonon easy oHonoHMII .Iuouom onchIL. ucoamomHa II. ImHQ mHanum> GOumHmuHuHsz II. ucoamomHa HmHemm rmHn uamumcoorlr eoueHm mHmeHm :oumamrl. unmaoomHa . Imwfl ®H£QHHw> HMHX< usoESOSHa ImHn uamumcoo vooomHmnas ~.uouMum :H ocm>III poocmHmm umamHHm ucosoomHmmHn uo.meHoeg. oom>Il. oHanum> uoxosm ovam acuom oH oom> poocmHmmmwwcoaoomHmmHn pooamenoa unmumcoo mmH%um mma%9 mommmHo mmzbm wmomuloumm umoem ucmumcou u< ousumuoaaos msmum> qumoomH> .N.Q oustm Am.v mepe muHmoomH> .H.a ousmHm a: mmseémmzma 00 MN 00 me om mm om me pI . b P P b . L I d . q a W a 1 llooN LrmH II- [T lruoom Li. C” pomaov or 2.38m; I iuooe I Q .roo... .. 8 LI Lr wfiHmOUmH> ZO muszAmzH mmDH .qu cH>omrroumm amonm ucmumcoo u< .mo ouHEOIISumm ummem ucmumcoo u< ousumuoa5oe msmuo> zuHmoomH> .e.o oustm opsumuoasoe msmno> mUHmoomH> .m.o oustm EL “28,23sz EL mmaémmzme. ca me me me co mm so we fi q .4 “ 00m v w u .IV JV L... 1. cos ..mm . 6 , nomaoe Ammaov 7. 1. a: +2 NHHmoomH> wHHmoomH> .v cow 1.moH Lu com 1.0+: H.Hzoov o xHezmmm< 77 .Um¢ wcmzuhmmrrhmwfim mo mumm ucmumcou u< ousumuoaSmH msmuo> muHmoomH> .o.o ousmHm fieov mmae T o v m ousumuoaeoH mamuo> zuHmoomH> .m.o opstm Aeveaéfifie ow we 00 me .1 m u " .rmH Hemauv 1.. CM weHmoomH> arms 1100 a- H.Hzoov o mezmmm< Hmoov 78 - oumHmnum quooHaurwomam aesm msmuo> uaauso .N.m oustm .Ume ocmfiuuomripmoam nasm msmuo> usauso .H.m mustm ea: 85$ .22 Eat mmmmm exam qu 00H 00 ON . OdH ooH o0 ON .4 x x m H .1 H i a .r .1 x H x .r x r 8 I; T om .8 HcHE\HEv »:HE\HEV 1. 39:5 .1 Seer: mzam .u ooH mama .rooH 41 .r I 0.: +0.: gr Ir mZDm UHHHmmurvooam mesa msmpm> usauso .q.m opswwm Haeev mmmmm exam oeH ooH om oN r P P b b p p L 1 d d d u N 1 N ivooH .mo ouHEOIIpooam assm monum> usauao .P roN woo H:He\HsV Hamaao mZDm A.Hzoov m mezmme< qu -r db Hague ommmm exam ooH - - 00 F d db .m.m mustm ON dI- L . ON T ow AcHs\Hav . Hamepo r ooH mZDm . oeH - 80 Hmoov oumHmzum quooHOrrOooam assm msmuo> usauso .N.m oustm Hague mmmmm exam .qu oconuuomuroooam maze msmpo> usmuso Agape mmmmm exam mm - .H.m meemmm mes me .me mH P b p n n - b p r 1 a d d q q d u 1 row T l 8 Heee\HeO .. Homeeo exam -1 2: :1 er wish mmmrrpmmam ease msmpm> unauso .q.m oustm .mO ouHEOIIOomam masm mompm> usauso .m.m mustm HeaeV ommmm exam Haeev mmmmm exam mOH me me mH mOH me we mH - n p p b p p p p F p b L p b b q a q a 4 H a d u q - J 4 a 4 q IHON rON AGHE\HEV AcHe\HEV 1' oo LIOQ HOmHOO HOmHOO 1. exam .. maze .uOOH .TOOH LT 1.! IIOdH 1::qu A.Szoov a xHozmme< MICHIGAN STAT NIVERSITY LIB 3 9 062 29 ARIES III I 1 EU 3 12931