ABSTRACT AGITATION IN LIQUID MANURE TANKS AS AFFECTED BY PHYSICAL PROPERTIES OF MANURE AND SHAPE OF TANK by Robert Harry Shaw The primary objective of this thesis, carried out by building models and using model analysis, was to deter- mine which shape of liquid manure holding tank could be agitated with the most efficiency. Fluid flow laws must be followed in the model analysis; and to maintain these laws, certain physical prOperties, viscosity and density of the liquid manure was needed. The study of physical properties was carried out on samples collected from several dairies in Michigan. Viscosity tests were run on each sample at varying moisture contents using a rotating Brookfield viscometer. The density and the moisture content of the original samples were de- termined by the Bio-Chemistry Laboratories at Michigan State University. Two model tanks (square and rectangular) were used for the agitation studies. The sides of the tanks were con- structed of plexiglass, so that the movement of fluid could Robert Harry Shaw be observed through the sides. A photographic technique was developed and used to record the flow lines of the fluid in the tank. The viscosity of the liquid used in the model tank was varied (to meet the model laws) using a cellulose product as a thickening agent. A recirculating agitation system was chosen for the study, which is commonly used for agitating liquid manure in holding tanks. The study of physical prOperties of liquid manure shows that it behaves as a pseudoplastic liquid with the density being similar to that of water. The study also showed that the viscosity of the liquid manure is primarily dependent on the moisture content and not on type of bedding. Of the two different shapes of model tanks under study, it was shown that the square tank lends itself to most efficient agitation. It is desirable to be able to rotate the pump discharge nozzle 360° and be able to adjust it in elevation. The study carried out in this thesis was primarily for dairy cattle, but similar methods could be used, and results obtained for liquid manure from other farm animals. ApproveZ/ $44 @144 Date: 3% /?é7 V / AGITATION IN LIQUID MANURE TANKS AS AFFECTED BY PHYSICAL PROPERTIES OF MANURE AND SHAPE OF TANK By Robert Harry Shaw A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Agricultural Engineering 1967 ACKNOWLEDGMENTS The author wishes to express his sincere appre- ciation to his major professor, Dr. James S. Boyd, for his guidance and patience during the course of this study and during the preparation of this manuscript. The author also wishes to eXpress thanks to his wife, Linda, for her encouragement and help in the prepara- tion of this manuscript. The author is indebted to Mr. Kenneth Fox, gradu- ate student in Food Science, for his advice and help ren- dered in the photography and photographic techniques used in the study and preparation of this thesis. ii TABLE OF CONTENTS LIST OF FIGURES . LIST OF TABLES I. INTRODUCTION . . . . . . . . II. OBJECTIVES . . . . . . . III. REVIEW OF LITERATURE . . . . . . . A. Liquid Manure and Liquid Manure Systems . . 1. Moisture content and contents of liquid manure . . . . . 2. Density of liquid manure 3. Excrement of a dairy cow A. Shape and sizes of hold- ing tanks . . . 5. Capacity of pumps B. Theoretical Considerations 1. Theory of models . . 2. Theory of viscosity 3. Rotational viscometer C. Review of Different Types of Fluid Flow . . . . . l. Laminar flow . 2. Turbulent flow . 3. Agitation and carrying capacity of fluids iii Page vi 10 10 12 1A 15 15 15 16 IV. V. VI. VII. APPARATUS AND TESTING PROCEDURE . A. Part I - Physical Properties 1. Collection of samples 2. Notes on bedding 3. Instruments used to measure viscosity. 4. Testing procedure. B. Part II - Agitation. l. Assumptions made in the design of model tanks UT-CLUN 6. Testing procedure RESULTS . A. Part I - Physical Properties 1. Description and characteristics of each sample 2. Density and moisture content of Thickening agent the samples. 3. Viscosity. B. Part II - Agitation. l. Viscosity Pump and nozzle size Photograph technique 2. Study of flow lines. CONCLUSIONS SUGGESTIONS FOR FUTURE STUDIES iv Construction of model tanks 17 17 l7 18 18 21 23 23 2A 2A 26 27 28 30 3O 3O 31 31 A6 A6 A8 6A 66 VIII. BIBLIOGRAPHY IX. APPENDIX 67 69 Figure 10 11 LIST OF FIGURES Shows Opening through which the liquid manure is put into a holding tank. Bruce Miller, Howell, Michigan . . . Tank that is used to spread the manure in the field. Quandt Dairy, Watertown, Michigan . . . . . . . . . . . . . . Another example of a tank that is used to spread the liquid manure on the fields. Bruce Miller, Howell, Mich. A typical example showing that the tOp of tank is at ground level. Ross Dairy, -Marlette, Michigan . . . . . . . . . Shows the chOpper impeller of a typical recirculating pump. Fred Braun, Saline, Michigan . . . . . . . . . . . . . . . Shows the recirculating pump in place. Quandt Dairy, Watertown, Michigan . Shows the relationship of different types of fluids; a - Newtonian, b — Pseudoplastic, c - Dilatant. . Shows the Brookfield Viscometer with the guard installed, and the beaker used for the test. . . . . . . . Shows the spindles, number A through 1, from left to right . . . . . . . Model tanks, built for the agitation tests. . . . . . . . . . Small roller radial pump used in the agitation test . . . . . . . . . . vi Page 13 20 2O 25 25 Figure l2 13 1A 15 16 17 18 19 2O 21 22 23 2A 25 Positions of pump nozzle and camera (9 of nozzle about 1 l/A" below the surface of liquid). . . . . . Viscosity vs. spindle speed of the viscometer for Case A . . . . Viscosity vs. spindle speed of the viscometer for Case B . . . . . . Viscosity vs. spindle speed of the viscometer for Case C . . . . . Viscosity vs. moisture content of liquid manure, with a constant spindle speed of 6 R.P.M. . . . . . . . . . Viscosity vs. moisture content of liquid manure, with a constant spindle speed of 12 R.P.M. . . . . . . . . . . Viscosity vs. moisture content of liquid manure, with a constant spindle speed of 30 R.P.M. . . . . . . Viscosity vs. moisture content of liquid manure, with a constant spindle speed of 60 R.P.M. . . . . . . . Rate of shear vs. shear stress for Case A. O O I O O O O O O O O O O 9 Rate of shear vs. shear stress for case B. O 0 O O O O O O O O 0 0 Rate of shear vs. shear stress for case C. O O I O O O O O O O O O Slope, n (from Figures 20, 21 and 22) vs. moisture content of liquid manure Flow lines in the rectangular tank. Viscosity of liquid is 157 ops. (Nozzle position B, view II, Ref. Figure 12). Flow lines in the rectangular tank. Viscosity of liquid is 157 cps. (Nozzle position C, View II, Ref. Figure 12). vii Page 33 3A 35 36 37 38 39 A0 A2 A3 AA “5 50 50 Figure 26 27 28 29 3O 31 32 33 3A 35 Flow lines in the rectangular tank. Viscosity of liquid is 62.5 cps. (Nozzle position A, view II, Ref. Figure 12) . . . . . . . . . . . Flow lines in the rectangular tank. Viscosity of liquid is 157 ops. (Nozzle position A, view II, Ref. Figure 12) . . . . . . . . . Flow lines in the rectangular tank. Viscosity of liquid is 157 cps. (Nozzle position C, View IV, Ref. Figure 12) . . . . . . Flow lines in the rectangular tank. Viscosity of liquid is 157 cps. (Nozzle position B, view IV, Ref. Figure 12) . . . . . . . . . . Flow lines in the rectangular tank. Viscosity of liquid is 157 cps. (Nozzle position A, view IV, Ref. Figure 12) . . . . . . . . . . . Flow lines in the rectangular tank. Viscosity of liquid is 62.5 cps. (Nozzle position A, view IV, Ref. Figure 12) . . . . . . . . Flow lines in rectangular tank. Viscosity of liquid is 157 cps. (Nozzle position A, View V, Ref. Figure 12) . . . . . . . . . . Flow lines in the rectangular tank. Viscosity of liquid is 157 cps. (Nozzle position B, View VI, Ref. Figure 12) . . . . . . . . . . Flow lines in the rectangular tank. Viscosity of liquid is 62.5 cps. (Nozzle position C, View I, Ref. Figure 12) . . . . . . . . . . Flow lines in the square tank. Viscosity of liquid is 1AA cps. (Nozzle position A, view II, Ref. Figure 12) . . . . . . . . viii Page 51 51 52 52 53 53 55 55 56 56 Figure 36 37 38 39 A0 A1 A2 A3 AA A5 Flow lines in the square tank. Viscosity of liquid is 1AA cps. (Nozzle position A, view V, Ref. Figure 12) . . . . . . . . Flow lines in the square tank. Viscosity of liquid is 68. 3 cps. (Nozzle position A, view IV, Ref. Figure 12) . . . . . . . . . . . . Flow lines in the square tank. Viscosity of liquid is 68.3 cps. (Nozzle position A, view IV, Ref. Figure 12) . . . . . . . . . . Flow lines in the square tank. Viscosity of liquid is l ops. (Nozzle position A, view VI, Ref. Figure 12) . . . . . . . . . . Flow lines in the square tank. Viscosity of liquid is 1AA ops. (Nozzle position B, View IV, Ref. Figure 12) . . . . . . . . . . . Flow lines in the square tank. Viscosity of liquid is 68.3 cps. (Nozzle position B, view I, Ref. Figure 12) . . . . . . . . . . Flow lines in the square tank. Viscosity of liquid is l cps. (Nozzle position B, view VI, Ref. Figure 12) . . . . . . . . . . . . . . Flow lines in the square tank. Note the fluid movement in center of pic- ture near bottom. Viscosity of liquid is 1AA ops. (Nozzle position A, view I, Ref. Figure 12) . . . . . . . . . Flow lines in the square tank. Center line of nozzle is at surface level. Viscosity of liquid is l ops. (Nozzle position A, view II, Ref. Figure 12) . Flow lines in the square tank. Center line of nozzle is 3 1/2 inches from bottom of tank. Viscosity of liquid is l ops. (Nozzle position A, view II, Ref. Figure 12). . . . . . . . . . ix Page 57 57 58 58 59 59 6O 6O 62 62 Table LIST OF TABLES Constants for equation No. 8. Constants to multiply the scale reading by in order to obtain the viscosity in centipoises . . . The amount of water added to each sample . . . . . . . . . Density and moisture content Equivalent viscosity of the prototype Viscosity data obtained on the three samples of liquid manure. . . . Page l5 19 22 31 A8 69 INTRODUCTION The methods of handling the waste products of farm animals as a liquid has progressed in the United States until we have what is known today as the liquid manure sys- tem. In this system, the manure is pushed, scraped or pumped from the barns or holding areas of farm animals into a large holding tank, usually through a slit in the top (see Figure l) in which the manure is stored. Some addi- tional liquid may be added to the manure to make it handle more like a liquid. After a period of time, the mixture of manure and liquid is agitated together, then the liquid manure is removed from the holding tank and spread on the fields. (Tanks are used as shown in Figures 2 and 3 to spread the liquid manure in the field.) To begin with this method was most widely used by swine producers. These systems were relatively small and ineXpensive. Today the system is also used by beef feeders and dairymen and the systems have become large, SOphisti- cated and eXpensive. The liquid manure system provides an efficient method of handling leppy manure. This method uses a mini- mum of physical labor to clean the holding and bedding area ....-. ,, . ' I Fig. l.--Shows Opening which the liquid manure is put into a holding tank. Bruce Miller, Howell, Mich. ,zuy-‘x‘ O o . OIOJOI' - a 8 Fig. 2.-—Tank that is used to spread the manure in the field. Quandt Dairy, Watertown, Mich. Fig. 3.-—Another example of a tank that is used to spread the liquid manure on the fields. Bruce Miller, Howell, Mich. Fig. A.--A typical example showing that the top of tank is at ground level. Ross Dairy, Marlette, Mich. of farm animals in that these areas may be washed down or the manure can be scraped into the holding tank periodi- cally. The manure handled in this manner has greater ferti- lizer value (Fitzgerald, 1965) than manure handled by some of the other conventional methods. There is less loss of plant nutrients due to evaporation and leaching and the liquid manure can be applied to orOps (Fitzgerald, 1965) when it will be most advantageous. Also, it provides a means of storing the manure through the busy planting and harvesting seasons and through the periods of bad weather. The manure handled in this manner helps to keep the fly and odor problems (Fitzgerald, 1965) at a minimum and usually the bedding requirements are less. Bedding is not required to absorb the excess liquid in the manure when it is handled as a liquid. The holding tank, usually built about eight or ten feet deep with the top or cover level with the ground (see Figure A), has been built to fit the space available for the structure. Also, many of the manufacturers of liquid manure systems have a size and shape that is stan- dard with them for manufacturing reasons. Little consid- eration has been given to determining an Optimum size or shape of the holding tank with respect to agitation. Agitation is the mixing of solids and liquids in the holding tank. The length and number of times that agitation is required with respect to filling and emptying the holding tank has the greatest dependence on the type of agitation system used. Some systems require that agi- tation of the holding tank be done periodically every day while other systems require agitation only at emptying time. In this thesis the recirculation type of agita- tion system will be considered. This system of agitation is in common use today. This system removes the liquid from or near the bottom of the tank. The liquid containing solids goes through a chOpper impeller pump (Figures 5 and 6) which cuts or breaks up the solids into smaller particles. The liquid or slurry is then ejected through a nozzle near the surface of the liquid. The Jet of liquid from the noz- zle is then directed toward the crust layer that may have formed on the surface of the tank. As the liquid comes in contact with the solids, they tend to dissolve and become a part of the slurry. Agitation is usually required Just before the holding tank is emptied. Fig. 5.--Shows the chopper impeller of a typical recircu- lating pump. Fred Braun, Saline, Mich. Fig. 6.—-Shows the recirculating pump in place. Quandt Dairy, Watertown, Mich. OBJECTIVES: To study the physical prOperties of liquid manure. To determine the effect of size, -shape of holding tank and certain manure characteristics on the ef- ficiency of agitation. REVIEW OF LITERATURE Liquid Manure and Liquid Manure Systems The following section concerns itself with general information on liquid manure and liquid manure systems. Moisture Content and Contents of Liquid Manure Liquid manure for dairy cattle is defined (Sobel, 1965) as a liquid above 95% moisture content and as a semi- liquid between the moisture content range of about 88% to 95%. In a survey (Speicher, 1965) of eight dairies in Michigan with liquid manure systems, it was determined that the average moisture content was 91.55%. The survey showed that the liquid manure contained an average of .OA7% phos- phorus, .A2% potassium and .30% nitrogen. It consisted of 1.65% ash and 6.93% organic matter. Density of Liquid Manure The average density (Speicher, 1965) from the sam- ples of the eight dairies was 8.52 lbs. per gallon. Excrement of a Dairy Cow Fitzgerald (1965) quoted the following figures collected by one manufacturer of a liquid manure system. The total excrements of a dairy cow is about 8% of the body weight per day. Therefore, a 1000 pound cow would excrete 80 pounds or 9.6 gallons per day. This is equal to 2A00 pounds per month or 290 gallons. The moisture content is about 8A%. These figures were comparable to other studies (Speicher, 1965) that about a 500 gallon capacity tank is required for a dairy unit1 per month when the moisture content of the liquid manure is about 91%. Shapes and Sizes of Holding Tanks Two shapes of holding tanks are commonly used. They are the round tank and long rectangular. The diameter of the round tank varies from 20 to 25 feet and is usually about ten feet deep. The common dimensions for the rec- tangular tank are eight feet deep, ten feet wide and some multiple of 30 feet in length up to about 120 feet. Another common size of a rectangular tank is 20 feet by 60 feet by 10 feet in depth. The latter example is made of precast concrete slabs. Capacity of Pumps According to the survey of manufacturers, the capacity of the pump for the recirculation system varies up to about 1500 gallons per minute. lMature dairy cow equivalent to 1 dairy unit; replacement heifer or steer equivalent to .75 unit. 10 Theoretical Considerations Models and model analysis can be used in the compilation of objective two. In order to relate the models to the prototype certain laws must be maintained. In this section, these laws and the theory of models are reviewed. In the second part, the theory of viscosity with certain relationships are discussed. These were found necessary to study physical properties of liquid manure. Theory of Models The objective for using a model is to determine or predict the behavior of the prototype. Models may be used, and are used in many cases, when the problems are too complicated for a mathematical or graphical solution. Models are also used to illustrate or study certain pheno- mena that takes place. The main reasons for using a model in this study were size considerations and the mechanics of handling large volumes. A general equation may be written for the proto- type-with the use of the Buckingham Pi theorem. n1 = F(n2, n3, "A°"°"s) (1) Two restrictions are placed on the N term. They must be dimensionless and independent. The number of n terms in the general equation is determined by the formula 8 = n - b where s is the number of w terms, n is the total 11 number of quantities involved and b is the number of basic dimensions. The same general law governing equation (1) can be applied to the model, so that N = F(n 1m 2m’ 1T3m’ “Am'°'°"sm) (2) Now by dividing equation (1) by equation (2) and letting it be equal to one: n1 F(n2, n3, flu....fls) l = n = F(i N N w 2m’ 3m’ Am"" (3) SITl It follows from equation (3) that n thus the dif- 1m = "1’ ferent dimensionless parameters for the model can be de- termined from the prototype so that for a true model. Textbooks on Fluid Mechanics (Olson, 1961 and Murphy, 1950) have indicated that for a small model that has a free surface of liquid both Reynolds law and Froude law are important for a dynamic similarity between the model and the prototype. Reynolds number Re, equal to the product of density p, velocity V, and length L divided by dynamic viscosity u, R =9? (u) e is important in the flow of Viscous fluid. Froude number Fr’ equal to the velocity V, divided by the square root of the product of gravity force g and length L, .V F = (5) r (gL)l72 12 is important where gravity forces govern the flow. A study of the dimensions of Reynolds and Froude numbers, where density is equal to lbm/Ft3, velocity is equal to ft/sec., length is equal to ft., dynamic viscosity is equal to slug/ft. sec. and gravity force is equal to ft/sec2 (American Engi- neering System); can be shown that they meet the requirement of the w terms in that they are dimensionless and independent. From the above discussion, it can be determined for a geometrically similar model such as the holding tank of liquid manure that the prOperties and the flow of the liquid should be controlled by Reynolds and Froude numbers. The dimensionless terms Of the model should represent as closely as possible those of the prototype in order to maintain a dynamic similarity between the model and prototype. Theory of Viscosity Viscosity n is a proportionality constant called the coefficient of viscosity which relates the shear stress 91 T between the layers of liquid to the rate of shear dt. The relation follows the equation _ T dt 'where n may be expressed in poises, T in dyne/cm2 and the IPate of shear has the unit of l/sec (metric system). Newtonian, pseudoplastic, and dilatant fluids are tShree common types of fluids. For an ideal Newtonian fluid Slich as pure water and many lubricating oils, the viscosity 13 remains constant as the shear rate increases. The viscosity is not constant (Behn, 1960) for the other two types of fluids. It varies depending on the molecule characteristics of the fluid and on the amount of certain suspensions that are in the fluid. The viscosity of a pseudoplastic fluid decreases as the shear rate increases (Illustrated by curve b or Figure 7). This type of fluid is very common. Curve c or Figure 7 illu- strates the behavior of a dilatant fluid. The viscosity in— oreaSes as the shear rate increases. This type of fluid is less common than the pseudoplastic fluid. Examples are found in some types of enamels and in some clays and sands. 3.7- A Stress dyne 2 cm at: , Shear Rate, l/sec. Fi . 7.--Shows the relationship of dif- ferent types of fluids, a - Newtonian, b - pseudo- plastic, c - dilatant. 1A The relationship between shear stress and shear rate of a non—newtonian fluid may be represented by the equation 2L3 = BTn (7) where B and n are constants depending on the characteristics of the material. However, when n = 1, it can be concluded that the fluid is newtonian. When n > 1 the material is pseudoplastic in character and when n < 1 it is a dilatant fluid. Rotational Viscometer Van Wazer, Lyons, Kim and Codwell (1963) stated that for a rotational viscometer the viscosity will have a relationship where n=K(stress term/rate of shear term) (8) K is known as an instrument constant. The stress term may be given in terms of dyne-cm, degrees of deflection, etc., and rate of shear term may be expressed in rpm, rps., and etc. In case of the Brookfield viscometer, model L.V.F., the scale reading may be equivalent to the stress term and the rpm would equal the rate Of shear term. The constant K varies depending on the spindle number. The K values are given in Table One. 15 TABLE 1 CONSTANTS FOR EQUATION NO. 8 Spindle Number 1 2 3 A K 6 3 12 60 Review of Different Types of Fluid Flow In the following section, short definitions of laminar and turbulent fluid flow are given. Laminar and turbulent fluid flow are important in relation to agitation and the carrying capacity of fluids. Laminar Flow In laminar flow in a liquid each layer travels parallel to the adjacent layers. The velocity of the layers are the same or slightly different and the particles in the fluid do not change layers. Turbulent Flow In this type of fluid flow, there are cross cur- rent velocities and there is a mixing of the fluid. The forming of eddies in the fluid is a characteristic of tur- bulent flow and their presence is resposible for the addi- tional amount of energy loss. l6 Agitation and Carrying Capacity of Fluids The greatest amount of agitation or mixing would occur in liquid where there is turbulent flow present be- cause of the cross current velocities. Little information was found on the carrying capacity of flowing liquids. The amount of particles that fluid will carry at a certain velocity depends on the size, density and shape of the particles. No information was available for liquid manure. APPARATUS AND TESTING PROCEDURE Physical PrOperties Collection of Samples Three liquid manure samples were collected by the farmers on different dairies in Michigan. The names and addresses Of these dairies are as follows: Case A - Virgil Pung, Ionia Case B - Ross' Dairy, Marlette Case C - Quandt's Dairy, Watertown Throughout the remainder of this thesis the above three samples will be referred to as Case A, B, or C. No set procedure was specified in the collection of the samples. Each farmer used a different method. Case A collected the sample by catching the drippings from the filler pipe as the holding tank was being emptied. This was accomplished just after the power to the pump was dis- engaged. Case B collected the sample of the liquid manure Off the top of the holding tank after it was agitated for ten minutes. The tank was full. Case B had a type of agitation system that required agitation twice daily during the summer months and A or 5 times a day during the winter months. Cases A and C had the recirculation type of agi- tation systems. Case C had collected a number of samples for the dairy department and one of the two quart containers 17 18 was given to the author for the study of physical properties. The samples were collected by taking a small sample from each load of liquid manure removed from the holding tank. Notes on Bedding Each farmer used different kinds and amounts of material for bedding. The bedding characteristics of each case were recorded. Case A used sawdust and bark shavings. This farmer used a sufficient amount of the bedding material and noticeable amounts were present in the sample. Case B also used sawdust for bedding but his supply had become de- pleted two months ago. There was no evidence of sawdust in the sample but there was some hay present. Case C used chopped straw for bedding. There was a noticeable amount of straw present in this sample. Instruments Used to Measure Viscosity A Brookfield viscometer, model L.V.F. was used as the instrument for measuring the viscosity of the liquid manure samples. This instrument measures the torque of the Spindle at a predetermined speed. Speed is defined as the revolutions of the spindle per minute. The torque required for a full scale reading of the instrument is equivalent to 673 dyne-cm. The scale is graduated in one-half units from 0 to 100. The instrument is calibrated by the manufacturer and maintains an accuracy of one percent of full scale with l9 guard on and in a container larger than 800 o.c. (see Figure 8). It is desirable to use the instrument near full scale in order to keep the error to a minimum. The range of viscosity reading at full scale is varied by using dif- ferent spindles at different speeds. The Brookfield viscometer, model L.V.F. has a range of four Speeds; 6, 12, 30 and 60 rpm. It has four standard spindles, one cylinder, two disks and one Straight shaft (see Figure 9). The viscosity values, in centipoise, are Obtained by multiplying the scale reading by the fac- tors given in Table Two. TABLE 2 CONSTANTS TO MULTIPLY THE SCALE READING BY IN ORDER TO OBTAIN THE VISCOSITY IN CENTIPOISES SPINDLE NUMBER RPM 1 2 3 u 6 10 50 200 1000 12 5 25 100 500 30 2 10 no 200 60 1 5 20 100 The constants given with the viscometer (as in Table Two) are usually valid only for Newtonian fluids. When this constant is used for computing viscosity for a 20 Fig. 8.--Shows the Brookfield viscometer with the guard installed, and the beaker used for the test. Fig. 9.—-Shows the spindles, number A through 1, from left to right. 21 non-Newtonian fluid, the viscosity should be referred to as apparent viscosity. Apparent viscosity is the ratio of the total shearing stress to total rate of Shear at a given value. Testing Procedure The procedure followed in taking the viscosity readings for the liquid manure samples was as follows: 1. The sample was mixed and a portion of the sample placed in a one liter beaker. 2. The liquid manure in the beaker was mixed again and a viscosity reading taken. The viscometer spin— dle was removed from the beaker and the liquid manure mixed again and another viscosity reading was recorded. The pro- cedure was followed until five readings were Obtained. 3. Step 2 was repeated for each of the four speeds of the viscometer. The lowest number of spindle was used at all times in order to utilize the maximum amount of the scale on the viscometer. A. Steps 2 and 3 were carried out and repeated with the liquid manure at the same moisture content as col— lected, and for four other varying amounts of water added to the sample. The amount of water added is indicated as a percentage of final volume (see Table Three). 22 TABLE 3 THE AMOUNTS OF WATER ADDED TO EACH SAMPLE % H20 ADDED % OF TOTAL VOL. AS COLLECTED a. 0 100 b. 12.5 87.5 c. 25 75 d. 37.5 62.5 e. 50 50 Steps one through four were carried out for each case; A, B, and C. The temperatures of the liquid manure in the beaker were recorded as the tests were run. Also, general notes on the qualitative characteristics of the liquid manure and the material (bedding) in the liquid manure were recorded. The final step of the test procedure was to de- termine the density and moisture content of the remaining portion of the liquid manure samples. This was accomplished in the biochemistry department laboratory under the super- vision of Dr. E. J. Benne. After the moisture contents of the original sam- ples were determined, the moisture content (appendix A) was calculated for the different stages (part a, b, c, d and e of Table Three). This gives a common reference for the data of each liquid manure sample. 23 The average, standard deviation and the 95% con- fidence range of the viscosity readings recorded in step two, were calculated. The confidence range gives the limits which may be assumed to contain 95% of the data from any future tests. Also it gives an indication of the accuracy and reliability of the data. The calculations were done by using the t distribution test, with a program that was avail- able for the digital computer. Some modifications were re- quired on the program in order to adapt it to the data. Agitation Assumptions Made in the Design of Model Tanks In order to apply the information to actual con— ditions, the tanks were models of a prototype tank of A0,000 gallons capacity. This tank would have usable space Of about 3A,OOO gallons. The loss of usable volume in the tank is accounted for by the material left in the tank when it is emptied and some space near the top of the tank which cannot be used. The accumulation in the bottom is due to imprOper agitation and tank characteristics causing some material to settle. The tank of this Size would have capa- city to store the excrement from a 35 cow dairy unit for two months. The prototype tank would therefore be ten feet deep. The rectangular tank would be ten feet wide by 2A 53.5 feet long and the square tank would be 23.1 feet on a side. The reduction factor for length in the models was eight which resulted in the model tanks inside dimensions being as follows: The square tank was 15 1/8 inches deep and 3A 5/8 inches square. The rectangular tank was 1A 15/16 inches deep, 80 3/A inches long and 15 1/16 inches wide. Construction of Model Tanks The bottoms for the model tanks were 20 gage sheet metal. One-fourth inch thick plexiglas was used for the sides (see Figure 10). Vertical corners were reinforced with an extra thickness of plexiglas and the sides were bolted together. A liquid cement was used in the corners and other areas to join the plexiglas. This made the seams waterproof and added strength as well. The edges of the sheet metal bottom were turned up and the plexiglas sides set inside. The sides were attached to the bottom and all joints were caulked. A steel frame constructed of 3/8 inch angles to fit around the top edges of the tank gave rigidity to the tank. Pump and Nozzle Size For the design of pump capacity, Froude modeling was used. The following equation was used, where Q is 25 . s , _ Q - .' l' I 'f“ 2" o Fig. 10.--Mode1 tanks built for the agitation test. Fig. ll.--Small roller radial pump used in the agitation test. 26 equal to the output of the pump in gallons per minute. LE 5/2 LP (9) Qm=Qp In using equation (9), it was assumed that the output of the prototype was 1500 gallons per minute and that gravita— tional acceleration for the model and the prototype are the same. In carrying out the calculation of equation (9), it was determined that the Qm desired was equal to 8.32 gallons per minute. The Size of the nozzle was determined by assuming that nozzle Size of the prototype was five inches in diam- eter. Dividing this by the scale factor eight indicated a model nozzle of 0.625 inches. A small radial pump with a small electric motor was used (see Figure 11). A primary check showed that the pump capacity was about eight gallons per minute when a one-half inch pipe (inside diameter equal to .622 inches) was used for a nozzle. Thickening Agent To meet the model similarity requirements it was necessary to use some thickening compounds that could be added to water, which would remain reasonable clear. It was desired to regulate viscosity of the liquid in order to satisfy Reynold's model number. A number of samples of a cellulose product were acquired for testing and a material called methocel Hg, Grade Standard, type 65 Hg, A000 cps, manufactured by Dow 27 Chemical Co., was best suited for the purpose. It was necessary to heat the water to near boiling point and then agitate it as the thickening compound is being added. The mixture should be allowed to cool to nor- mal room temperature before the viscosity is checked. The viscosity is temperature dependent. Water may be added and mixed into the cooled solution to lower the viscosity. Photograph Technique A study of flow lines by the use of models and model analysis was done in an attempt to accomplish objec- tive two. In the study of the flow lines, it was desirable to maintain a record. This was accomplished by developing a photographic technique. Aluminum powder was put in the liquid so that flow lines could be traced and photographed. A 35 mm camera with panatomic-x film was used for the photographic work. Kodabromide F5 print paper was used to acquire the maximum contrast. A lens speed ci'l/A second gave the best results for observing the flow lines. It was also desirable that only a narrow vertical strip of the fluid flow lines appear on the film to elimi- nate a complete blurr caused by all the aluminum powder in the solution. This was accomplished by providing a narrow Slit of light through the liquid so only the aluminum in this slit would be photographed. A box with a Slit 1/2 inch wide and 15 inches high was placed over a 750 watt photospot 28 lamp. A stand constructed of two 5 inch boards about 18" long placed 1/2" apart was placed between the light source and model tank. This helped to keep the narrow light ray from dispersing. The bottom of the tank was painted flat black and a piece of sheet metal painted black was placed in the tank on the Opposite Side from where the photographs were taken. The photographs were taken at night which helped to keep reflections to a minimum. Testing Procedure Tests were run from two positions in the square tank and at three different positions in the rectangular tank (see Figure 12) for each liquid. Tests were run with water and two other liquids of different viscosities. For each test, photographs were taken of each side of the tank and the top. Where the clearness of the fluid permitted, photographs were also taken with the light strip at varying distances from the side. As the photographs were taken, a sketch was made so a record could be maintained as to the direction of movement of flow lines. The movement on the film was a white straight line and could not indicate direction. Series of tests were also run with the discharge nozzle at varying depths. Tests were run with the nozzle 29 at liquid level and at four other depths. These tests were run to try to study and determine the desirable depth for the discharge nozzle for most efficient agitation. RESULTS Physical Pr0perties Description and Characteristics of Each Sample The liquid manure sample from Case A, which used sawdust and bark shavings for bedding, appeared to be quite thick. The sample had the flow characteristics that would be represented by a Bingham model. It requires a certain amount of shear stress before the fluid would flow. The sample appeared to be quite homogeneous in that the sawdust and bark shavings were evenly distributed. The bedding material did not tend to settle to the bottom or float to the tOp. Case B, which had hay present in the sample, ap- peared to flow more like a liquid than Case A. However, the hay material tended to settle to the bottom in a short period of time after the sample was agitated. Itwas also noted that the hay tended to matt up in a ball during agi- tation. The liquid manure sample with straw present (Case C) was very thin. The bedding material settled out as in Case B, but the straw did not tend to matt up in a ball as the sample was agitated. 30 31 Density and Moisture Content of the Samples The density and moisture content of each liquid manure sample is given in Table Four. TABLE A DENSITY AND MOISTURE CONTENT Case Density Moisture lbs/gal Content (% w.B.) A. 8.27 87.93 8.25 90.17 C. 8.33 92.71 The average density of the liquid manure samples was 2.8% less than the average in the survey indicated in the literature review. The average moisture content was l.A% less. Considering the type of material and methods used in the collection of the samples the density and moisture content of the samples were very close to the surveys. Viscosity The viscosity and other data related to viscosity or calculated as the result of the viscosity are recorded in Appendix A. 32 Figures 13 through 15 indicate that there is a linear logarithmic relationship between the apparent vis- cosity and the speed of the spindle of the viscometer. The 95% confidence limits for each average viscosity reading were also plotted. If in theory, there is a linear rela- tionship and in reality a straight line falls within the 95% confidence limits, then the straight line is justified. The linear straight line fell within the limits in all cases except two. This is good in that the linear logarith- mic relationship between the viscosity and Spindle speed is valid and also indicates that the viscosity data are reason- ably accurate. The curves were drawn in graphically to il- lustrate the different relationships between the variables. On these chart lines of nearly the same moisture content fall within the same region on the graphs and with similar SIOpes. The difference may be due in part to the different types of material in suspension. It is also noted that the Slope of the curves for each case becomes more posi- tive as the moisture content of the sample increases. This indicates that the viscosity becomes more of a constant with respect to the Speed of the spindle of the measuring instru- ment as the moisture content of the sample increases. The apparent viscosity was plotted against the moisture content of the samples with a semi-log scale (see Figures 16 through 19). Data from the three cases were plotted with the Speed of the spindle on the viscometer held 33 A W" B c H erad- I; e w—I v: p m l 80%” A v RECTANGULAR TANK A $T1 VHI 2 HI _’ 2%" I; 8 ea» «JL— Ho q. r) V” JLIV t—34I"——>1 VI V .SQUARE TANK Capital Letters — Position of Pump Nozzle Roman Numerals - Position of the Camera (view of the picture) Fig. l2.--Position of pump nozzle and camera (C of nozzle about 1 l/A" below the surface of liquid). IOO 90 80 70 ._ 6C) VISCOSITY, '7 x :02, CPS. 40- 3A — LINES OF CONSTANT o MOISTURE CONTENT (w.b.) o 87.93 °/. v 90.95 °/. A 89.49 °/. C] 92.6! °/. 0 93.96% TI‘II [> .n C» OT~JGHDC3 l l l J l2 3'0 60 SPEED, RPM. Fig. l3.—-Viscosity vs. Spindle speed of the viscometer for Case A. L 6 VISCOSITY, '9 x no“, CPS. 35 IOO - 9 LINES OF CONSTANT O MOISTURE CONTENT (w.b,) 60- o 90:? °/. v 92.63 % 50‘ A 9l-4O °/o D 93.86 % O 95.08 % w ..O 1 45 U‘ 05' mecoo l #L 6 l2 30 so SPEED,RPM. Fig. lA.-—Viscosity vs. Spindle speed of the viscometer for Case B. VISCOSITY, T x Io“,CPs. 20 A mm wa0 36 LINES OF CONSTANT MOISTURE CONTENT (w.b.) o 92.7I °/. V 94.53 '7. A 93.62 % CI 95.44.0/0 O 96.35 % [ . l l l 1- 30 60 2 SPEED,RPM. Fig. 15.—-Viscosity vs. viscometer for Case 0. spindle Speed of the VISCOSITY, ’9 XIO’, CPS. 20 OI\JQHDE3 Inca U: 37 r I 1 I 0 CASE A A CASE 5 D CASE C C ‘ “bosoo~ (9'88 89 L'Li i l l L L_ L l 1 0 9! 92 93 94 95 96 MOISTURE CONTENT % (w.b.) Fig. l6.-—Viscosity vs. moisture content of liquid manure, with a constant spindle speed of 6 R.P.M. 38 VISCOSITY/’7 x I 0‘, C PS. 50~(3 40. A 30— O 20- o A . El 8‘ l 3. 7w (3 6.- 5- 4, I A 3I' . 2, I 0 CASE A A CASE 8 IF A '3: I: CASE C D ‘7? .6*' 055 .4‘ .3“ IL I .'L._|____.|§_____§.U i 1 LL I ! l O 88 8 9| 9? 93 94 95 96 . MOISTURE CONTENT °/o Mb.) Fig. l7.--Viscosity vs. moisture content of liquid manure, with a constant spindle speed of 12 R.P.M. VISCOSITY, ’7 x I02, CPS. UICDNCDUJO l 39 20- 0 CASE A . A CASE 8 [:1 CASE C b umaam— 111T 90 SI 92 93 94 95 9E MOISTURE CONTENT °/. (w.b.) Fig. l8.--Viscosity vs. moisture content of liquid manure, with a constant spindle speed of 30 R.P.M. VISCOSITY, '7 x I0“, CPS. A0 20. O I0L 9.— 8r- 7.- 6- 5b- 4L. 3- 2— I _ .9- .8~ '7’ o CASE A .6I- .- .St A CASE 8 A“ El CASE C .3e .2- II ‘J I _L l l l l L -'0‘4'88 89 95 96 90 9| 92 93 94 MOISTURE CONTENT °/o (w.b.) Fig. l9.--Viscosity vs. moisture content of liquid manure, with a constant spindle speed of 60 R.P.M. A1 at constant. The data approximates a smooth curve which indicates that_the viscosity of the liquid manure samples was primarily dependent on the moisture content. The theoretical considerations indicated that the 91 rate of shear, dt, was equal to a function of the shear gl=BTn stress, I. It followed the relationship dt where B and n are constants. N is the Slope of the curve which indicates the type of fluid. (Speed of the viscometer represents the rate of shear while the scale reading repre- d_Y. sents the Shear stress.) The rate of shear term dt was plotted against the shear stress term (T), with a constant moisture content, on the log-log scale; Figures 20, 21 and 22 indicate that the relation- ship %% T BTn is valid for the liquid manure samples. The value of constant n was evaluated for each curve. Constant B was not determined because it is dependent on which spin- dle was used during the tests. The n value, larger than one, indicates that the liquid manure closely represents a pseudoplastic fluid. Graphs (Figures 20, 21 and 22) show that the n values de- crease toward one as the moisture content of the liquid manure increases. This indicates that the liquid manure becomes more Newtonian in nature as the moisture content approaches 100%. This can be verified by observing graph Figure 23, where the slope, n, is plotted against the moisture content of the sample. SPEED,RATE OF SHEAR TERM,'d7/dr m 0 0,0 O Case A. A2 o 87.93 °/.~I.C. (w.b.), n = 3.2I6 A 89.49% MC. (v.10), m 2.937 v 90.95 °/oM.C.Iw.h), I: 2.250 E] 92.6I °lo M.C.(w.b.), n: 2.406 O 93.96 *7. MC. (w.b.), n = 2.203 O NO, 4 SPINDLE USED A NO. 3 SPINDLE USED V, I: AND 0 No.2 SPINDLE USED dT/dIE-BT A A P V IA V I A 1 l L l 50 60 70 O90IOO I5 20 25 30 40 SCALE READING SHEAR STRESS TERM Fig. 20.--Rate of shear vs. Shear stress for SPEED, RATE OF SHEAR TERM, dT/dt 60 0‘ O I D Q. 0 A3 0 90.!7 °/.« M.C. (w.b.) , n = 4.484 A 9I.40°/oM.C.(w.b.), II = 5.328 V 92.63% M.C.(w.D.I, n = 2.2I8 CI 93.86 % M.C.(w.b.), n . 2.203 O 95.08 °/.'.M.C.(w.b.), n: 1,750 C AND A No.3 SPINDLE USED v NO.2 SPINDLE USED E] AND O N0.I SPINDLE USED dT/CII 2 BT" r I2- A O V 6r 0 A v 0 .IL IT L I I I I L I L I l I I0 50 60 70 80 90 I00 I5 20 25 30 40 SCALE READIN6,SHEAR STRESS TERM,T Fig. 2l.--Rate of shear vs. shear stress for Case B. SPEED, RATE OF SHEAR TERM, d7/dI AA 92.7I ox. M.C.(w.b.), n = 2.960 93.62 °/. M. C.(w,b.), n = 2.093 94.53 ox. M.C.(w.b.), n = I.875 95.44 °/.M.C.(w.b), n= I.703 96.35 °/. M.C.(w.b.), n= l.640 O ODO AND A No.2 SPINDLE USED Ap AND 0 N0.I SPINDLE USE-:3 OT/dI=BTn l J l l l L l l l L L No I5 20 25 30 40 50 60 70 8090 IOO SCALE READING,SHEAR STRESS TERM)?" Fig. 22.--Rate of shear vs. shear stress for Case 0. A5 .chscme OASUHH mo ucoucoo cthmfioe .m> Amm cam .Hm .om .mmfim Eopmv .c aOQOHmII.mm .mHm 4. .5328 $55.02 00_ mm mm hm mm mm vm mm mm 5 0m mm mm PIOQ / / // I_ // D / / 3.4/0 /./. nu IN od 4 o O n_ O In U mmdu D q 0 m mm