l ' . -‘- mwv' fig- . H ”L4H,“ .v “t"! n...» L'l-‘flv. v ,.\ 7‘ e v I :: 1"“) I ”3- 112' , v...,-.. u. .- . . ’o' 4 I“ «mu. on-r' '- . Q ~ ‘- I 1'“ _. '. 3‘," 9 ‘ 4fo . HELP-'51: , $30 9: ‘ : "*5: «2 Km " .VELLEEQT, «r «ya . 1—".1| '1 . _\ '4 Jlfii, . VJ. . . .., “Us: .2 "I. ,.-,,t K ’7 TH— ! ‘ ~ ' ‘ ' ' f. y- "E' .o. *1 #2:. . £2!- .37 =2? ’;. .h :g ' jJ :v ..=: 1‘4 ' "fl“ #5 ’w mafia" 4 ‘9'- It‘ a J“. 'i: '5"; a- "h 1.43%? .j.. :M-. \t‘ . '0‘. - X . .' ‘1 A V: - i. , . "Vault“. .- . u 1. ‘ . ‘ - A ' T Y I? . .. 0.‘ - . ‘ - . ‘ .7 ‘ fl _ no! '5 ,‘ 1 ' ' ' ' I f ,‘ ‘ ‘ ~' ‘ _ u ‘ 4. . Y ' 1' ' ‘ E b' ’1 I " H ‘ I J ‘ "‘ 'V 'l‘ ~ W" V‘ 1’ “‘9' ‘ I . m: .y. M .L‘V‘z‘f - ~~ Hr”. , ‘y ‘- n . 4“ ~‘_ w‘ I‘Lk‘. ‘;I|'lt'ntl, ' I. 4!- ..hH‘ o 0. ~ ". ,— \ -«w _ v vofi . n"; . --- . :z'w‘ 1;". ”I ‘1‘ _-' aw nu- - fifi‘tfi '1 1‘ .r ' 5;.‘r ' ' .. ,t-e‘ u u! :3?!» w"? ' V l‘ 4. I 0 'z7 .. 4;“ Of..'.\': .47.. ... .5. W 9- a“: f’ 53" ' - 2 "Ho . h t . ‘ ' ’I, 4"“! :dkvivi' ‘ .~ 21' ' ‘ ' H? mug - - my “I? 53. . I .Mv . {fli’b‘i-f“ .. :1 ‘ :l‘ , ' ,- , E“, T‘- ‘1!» , , , > . 43 ., . Q .- -. 5‘. .1’ m .. -. M 4‘ CS”- -..'.::;,-' aria. MICHIGAN STATE UNIVERSITY LIBRARI . llHilllzlglglllllglllglflllllll “ 9 WW 2 57 7 n: . LIBRARY Michigan State University This is to certify that the thesis entitled THE MODELING AND MEASUREMENT OF THE RELEASE, PRODUCTION, AND RETENTION OF CLOTH FIBERS IN A TOP-LOADING WASHING MACHINE presented by David John Fanson has been accepted towards fulfillment of the requirements for Master of Science degree in Mechanical Engineering Major professor 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE If t w ,t' ‘ I (I MSU Is An Afirmaflve Action/Equal Opportunity Institution THE MODELING AND MEASUREMENT or THE RELEASE, PRODUCTION, AND RETENTION or cwrn FIBERS IN A TOP—LOADING VASEING ucEINE By David John Fanson A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Mechanical Engineering 1989 SUIS‘ZIX ABSTRACT TEE MODELING AND MEASUREMENT OF THE RELEASE, PRODUCTION, AND RETENTION 0P CLOTH FIBERS IN A TOP—LOADING WASHING MACHINE BY David John Fanson A new lint measurement technique has been developed to measure instantaneous lint concentrations which occur in a top-loading washing machine during agitation. The technique, which is based on the transmittance of light through a water sample, provides quick and accurate measurements. Using information revealed through these measurements, a mathematical model was developed to describe the physics of the processes which cause lint to be suspended. Using this new lint measurement technique and the mathematical model an experimental study was completed. The focus of this study was to determine the effect of agitator operating conditions (stroke length and oscillation frequency) on the suspension of cloth fibers (lint). Knowledge gained from this study verified the lint model and provided the groundwork for several new relationships which pertain to the release, production and retention of lint. To my wife Wendy and my family iii LIST OF LIST OF TABLE OF CONTENTS TABLES FIGURES NOMENCLATURE CHAPTER CHAPTER CHAPTER 1 INTRODUCTION 2 LINT RELEASE MODEL 2.1 Introduction 2.2 Lint release model 2.3 Reactivated, produced, and retained lint 3 EXPERIMENTAL TECHNIQUES AND PROCEDURE 3.1 Introduction 3.2 Test apparatus 3.3 Standard fabric load 3.4 Lint concentration measurement technique 3.5 Experimental setup iv vi vii ix 11 13 13 13 15 15 16 x‘ 3.6 3. 3. 3. 7 8 9 Calibration of lint concentration measurement technique Sensitivity levels of lint concentration measurements Experimental procedures Parameter estimation 3.10 Repeatability 3.11 Evaluation of partition coefficient CHAPTER 4 RESULTS 4. 4. 4. 4. 4. 4. 1 2 3 lo 5 6 Experimental operating conditions Evaluation of model parameters Average angular velocity Low sensitivity data points Partition coefficient Final model Dimensional analysis CHAPTER 5 CONCLUSIONS 5. 5. 5. APPENDICES 1 2 3 Overview Conclusions Suggested research Program LINT Experimental data 19 20 21 25 31 31 35 35 37 39 42 45 45 49 53 53 54 57 58 58 96 3.1 3.2 4.1 LIST OF TABLES Repeatability test without rinse Repeatability test with rinse Order of washing runs vi 32 33 38 LAND») wa 10 WWW .11 4.23-c 4.3a—c 4.4 4.5a-c 4.6 4.7a-c 4.8 LIST OF FIGURES Lint release model Washing machine setup Light Transmittance Experimental setup Calibrations for sampling chambers Sensitivity of lint concentration measurement Typical data file Plot of lint concentration versus time Sensitivity coefficients of model parameters Typical plot of model fit to data Repeatability test without rinse Repeatability test with rinse Operating points for experimental matrix Model parameters versus stroke length Model parameters versus frequency Operating conditions with velocities Model parameters versus velocity A K versus measurement sensitivity Model parameters versus velocity Alpha versus average angular velocity vii 10 14 16 17 20 22 26 27 28 30 32 33 36 40 41 42 43 44 46 47 4.9 4.10 4.11 4.12 4.13 Beta versus average angular velocity Complete lint model Comparison of lint model to actual data Fanson linting constant versus velocity Nondimensional lint concentration versus time A nondimensional plot of all the data viii 47 48 49 51 51 52 Symbols A,B,C SI >’@ NOMENCLATURE Description Final lint model constant Estimation parameters Lint concentration Oscillation frequency Fanson lint constant Relative lint concentration Lint production rate Release rate Output power Input power Sensitivity Time Minimum time for linear model Normalized transmittance Partition coefficient Lint ratio Difference Volume fraction Average angular velocity Agitator stroke angle Calibration constant ix Units mg/L mg/L mg/L/min 1/min mV mw rad/min deg mg/L Subscripts 1,2,3 A,B,C d f Superscripts a e Final lint model constants Estimation parameters Drain Fabric Measurement Rinse Total Vater Dimensionless Reactivated Equilibrium Final Initial Produced CHAPTER 1 INTRODUCTION The cleaning of clothes using an automatic washing machine has become quite common in American households over the past three decades. To the consumer, washing clothes consists of placing laundry into the washing machine, adding detergent, selecting a particular type of wash cycle, turning on the machine, and removing the washed clothes to dry. The most important aspect of washing machine performance to the consumer is that the washer removes all visible stains and offensive odors. Commercial manufacturers have found that removing soil from clothes in the washing machine is relatively easy. All that is needed is water, detergent, and mechanical agitation. The more intense the agitation, the faster and more thoroughly the clothes are cleaned. But also with more intense agitation greater cloth damage occurs. Fabric damage exists in two forms 1) visible tears and 2) lint. Obviously, visible damage to the cloth must be avoided, but what about lint? To the consumer lint is only objectionable if it can be seen. Visible lint deposits occur when a very large number of lint fibers have aCcumulated in one location. Because lint is only a concern when very high concentrations exist, engineers have not put a high priority on thoroughly understanding the process of lint suspension. Even though, lint is always present during agitation. 2 Because the washing process has not undergone any major improvement in recent years, research interest has been allowed to shift from a "make it work" mode to a "why and how does it work" mode. Because of this change funding was made available by Whirlpool Corporation to study lint. Being in a university setting, a thorough and original scientific analysis was allowed to be completed. Before this study engineers have been satisfied with knowing only where lint ends up, and not knowing where, when, or how lint is produced. The scope of the research presented in this paper is summarized by these three objectives: 1. To develop a theory to describe the suspension of lint using a mathematical model. 2. To develop and implement a technique for the quantitative measurement of the amount of lint suspended in water, 3. To study the effect of stroke length and oscillation frequency on the suspension of lint. While completing these three objectives, many new and valuable observations have been made about lint suspension. By increasing our understanding of the processes associated with lint, it will become easier to diagnose problems related to linting. The chapters which follow in this thesis generally follow the objectives listed above. Chapter 2 describes the development of a theoretical lint model which includes the amount of lint suspended in water as a function of time. This model is based on several assumptions which have been theorized or inferred from observations. 3 The lint model includes parameters describing the lint production, the lint transfer from the fabric surface to the water, and the amount of lint which is initially adhered to the fabric. Chapter 3 describes the lint measurement technique used to produce the experimental lint concentration profiles. An explanation of the underlying physical principles of the technique (light transmittance) along with a specific example of its implementation in a laboratory environment are included. The functional relationship between amount of light transmitted through a lint-water sample and the mass of lint per unit volume of water is revealed from the calibration of the technique. The chapter also describes the experimental procedure used to obtain the experimental data. Using a parameter estimation technique raw data are reduced to a set of parameters which describe the dynamics of the linting process. Chapter 4 outlines a sample experiment which was completed in order to explore the potential benefits and problems associated with the new technique. The analysis includes a study of the effect that changes in stroke length and oscillation frequency of the agitator have on the production, the release and the retention of lint. Finally in Chapter 5 a summary of the most important conclusions and a list of recommendation for future research are presented. CHAPTER 2 LINT RELEASE MODEL 2.1 INTRODUCTION A mathematical model has been developed to help in the basic understanding of the suspension of lint during the washing of clothes. The model is used to quantitatively illustrate the origin and residence of lint fibers during the wash cycle. The theory is based on the following conceptual model of the linting process: Suspended lint fibers occur in the washing machine in two different states: 1) lint is suspended in the water and 2) lint is adhered to the fabric. The lint suspended in the water is produced by two mechanisms, 1) the continuous breakage and subsequent release of fibers due to mechanical agitation and 2) the release of loosely embedded fibers which were either deposited on the fabric by an earlier washing or broken in some way before the washing began (i.e. in the drier or from wear). Fibers not only move from the fabric into the water under agitation, but fibers in the water may adhere again to the fabric and reside there for a certain time. Therefore, the notion of residence time, well known in surface renewal theories, can be applied to describe the dynamics of textile fibers. 2.2 LINT RELEASE MODEL Several terms which are used in the development of the lint release model are defined: Equilibrium conditions - conditions under which the mass of fibers leaving the fabric equals the mass of fibers deposited onto the fabric, Lint concentration - lint mass per unit volume (fabric, water, or total), Relative lint concentration - lint mass per unit volume pertaining to the total volume (water plus cloth), Volume fraction - partial volume divided by the total volume, Partition coefficient - the ratio of suspended lint concentration to attached lint concentration under equilibrium conditions, Attached lint - lint fibers which are deposited on the fabric, . Suspended lint - lint fibers which are suspended in the water, Total lint — attached lint plus suspended lint, Lint production rate - the rate at which total lint is generated, and 6 Lint release rate - the rate at which attached lint is released from the fabric. In addition to these definitions the following variables are defined to facilitate the development of the mathematical model: a - partition coefficient, 8 — ratio of suspended lint concentration to total lint concentration under equilibrium conditions, 7 - volume fraction, C - lint concentration (mg/L), L - relative lint concentration (mg/L), K - lint production rate (mg/[L-minl), K' — release rate (1/min), t - time (min). In addition to these variables, descriptive subscripts and superscripts are implemented. The subscripts are used to describe location or type, for example, f - fabric, w — water, and t 7 total. 7 Similarly, the superscripts are used to describe a time or condition, as seen in, i - initial, f - final, and e — equilibrium. Using the aforementioned definitions and variables the following governing equations emerge: From the definition of the partition coefficient, e C" (2 1) E=E e f and Ce 3 . .3 (2.2) C H Using the following relation (for all times) between Ct, C", and Cf, Ct a 7“ C" + (1 - Yw) CE (2.3) a and B can also be expressed as, 6-87 a.__! (2.4) 1—81" and a 5.____ (2.5) a Yw + 1 - 1v 8 If the rate at which the total concentration increases in time (production rate) is assumed to be a constant, then dCt ——- a Kt (2.6) dt Similarly, it is assumed that the rate at which the lint concentration in the water increases with time (release rate) is proportional to the deficit lint concentration in the water. Thus, dC w , e 3:. a K (Cv - Cw) (2.7) Note that the rate of change in Cw in Eq. 2.7 is the net result of the lint being released from and redeposited onto the fabric surface. To solve this system of equations (Eqs. 2.1 - 2.7) and to find the concentration of lint in the water at time t, initial conditions must be given. We assume that at t . 0 all lint is adhered to the fabric giving the following equations: C a C at t . 0 (2.8) and C - 0 at t s 0 (2.9) The total lint concentration can be found using Eqs. 2.6 and 2.8, namely 1 Ct : Ct + Kt t (2.10) 9 Using Eqs. 2.2, 2.6, and 2.10, Eq. 2.7 can be rewritten as follows: dcw 1 ___ = e K' ct . a Kt K’ c — K' c (2.11) dt “ Finally, the solution is c = 5 [Ci - 55] (1 — e'K") + e Kt t (2.12) K’ As stated in the introduction, the suspension of lint into the water during agitation can be considered as the result of two separate processes 1) resuspension (or reactivation) of old lint and 2) the production of new lint. If we separate the suspended lint concentration from Eq. 2.12 into a reactivation (0:) and production (C5) concentrations, we get, a i —K't Cw a 8 Ct (1 - e ) (2.13) and p t — (1 - e‘K't) Cw . B Kt —'——— (2-14) K where a P cw . cv . cV (2-15) From Figure 2.1 in which Eqs. 2.12, 2.13, and 2.14 are plotted using arbitrary values for B, C1 K and K’ (0.5, 200, 0.5, and 5, t’ t’ respectively), we can observe qualitatively how each of these processes (reactivation and production) contribute to the suspension of lint fibers during agitation. 10 200- ‘ C v . Ca __— — A 150— ‘5 _J . c —.-_ j; w 15 . .5 12°? 1% __m____._ _.__._ é . _/" 8°: o —”T u T/ I: . —/ ':. -/ 40- -a—T’ I —/ 4/ ./ /‘/ 0 ‘fi‘fi’. I I I I I I I I I . . I . . I . I . I . I I . . 14T~1 0 5 10 15 20 25 30 Time (min) Figure 2.1 Lint release model As seen in this plot the reactivated lint concentration increases rapidly when the agitation begins but levels out as time increases. This initially rapid increase is due to the large difference between the equilibrium and suspended lint concentrations at the beginning of agitation. As the amount of reactivated lint approaches the initial equilibrium concentration, the reactivation term becomes constant. The production term appears to be proportional to time. At early times the reactivation term dominates the production term. As time increases and the total amount of lint suspended in the water increases, the slope of the plot approaches a constant value. This result is caused by the decreasing lint deficit and eventual dominance of the lint production term. 11 A series of experiments was carried out to verify the assumptions made in the model and to analyze the linting process. 2.3 REACTIVATED, PRODUCED, AND RETAINED LINT Lint is categorized to describe its origin or its fate. All lint concentrations are multiplied by their corresponding volume fractions (7) in order to make all concentrations pertain to the total volume. The following types of lint are described here: Reactivated Lint The total amount of lint which was initially adhered to the fabric that is resuspended in the water during agitation is given by, L: = e c: Yw (2.16) Produced Lint The total amount of lint which is generated during agitation and is suspended in the water is equal to, P , LV 3 8 7w Kt (t - 1/K ) (2.17) Final Lint ‘ The total amount of lint which is suspended in the water at the end of agitation is f LV a P - Lv + Lv ' 1 . a *v [ct + Kt (t — 1/K')] (2.18) Drained Lint After agitation the water containing suspended lint is drained. As a result some of the lint is removed. The amount of lint removed (drained) is shown by Ld a Cd Yd (2.19) Retained Lint When the water is drained from the washer the suspended lint is only partially removed. The remaining lint is deposited (retained) on the fabric. Thus, f Lr s Lv — Ld (2.20) Using these values along with the coefficients from the lint release model a quantitative comparison between washing runs can be made. With this ability, different operating conditions and/or washing setups may be evaluated as to their linting potential. CHAPTER 3 EXPERIMENTAL TECHNIQUES AND PROCEDURE 3.1 INTRODUCTION A series of experiments were performed to find the effect of particular operating conditions on the linting process. For these experiments washing runs were completed in a modified top-loading washing machine employing a newly developed technique for the measurement of lint concentrations in water. The lint concentrations recorded during the washing run were then used to solve for the unknown parameters in the mathematical model discussed in Chapter 2. A more complete description of the entire process is found in the following sections. 3.2 TEST APPARATUS For all experiments a basic top—loading Whirlpool washing machine was used. A number of modifications to the washer were made to provide control over some of the operating conditions. The steel outer housing of the washing machine was removed to allow greater accessibility to the drive mechanism and to the washing tub. The drive mechanism was modified so that the sweep (stroke length) of the agitator was adjustable. The standard drive motor was removed and replaced with a 13 14 1/2 horsepower variable speed DC motor. Also, the spin cycle and timer were disconnected causing the washer to run continuously in the wash cycle. In order to simplify the experiments and to concentrate on the washing cycle, all other cycles were disconnected. Since the spin cycle was not in use the washing tub suspension system was no longer needed and therefore, removed. The washer frame was then rigidly mounted to the floor. This configuration allowed experiments to be performed at maximum oscillation frequencies up to just under 3 Hertz with a stroke length range of 0 to 200+ degrees. Washing runs could be specified for any length of time. Figure 3.1 is a front view of the washing machine setup. Basket 7 T / Tub Agitator\~ / -——- Drive From / ~~ Motor /// /// Figure 3.1 Washing machine setup 15 3.3 STANDARD FABRIC LOAD The fabric load used in the experiments consisted of 30 pieces of white cotton fabric. The pieces were rectangular and measured 0.5 X 0.8 meters. Two fabric loads were used alternately to permit drying while another experiment was in progress. The sheets in both loads were selected at random from the initial fabric supply provided by Whirlpool Corporation. All of the fabric was then labeled to enable a record to be kept of the washing history of each group. 3.4 LINT CONCENTRATION MEASUREMENT TECHNIQUE Suspended lint concentration versus washing time plots are needed to implement the lint release model developed in Chapter 2. At the time this study was initiated a suitable quantitative lint measurement technique was unavailable. Therefore, to obtain the necessary raw data a new technique was developed. The new technique uses light transmittance through a lint-water sample as a basis for measurement. (Light transmittance is defined as the fraction of radiant energy that having entered a layer of absorbing matter reaches its farther boundary.) As light passes through a lint-water sample the lint suspended in the water causes the light intensity to decrease. The decrease is caused by the fine lint fibers in the water absorbing and dispersing the light. To determine the transmittance a coherent beam of light (a laser) with known power P1 is directed into a volume of lint-water mixture. The power of the light beam leaving the sampling chamber is measured. 16 This value P(Cw) is a function of the lint concentration of the water. The transmittance is then calculated by dividing P(C") by P1. Figure 3.2 illustrates the notion of transmittance. P1 P(C) ;/\’\‘,/\:/\‘\’,l\‘/\T\‘,l / \ / x x x \ / / \,/7 \ / / \ / Laser Beam Lint-water Mixture Figure 3.2 Light Transmittance In order to eliminate the effect of fluctuations in the water supply's transmittance, all measured transmittances are normalized with respect to the transmittance of the tap water. Normalized transmittance is calculated using the following equation P(C )/P1 P(C ) , _V__ 3 _V (3.1) P(0)/P1 P(O) where T’ equals the normalized transmittance and P(O) is the output power through the water with a lint concentration equal to zero. 3.5 EXPERIMENTAL SETUP To determine instantaneous lint concentrations in the washing machine during the washing cycle, water from the washing machine is sampled continuously and its transmittance measured. The experimental setup is illustrated in Figure 3.3. 17 Laser 50mpfing Power Chamber Meter :—— —— —— Laser Beam / -\ Circulation Pump Computer Washing Mochme Figure 3.3 Experimental setup The instantaneous sampling is accomplished by circulating some of the lint-water mixture from the washing machine through a transparent test chamber. The lint-water solution is continuously drawn from the area between the basket and the tub of the washing machine. This location was selected because it is close to the vicinity of the agitator while not allowing the fabric to obstruct the flow. Close to the agitator, the largest fluid velocities and the most complete mixing occur. It is assumed that this location will provide a representative approximation of the lint concentration in the washing load. After passing through the test chamber the liquid is returned to the top surface of the tub. The test chamber is a tube 15 millimeters in diameter. The intake and outlet ports are mounted on the sides of the tube and each end is covered with a 22 x 22 X 0.2 millimeter glass microscope slide cover. The slide covers are used to minimize the dissipation of light at the 18 boundaries. The length of the tube used is dependent on the range of lint concentrations to be measured. Higher concentrations need a shorter test chamber in order to maintain accurate measurements. Three different test chamber lengths were used in the experiments. The lengths are 79, 127, and 254 millimeters. The lint-water mixture is circulated by a rotary gear pump. The pump is driven by a 1/8 hp variable speed DC motor. This motor-pump system is capable of producing flow rates up to 45 liters per minute, but for all our experiments a flow rate of 9 liters per minute was maintained. A small impeller pump was used for the first series of test experiments but was found to be inappropriate. At higher lint concentrations lint began accumulating on the impeller. This accumulation eventually caused a flow restriction or blockage. The light source is a continuous 3 Watt Argon-ion laser made by Lexel (Model 95-3). For all experiments a one watt beam was selected. This amount of power was‘ not necessary but convenient for our equipment. A 10—15 minute warm—up time is required for the laser to reach a steady power level. A laser power meter is used to measure the power level of the incoming and outgoing beam. The power meter is a Surface Absorbing Disc Calorimeter. The. calorimeter converts the laser light to heat. A thermopile then produces a voltage proportional to the heat absorbed. A factory calibration data sheet states that 95.0 millivolts of electricity are produced per watt of laser light. The response time of the power meter is about 10 seconds. With this relatively long response time the meter averages out high frequency fluctuations. 19 The output voltage from the power meter is amplified by an operational amplifier with a 97.4 gain. Hence, the ratio of output voltage to light power is 9.25 volts per watt. The output voltage is measured and a normalized transmittance is calculated using a digital data acquisition system. The system consisted of a Digital Equipment Corporation, PDP 11/73 microcomputer with D/A and A/D capabilities. Output voltages are sampled at a rate of 21.25 Hertz. Because the washing cycles are run for 30 minutes and disc space is limited, every 12 seconds the average of 255 voltage values is calculated before the processing continued. These values are converted to lint concentrations using calibration data and stored in a data file. Details of the calibration procedure and of its results are discussed in the following section. 3.6 CALIBRATION 0P LINT CONCENTRATION MEASUREMENT TECHNIQUE The relationship between lint concentration and normalized light transmittance is obtained from a calibration experiment. For this experiment 7.5 grams of lint collected by the clothes dryer is rehydrated and suspended in the washing machine with 60 liters of water. While using the agitator to keep the lint uniformly suspended the normalized transmittance is measured. This process is repeated several times with different lint concentrations and for each of the three different test chambers. The results are plotted in Figure 3.4. Using this calibration a general functional relationship between normalized transmittance and lint concentration is established to be Cv a ln(T') A (3.2) 20 200- . 79 mm 1 135 mm -— 160- 250 mm ——--—- Lint Concentration (mg/L) o . .-..-fi” .-....fij 0.01 0.10 .00 Normalized Transmittance Figure 3.4 Calibrations for sampling chambers where A is defined as the calibration constant. The value of A was found to be a function of tube length and fabric type, but independent of fluid speed in the test chamber and laser power. For the three tube lengths of 250, 135 and 79 millimeters values of A for the white cotton cloth were found to be -17.1, -27.7, and -42.8 mg/L. 21 3.7 SENSITIVITY LEVELS OF LINT CONCENTRATION MEASUREMENTS The measurement technique’s sensitivity to changes in lint concentration is a function of both the lint concentration and the length of the test chamber. Using an accepted definition of sensitivity, the sensitivity of the measurement process is 3T' 5 =__c (3.3) w 3Cw . Solving Eq. 3.2 for T’ and differentiating with respect to Cw' Eq. 3.3 becomes S = [c—v] e(cW/X) (3.4) A A plot of the measurement sensitivity for the three different lengths of test chambers is found in Figure 3.5. From these plots it is observed that each test chamber has a peak sensitivity range. In general, the long tube has a greater sensitivity at low concentrations and the short tube, at high lint concentrations. 3.8 EXPERIMENTAL PROCEDURES In order to obtain accurate data using the lint measuring technique described above a detailed experimental procedure was established. A large portion of the measurement process is computer controlled. The FORTRAN program LINT was developed to assist the operator in performing lint experiment measurements and in following the experimental procedure. A listing of LINT is given in Appendix A. 22 0.50- ‘ 79 mm — 135 mm -——-—- 0.40- 250 mm —-—— Measurement Sensitivity Lint Concentration (mg/L) Figure 3.5 Sensitivity of lint concentration measurement The lint measurement procedure developed and used for data gathering consists of the following sequence of steps: - Select and enter parameters for the run - Using the program LINT the operator performs lint measurements for washing runs with various governing parameters. Responding to computer. prompts, the operator enters his/her name, the oscillation frequency of the agitator, the stroke length of the agitator, the duration of the test, the agitator being tested, the size and composition of the cloth load and the length of test chamber used. All this information is then stored in a data file created for this run. — Fill tub with filtered water - The washing machine is manually filled with 60 liters of filtered tap water. (The water filtration system consisted of two line filters installed in series, each 23 capable of removing suspended particles larger than five microns.) By filtering the transmittance of tap water approached that of distilled water. — Measure initial transmittance - The computer is used to start the circulation pump and measures the transmittance of the filtered tap water. This value is used to normalize all future transmittance measurements. After the data is collected the computer stops the circulating pump so the cloth load may be added. — Add cloth load — A dry fabric load (30 white cotton sheet) is added to the basket of the washing machine. - Agitate and continuously measure lint concentration - Before the experiment continues the drive linkage is manually adjusted to the correct stroke length. Then, at the operator’s, command the computer starts the circulating pump, starts the drive system at the desired oscillation frequency, and begins collecting the lint concentration data at the specified sampling rate (five. samples per minute). The computer maintains these conditions for the duration of the test (usually 30 minutes), then shuts off the circulating pump, stops sampling and signals the operator. All data gathered is stored in the established data file. - Drain and save linted water - After the washing stops the operator must quickly drain the tank. The tank is drained by gravity (neutral drain) by opening a value connected to the bottom of the tank. The drained linted water (drain water) is transferred to a holding tank for later measurement of lint concentration. 24 — Refill with filtered water - The washer (containing the wet cloth) is refilled by the operator to the original level with filtered water. (This step is added to enable rinsing) - Rinse and remove cloth — Each piece of cloth is taken out individually by hand, rinsed of excess lint and wrung of excess water. The clothes are then dried using a tumble dryer supplied by Whirlpool. - Measure concentration of rinse water - The computer again starts the circulating pump. The rinse water is agitated gently to ensure a homogeneous mixture. The computer then collects lint concentration data for the mixture. The averaged value is stored as the Rinse Concentration (Cr)' After the measurement has been taken the computer stops the circulating pump and the agitation. — Discard rinse water and refill with drain water - The operator removes the rinse water from the tank and pumps the drain water back into the washer for lint concentration measurement. - Measure concentration of drain water - The computer again starts the circulation pump. The drain water is agitated gently to ensure a homogeneous mixture. The computer measures the lint concentration of the mixture. This value is stored as the Drain Concentration (Cd)' After the measurement has been taken the computer stops the circulating pump and the agitation. — Discard drain water — The operator removes the drain water from the tank and prepares the washing machine for the next run. 25 The total time for the process is about one hour. All the data collected by the data acquisition system is placed in data files for further processing. An example of a data file is shown in Figure 3.6 and plotted in Figure 3.7. 3.9 PARAMETER ESTIMATION Referring back to the mathematical model developed in Chapter 2 (Eq. 2.12), it is observed that the concentration of lint in the water (Cw) is 1 function of the variables 8, Ci K’, Kt’ and t. From a data profile obtained from the lint concentration measuring technique, values for Cv and t are known. Using parameter estimation, values for the remaining variables and/or combinations of variables can be determined from the data profile. In order to estimate these parameters, Eq. 2.12 is simplified to the following form, cw = A (1 — e'3 t) + c t (3.5) where A . a be: - ES] (3.6) K, B . K' (3.7) c - e Rt 3 (3.8) Using a linear-nonlinear regression analysis, the constants A, B, and C are determined from a concentration versus time plot. The analysis technique was conducted such that the sensitivity coefficients are at maximums during evaluation. ~00 was ~0- ‘m ‘m we we we ‘0 van ~00 ‘- no we ~0- -m ~00 we- no no woo ~aa 26 Data File - D15018A.DAT Test Date — 03—MAR—88 Test Time - 19:28:00 Run Description — MATRIX Data Collected By - DAVE Cloth Load - A,B,C Agitator - REGULAR Dwell Time - 12 sec Oscillation Rate Samples — 1 Stroke Length Total Time - 30.0 min Agitator Motion Total Laser Power - 995. mW Initial Chamber Power — 660. mW Rinse Concentration - 56.6 mg/L Drain Concentration - 76.7 mg/L Figure 3.6 Time (min) Lint Concentration (mg/L) 0.0000000 0.0000000 0.2000000 0.3628047 0.4000000 2.963831 0.6000000 7.795936 0.8000000 13.49236 1.000000 18.95062 1.200000 24.60547 1.400000 29.65394 1.600000 34.01007 1.800000 37.25677 2.000000 41.24780 2.200000 44.45755 27.80000 109.0104 28.00000 108.7201 28.20000 109.0015 28.40000 109.5254 28.60000 110.0293 28.80000 110.1405 29.00000 110.0593 29.20000 110.2567 29.40000 110.3221 29.60000 110.6039 29.80000 110.7411 30.00000 111.2437 Typical data file - 1.50 Hertz - 180.0 degrees - Symmetric 27 150 125 C? \ ................... m .................. E 100 ................................ c ...................... '2 ........... g 75 ............ c on 3 .‘I' c 3 50 'E I '3 .' 25 f 0 I'- I u v V I v v w v I r r v I l v v v I l v I I v I ‘ r 7 ti 0 10 15 20 25 30 Time(min) Figure 3.7 Plot of lint concentration versus time The sensitivity coefficients are calculated with respect to each of the three parameters A, B, and C. Using Eq. 3.5, the following sensitivity coefficients are found, ac sAa—V-A.A(1—e'3‘) (3.9) M _ ac SB.—L'E.Aate'3‘ (3.10) an ac" SC . ——— C a C t (3-11) ac 28 Figure 3.8 is a plot of the above sensitivity coefficients versus time. For this plot estimated values of 50, 0.5, and 1.0 for the parameters A, B, and C, respectively, were used. The concentration versus time profile (Eq. 3.5) is also included in Figure 3.8. 200- : Cw . s __ 3160- SA a 1 . = 120‘ .2 I ‘5, . ._ _......_______ E . / § 80- ' 8 ‘ \ / / -: i/l' \ / / 4o« ‘ / ‘ \ / / I x / 0 ‘ / /\‘\ o 5 1o. 1.5....2b....2.5..fi30 Time (min) Figure 3.8 Sensitivity coefficients of model parameters The sensitivity coefficients are important because they indicate the magnitude of change of the response of the model due to perturbations in the values of the parameters. It is observed that the sensitivity of parameter B is high at small times, but decreases to almost zero at later times. 29 Because of the large differences in the sensitivity of the parameter B with respect to A and C, for large values of t, changes in B have very little effect on the value of Cw‘ Therefore, for large values of time, Eq. 3.5 becomes approximately equal to Cv = A + C t (3.12) The time in which this approximation is acceptable (t') has been found to be -ln(0.02) t' = (3.13) B Using a linear regression, the data from time t' to the end of the plot is used to determine the "best fit" values for the parameters A and C. After determining these two values, a value for B is found using a one parameter fit of Eq. 3.12 to all of the data. The basic Gauss_Newton method is used for this nonlinear approximation. Of primary interest are the values of the parameters 8, Kt’ K’, and Ci. Only three of these can be determined using Eqs. 3.6 through 3.8. In order to find a solution the following new parameters are introduced: ie 1 Cv . 8 Ct (3.14) and Kv . B Kt (3.15) 30 By substituting C:e and KV for c: and Kt’ the parameter 8 is eliminated from Eq. 2.12 leaving the equation c-cie Kw(1 4‘" Kt 311. w _ w — E7 — e ) + w ( - ) Using the values obtained from the parameter estimation and Eqs. 3.6, 3.7, and 3.8, the parameters K’, Kw’ and Cie are found to be ie Cw = A + C/B (3.17) K’ = B (3.18) Kw = C (3.19) The plot in Figure 3.9 illustrates how closely this model fits a typical data set. 150 N (1‘ 100 75 0| O Lht Concenhafion (mg/L) N U" 0 . . . . 5 . . 10 . . 13 . . . . 20 . . . 25 . . . 30 Time (min) Figure 3.9 Typical plot of model fit to data 31 ie The new parameter Cw can be interpreted as the equilibrium lint concentration in the water at time equal to zero. The meaning of the parameter Kw is the rate at which the equilibrium lint concentration in the water increases with time. By determining these three constants for a washing run, different operating conditions can be quantitatively compared with each other. 3.10 REPEATABILITY Two sets of preliminary experiments were carried out to test the repeatability of the lint measurement technique. For the first experiment six identical runs were performed without a rinse cycle. By repeating the experiments in this manner, lint concentrations "built-up" from run to run (See Table 3.1 and Figure 3.10). This accumulation of lint over several runs was considered undesirable as it violated the notion of repeatability. For the second set experiments a rinse procedure was added (described in detail in Section 3.8). This rinsing process removed the lint accumulated during the washing run. Table 3.2 and Figure 3.11 show the results of this experiment. Although some variations did occur between runs, they appear to be small (approximately 1 82). 3.11 EVALUATION OF PARTITION COEFFICIENT In Section 3.9 we found that all the constants from the ’ 1 t 9 simultaneously. To complete the data reduction one more governing mathematical model (B, C K', and Kt) could not be solved for equation is needed. 32 Table 3.1 Repeatability test without rinse Run Frequency Stroke Length Cie K' K (Hertz) (degrees) (mg/L) (1/min) (mg/Lymin) 1 2.00 100.00 52.63 0.4953 1.1283 2 2.00 100.00 67.73 0.3380 0.8940 3 2.00 100.00 73.50 0.3425 0.9191 4 2.00 100.00 78.88 0.2639 0.8540 5 2.00 100.00 79.63 0.2919 0.9627 6 2.00 100.00 86.63 0.2934 1.0242 120- 100$ 801 Lint Concentration (mg/ L) 8 l v—r 1 w I w v u u I w w I I I o ' ' ' 5 ' ' 1b ' ' ' '15 20 25 30 Time (min) Figure 3.10 Repeatability test without rinse In View of the previous discussion on repeatability it seems reasonable to assume that the total amount of lint generated during a washing run is removed by the drain and rinse processes. Thus, f 1 Ct - Ct = Cd Yd + Cr Yr (3.20) Lint Concentration (mg / L) 33 Table 3.2 Repeatability test with rinse Run Frequency Stroke Length Cie K’ K (Hertz) (degrees) (mg/L) (1/min) (mg/Lymin) 1 2.00 100.00 56.86 0.3464 0.7818 2 2.00 100.00 46.49 0.4277 0.8095 3 2.00 100.00 46.20 0.4083 1.0112 4 2.00 100.00 54.64 0.3028 0.7746 5 2.00 100.00 50.04 0.3317 1.0140 6 2.00 100.00 52.11 0.3403 0.9602 120~ 1005 80‘ .. : "muW"T35 E: 60‘ 4°? . :ItiI . II? 20" I':' 15' 0' ......,.........,.......-fi 0 5 10 15 20 25 30 Time (min) Figure 3.11 Repeatability test with rinse 34 Using Eqs. 2.10 and 3.15, assuming a 30 minute washing run, and solving for 8, Eq. 3.20 becomes 30 K B = ___—v (3.21) Cd Yd + cr Yr With a solution found for 6, all of the constants from the mathematical model can be solved for and the model is complete. CHAPTER 4 RESULTS In this chapter a complete linting experiment is presented. The effect of changes in agitator stroke length and oscillation frequency were studied. 56 washing runs were completed using various operating conditions. Lint concentrations were recorded during washing runs using the measurement technique discussed in Chapter 3. These data were then evaluated using the lint release model developed in Chapter 2. The following sections provide a sample of the insight to be gained by measuring and modeling the linting process. 4.1 EXPERIMENTAL OPERATING CONDITIONS The selected range of operating conditions used in the experiment was based on the range of conditions commonly used in consumer washing machines. The range of oscillation frequencies was 1.00 to 2.75 Hertz with a range of 100 to 200 degrees for the stroke length. The matrix of 28 operating points used for the experiments is shown in Figure 4.1. Each of these operating conditions was repeated twice to help reduce systematic errors and to investigate the repeatability of the process. Because of limited supply of fabric only two fabric loads 35 36 5.0011 1 + + l ,_‘1L501 + -+ + -5 1 3 . + + + + I 4 " ZJMIT + «+ + -+ + >. . 95’ j + + + + + E", 150‘: + + + + “' i + + + g . 33 L00-: + + ° 0.50: 0.001r'ir'r'lr"r'VTrr'rrfifvrfrrl 80 100 120 140 160 180 200 220 Stroke Length (degrees) Figure 4.1 Operating points for experimental matrix were used for the tests. When experiments were conducted consecutively, one load was used in the experiment while the other was being dried. In order to minimize the bias due to differences in cloth loads, a specific ordering for performing runs was established. When possible, the two runs at a particular operating condition were performed using different cloth groups. Each of these runs at a particular operating condition uses cloth groups with opposing histories (i.e. one run is performed with a cloth group from a wash with a low agitation frequency and the other run with a group from a high frequency). 37 Because of the high lint concentrations values encountered in some of the higher frequency runs, a shorter sampling tube was needed to make accurate measurements. As shown in Chapter 3, the shorter sampling tube generally has a greater sensitivity at the higher lint concentrations than a longer tube. Table 4.1 shows the chronological sequence of the washing runs. The table also includes, the stroke length, the oscillation frequency, the cloth load and the sampling chamber length used for each run. All runs are consecutive unless otherwise noted (for a few runs data were not collected due to equipment malfunctions). Lint concentration profiles for all runs are graphically displayed in Appendix B. A complete description of the operating conditions for each run in the matrix is given in Table 4.1. No other washing variables were varied during the testing. Any changes in the measured parameters, therefore, are a result of the differences in the operating conditions and/or the variability of the washing process itself. 4.2 EVALUATION OF MODEL PARAMETERS Using the parameter estimation technique from- Chapter 3, the _experimental data from all 56 runs were analyzed. Values were found for the model parameters Cie, K', and Kw' Figures 4.2a-c and b.3a-c are plots of these model parameters versus stroke length and oscillation frequency. As seen from these plots the parameters do not correlate well with either stoke length or oscillation frequency. 38 Table 4.1 Order of washing runs RUN STROKE FREQUENCY CLOTH Sampling Tube # (degrees) (Hz) LOAD Length (mm) 1 100 2.00 2 135 2 100 2.25 1 135 3 100 2.50 2 135 4 100 2.75 1 135 5* 100 2.75 2 135 6 100 2.50 1 135 7 100 2.25 2 135 8 100 2.00 1 135 9* 140 1.50 2 135 10 140 1.75 1 135 11* 140 2.00 1 135 12 140 2.25 2 135 13 140 2.25 1 135 14 140 2.00 2 135 15 140 1.75 1 135 16 140 1.50 2 135 17 180 1.00 1 135 18 180 1.25 2 135 19 180 1.50 1 135 20* 180 1.50 2 135 21 180 1.25 1 135 22 180 1.00 2 135 23 120 1.75 1 135 24 120 2.00 2 135 25 120 2.25 1 135 26* 120 2.25 2 135 27 120 2.00 1 135 28 120 1.75 2 135 29 160 1.25 1 135 30 160 1.50 2 135 31 160 1.75 1 135 32 160 2.00 2 135 33 160 2.00 1 135 34 160 1.75 2 135 35 160 1.50 1 135 36 160 2.25 1 79 37 160 1.25 2 79 38* 160 2.25 1 79 39 200 1.00 1 79 40 200 1.25 2 79 41 200 1.50 1 79 42 200 1.75 2 79 43 200 1.75 l 79 44 200 1.50 2 79 45 200 1.25 1 79 46 200 1.00 2 79 39 Table 4.1 continued RUN STROKE FREQUENCY CLOTH Sampling Tube # (degrees) (82) LOAD Length (mm) 47 180 1.75 1 79 48 180 1.75 2 79 49 180 2.00 l 79 50 180 2.00 1 79 51 140 2.50 2 79 52 140 2.50 1 79 53 120 2.50 2 79 54 120 2.75 1 79 55 120 2.75 2 79 56 120 2.50 1 79 * Non-sequential washing runs 4.3 AVERAGE ANGULAR VELOCITY Because no meaningful correlations were obtained using stroke length or oscillation frequency as independent variables, a new variable, average angular velocity (3), was introduced. Average angular velocity combines both the stroke length (8) and the oscillation frequency (f) into a new independent variable. More specifically, 2 n B . (2 9) -—— (60 f) (4.1) 360 simplifying ‘6 a 2.0944 8 f (4.2) The units for E are radians per minute. Figure 4.4 shows the relationship of average angular velocities to the operating conditions tested in the experimental matrix. 4O 1251 A ‘ + + ,3 1001 1 + 1 $ + 5 i g 75-: f $ $ 15 i .5» 1 + + * i f E 50 + '52} 251 o,....T.r.-r.........r.-.. O 50 100 150 200 250 Stroke Length (degrees) 0.8- 1 + + 1 + E j + i i i 'E + > 041 : $ + + t V 1 ‘ t x 0.2: J 0,0 ...e..........j..e......fi 0 50 100 150 200 250 Stroke Length (degrees) 3.0 ’5 + + + + g 2.5 + + i + E i + 3 2.0 4' :5- + + + } 1.5 + + + 5 + ; * g 1.0 f i + i + “E + V 0.5 ’ x 'rTrtT‘I‘rTrf' 000 h j—r ' r ff ' ' T r ' l O 50 100 150 200 250 Stroke Length (degrees) Figure 4.2a—c Model parameters versus stroke length 41 125 + + E100 # ¢$;+ i . ii;#* 575 $=l=+ 5., 1:4” 3; 50 '3325 ofifi f 0.0 0'5 ' ' ViTOr ' '1:5r Y ' .210 ' ' {Sm 3:0 Oscillation Frequency (Hertz) 0.0 + + + + 30.6 +++ +$+ +i§¥§$ + £04 1+++++i=+ 3‘0.2 0.0.. V'VU ittjfi 'IU' ' I 0.0 ”0.5 1T0 1:5 2:0 ”'23 H 3.0 Oscillation Frequency (Hertz) 3.0 2.5 2.0 'i' 1.5 '1.0 +++++ ++ ++ ++++ +0-40- + +00- ...—”+0.. «+40%- 4++++ ...,. ...,. ++ 0.5 0,0 r.......r-........r.,. 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Oscillation Frequency (Hertz) K. (milligrarns/liter/minute) 'ITI'V‘I‘ Figure 4.3a-c Model parameters versus frequency 42 .300 250 200 L50 L00 Oscillation Frequency (Hertz) (150 Y 1' T r I 0.00 r I Y I I Y Y Y I’ U T T 1 17 f I so 100 120 110 '100 180' ' 1200 Stroke Length (degrees) 200 Figure 4.4 Operating conditions with velocities Now, using this new variable as the independent variable, we again look at plots of the lint concentration model parameters (Figures 4.5a-c). As seen from these plots, definite correlations exist. These relationships will be discussed in Section 4.4. 4.4 LOV SENSITIVITY DATA POINTS As mentioned earlier in Chapter 3, the sensitivity of the measurement procedure, in general, decreases with increased lint concentration. This raises the question of when does low measurement sensitivity affect accuracy? To give some indication a plot of the variation in the parameter K versus measurement sensitivity is shown in Figure 4.6. A K is defined as the deviation of K from a power equation (calculated from a least-squares regression). 43 Figure 4. 125 g 100 + if. it ++ < +1: +++ + E 75 ++ ++ + g + + + g + 1* j; 50 + '73; 25 0 T T ' T I fit r r 1 T T ' ' fj 0 200 400 600 800 Average Angular Velocity (radians/minute) (18* 1 it . + ++ j§.(L6: + ft+ $+ a 1:? i; E 0.4- I I :3 ‘ + 5‘ 0.2-1 0.0‘. ....-fi 0 200 400 600 800 Average Angular Velocity (radians/minute) A 3.0 o E E 2.0 $+i E ++: ll: \ + E 1'5 +g #+ + ._«_:» 1.0 + tfi * =5- + ~’ (15 x. 0.0 Irat ' ' U ' ' I i ' 1 I ' ' I 0 200 400 600 800 Average Angular Velocity (radians/minute) Sa-c Model parameters versus velocity 44 1.0a 1 ’5 . g 0.s~ + .E ‘ é . s 1 + 'i 0.6- + a ‘ ++ 5 + fa“ - V r X, : ++++ _% 0.2' + +++ + + + c: ‘ + + j + t ##J ¢ +++ + + 0.0 r1 [#t+fv¢fvr*t r¥tur—ri7..uj #.r$11 0.00 0 05 0.10 0.15 0.20 0.25 0.30 Measurement Sensitivity Figure 4.6 0.: versus measurement sensitivity From this plot we see that the highest values of A K occur at low measurement sensitivities. Data collected at the low measurement sensitivities are the most likely to be influenced by measurement noise. By eliminating the data collected at low measurement sensitivities, the overall spread of the parameters is greatly reduced (from i 0.8 to 1 0.25 mg/L/min). The minimum acceptable measurement sensitivity is found to be, .SIn a 0.06 (4.3) 45 with the low sensitivity data points removed, a regression using the power equation reveals the following relationships between the model parameters and average angular velocity: cv . 0.669 0 ° (4.4) xv 2 0.00884 8 (4.5) Kw = 0.000248 8 1'75 (4.6) Figure 4.7a—c illustrates how well these functions fit the data. Functions other than the power equation may provide better "fits" or may even be closer to the actual process, but because of the limited range of 6’s tested we were unable to determine how the parameters behaved outside the test range. Therefore, a simple power law function (which passes through the origin) was used. 4.5 PARTITION COEFFICIENT The partition coefficient a and the related parameter 8 also ' correlate well when plotted versus 8, as illustrated in Figures 4.8 and 4.9. Applying the power equation regression to the data the following relationships are found, - 1.5 a . 0.0000125 m (4.7) 5 . 0.000775 8 (4.8) 46 125: i ‘gg roof t *+ 5 1 +1: + - + E 751 ++ + + an 1 :ij 1 ‘1' E, 50: + 2.- zs-j 0‘...,....e......1 0 200 400 600 800 Average Angular Velocity (radians/ minute) (LB- ‘ $ - + 3 06 . + at. 2 ‘ + + 'E . + \ 04. + + + + C J '1- =‘ (123 i 0.0 ' V ‘ I ' T I I r v 0 200 400 '600' t r800 Average Angular Velocity (radians/minute) 30 2J5 20 L5 'L0 0.5 K. (milligrarns/liter/minute) 01) - . , . . ,. . -.. .... , .3. 0 200 400 600 800 Average Angular Velocity (radians/minute) Figure 4.7a—c Hodel parameters versus velocity 47 114- 011 v r . r 0 200 Average Angular Velocity (radians/minute) T T 400 "600r ' '800 Figure 4.8 Alpha versus average angular velocity 118- +4- 0.6-1 * * 4: + O * + + + ~ . + cs 0.4 1 '1' + + + + t - + (124 0.0 V V I I 1* 1 l r j t I fie fifi O 200 400 600 800 Average Angular Velocity (radians/minute) Figure 4.9 Beta versus average angular velocity 48 4.6 FINAL MODEL Substituting Eqs. 4.4, 4.5, and 4.6 into Eq. 3.14 we find that, E 0.75 - _ 1.75 (1 - e‘az ‘” t) + a3 6 t (4.9) w a 81 where 0.641 O) H I 0.000884 m N I 0.0000248 N U) I (Eq. 4.9 is dimensionally homogeneous as long as ‘5 is in rad/sec.). The concentration of lint in the water appears to be only a function of average angular velocity and time. Figure 4.10 shows Cw versus time for several different B's. I I I I I I I fl I I I I' O ‘r rfi I I T r r I r I I I 0 5 10 ' 15 ' 20 25 30 Time (min) Figure 4.10 Complete lint model 49 Using Eq. 4.9, the data collected during the test runs can be compared to a generalized theory. By calculating the percent deviation of the actual data points from the theoretical Cw values (calculated using Eq. 4.9), a A CV is defined. By plotting this value (for all data points) versus time (Figure 4.11), an approximation of the equation's accuracy (1 20%) is demonstrated. L0 0.8 0.6 0.4 33 02 . . "._ - f‘ 00 4' -- ....-w ‘l:0'. H - _ ..[LI\LI',,. ---------- ‘émlaz-uunnu-H": H I "m'" """':"“""'""""‘ "1‘5"“ x\' 4 ;:'1:,,, ... wmnr. .mn: .. Dena ljnt Concentrafion -0£ —1_0fi2...,....,......-..,....,...., 5 10 15 20 25 50 fime(mh) Figure 4.11 Comparison of lint model to actual data 50 4.7 DIMENSIONAL ANALYSIS As seen previously, CV is a function of the model parameters 8, ie K' Cw’ , and Kw’ and t. Using these variables, their corresponding units and employing dimensional analysis, the following dimensionless values emerge: * Cw 4 C 8 01; ( 010) V '1: KW F . (4.12) K' cIe Using Eqs. 4.4, 4.5, 4.6, and 4.12, F (Fanson linting constant) is found to be a constant equal to 0.0417. Figure 4.12 shows F plotted versus average angular velocity. By substituting Eqs. 4.10, 4.11, and 4.12 into Eq. 3.14, the following dimensionless lint equation develops, 0* . (1 - r) (1 - e“*) + r t* (4.13) Figure 4.13 is a plot of this relationship. By nondimensionalizing the experimental lint profiles using Eqs. 4.10 and 4.11, all of the data collapses down to the single curve formed by Eq. 4.13. Figure 4.14 is a plot of 0* versus t* for all data points. The constant F is hypothesized to be a characteristic of the washing load. Because only one cloth type was used during this experiment, no concrete conclusions can be drawn. However, one could 51 Figure 4.13 0.10-[ 0.084 1 0.065 ‘ + “- + ‘1' + L + : i+ '1' .l. + +$ 0.04- + writ-1+ ¥ is ++ ‘ ++ 1 (102'1 l (100 . . . . . r r . . . .fi 0 200 400 600 800 Average Angular Velocity (radians/minute) Figure 4.12 Fanson linting constant versus velocity 2.0- 1.5~ 13' 1111 1155 0.02 . -.f...-...s-fi O 5 10 15 20 t0 Nondimensional lint concentration versus time 52 '20 ' 25 Figure 4.14 A nondimensional plot of all the data assume that different fabric types would have different F constants. Controlled experiments, like this one, could be performed for different cloth types in order to determine the effect on the linting constant. CHAPTER 5 CONCLUSIONS 5.1 OVERVIEV In the previous chapters of this thesis, a study of the suspension of cloth fibers in a top—loading washing machine was presented. The focus of the study was based on the following objectives: 1. To develop a general theory for the physical processes associated with the suspension of lint, 2. To develop and implement a technique for the quantitative measurement of the amount of lint suspended in water, 3. To study the effect of agitator stroke length and oscillation frequency on the suspension and retention of lint. Some of the main developments which occurred during the completion of these objectives were: - The development of a mathematical model which incorporated parameters that characterize the release and production of suspended lint, 53 54 - The development, calibration, and implementation of a suspended lint concentration (mass per unit volume) measurement technique which was based on light transmittance, - The development of software for digital data acquisition and parameter estimation procedures, - The utilization of the lint model, lint concentration measurement technique, and associated software to gather ‘ lint data from a set of experiments in which the stroke length and oscillation frequency of the washing machine agitator were varied, - The systematic analysis of the experimental data to determine the effect of changes in agitator stroke length and oscillation frequency on the release, production, and retention of lint. 5.2 CONCLUSIONS Before the conclusions of this research are summarized, it should be noted that the quantitative results obtained from the experimental study partially depend on the particular experimental setup. The main area of concern is the lint measurement sampling location. Water was continually drawn from beneath the agitator, and it is not known if this location is truly a representative of the entire washing machine. 55 Keeping this in mind, the following observations were made and conclusions were drawn from the results of this research. The comments have been sorted in to three groups, physics, methodology, and measuremen IS . Physics - Lint exists in two states, adhered to fabric and suspended in water. - The transfer rate of adhered lint into suspension is dependent on the amount of lint currently in suspension. - The rapid increase of suspended lint in the early stages of a wash cycle is primarily due to the suspension of adhered lint. - Increases in the later part of a wash cycle are due to the generation of new lint. . Methodology - Quick and accurate quantitative lint concentration measurements are obtained using the light transmittance technique. - The calibration constant for and sensitivity of this measurement technique are functions of sampling tube length and cloth type. - The build-up of adhered lint is greatly reduced by using a rinse cycle. 56 - Agitator average angular velocity effectively characterizes stroke length and oscillation frequency. Measurements The rate at which new lint is produced increases with angular velocity. - The transfer rate of adhered lint into the water increases proportionally with angular velocity. Using the mathematical lint model and parameter estimation good (i 2 X) correlations are found between theoretical and experimental lint concentration profiles. - A general lint equation which is only a function of angular velocity and time, satisfactorily models the amount of suspended fibers at any time for any of the tested agitator average angular velocities. Lint concentration profiles can be nondimensionalized using the dimensionless parameters, t*, 0*, and F. The lint constant (F) may offer comparisons between other untested conditions. Continued research and industrial implementation will refine the lint model and lint measurement technique into valuable developmental and evaluation tools. 57 5.3 SUGGESTED RESEARCH In the experimental study only the agitator’s stroke length and oscillation frequency were varied. Variations in any of the other operating conditions may produce different results. For example, changes in any of the following items will most likely have an effect on the release, production, and/or retention of lint: - Type of cloth load (i.e. cotton, polyester, terry cloth) — Size of cloth load (total mass of cloth) — Size of cloth pieces - Cloth to water ratio - Water sample intake location — Agitator - Cloth's age - Length of washing cycle - Rinse cycle - Method of drying cloth Since no studies (using these new methods) have been completed in which these conditions were varied, their effect on linting can not be evaluated at this time. Each of these conditions should be studied in order to fully understand the process of lint suspension. APPENDICES 0000000OOOOOOOOOOOOOOOOOOOOOOOOOOOOOCO APPENDIX A PROGRAM LINT LINT is the main driving program for preforming lint measurement. The measurement technique is based on light transmittance. Linking Procedure: Using the taskbuilder TKB type the following at the TKB) prompt: TKB> LINT . LINT, KSAM, TNKFRQ, NFILE, LNTUTL, LNTTXT, PDLDAT TKB> @[l,54]LNK2KLAB TKB) @[1,54]KCOM Written by: David J. Fanson Michigan State University Last Modification: April 1, 1988 Variable List: DWELL Sampling dwell FILEN Name of output data file FREQ Frequency of tank N Number of samples NFILE Subroutine to make plotting files 58 OOOOOOOOOOOOOOOOOO 0000 O 0000 95 POVER RATIO SAMPLE TNKFRO TNKPVR TOTPVR 59 Power of laser beam in milliwatts POWER/TNKPWR (Power Ratio) Subroutine to sample voltages from A/D board Subroutine to set frequency of oscillating tank Laser power through tank Total laser power An array of all points taken Average voltage from thermopile REAL VAL(256),CONC(1000),TIME(1000),B(5) INTEGER DVELL,MODE,N,I,J,NCHAN,SCHAN,MOTOR,TUEE REAL RATE CHARACTER*15 FILEN CHARACTER*40 XTI CHARACTER ANS BYTE ESC T,YTIT Create temporary storage file of measured values (in case program crashes) CALL ASSIGN(3,'LINT.TMP') ESC a "033 CALL DTOA(0,0.0) CALL DTOA(1,0.0) J z 0 N - 256 RATE . 50. SCMAN - 0 NCHAN - 1 MODE - 1 Set both 0 to A channels to zero Set sampling parameters Set plotting labals XTIT . 'TIME (min)’ YTIT . 'LINT CONCENTRATION (mG per L)’ MOTOR - 2 WRITE(5,95) FORMAT(/' Using READ(5,*) MOTOR Set motor calibration (1) Linear Tank or (2) Washing Machine : '3) Ask which sampling tube COCO O 96 98 ... 60 used WRITE(5,96) FORMAT(I' Which Sampling Tube Length ' /' (1) 25.4 cm, ' /' (2) 12.7 cm, or ' /' (3) 7.9 cm : 'S) READ(5,*) TUBE IF (TUBE.EQ.3) THEN Constant for 7.9 cm sampling tube CONST . -98.6 ELSE IF (TUBE.EQ.2) THEN Constant for 12.7 cm sampling tube CONST . -63.7 ELSE IF (TUBE.EQ.1) THEN Constant for 25.4 cm sampling tube CONST . -39.4 END IF Open output file for results TYPE * CALL OFILE(FILEN,XTIT,YTIT) Enter run text CALL TXT(DVELL,FREQ,NUM) Start run TYPE 122 FORMAT(' Hit "RETURN" to Start Sampling ..... 'S) READ(5,*) Turn on circulating pump CALL DTOA(1,3.1) Sample voltage produced by laser CONTINUE TYPE * TYPE * TYPE *,'Measure laser beam power in air : ' TYPE * CALL FSAM(VAL,VOLT,SCUAN,NCHAN,RATE,N,MODE) Convert voltage to milliwatts of power using known constants CALL CALPOV(TOTPWR,VOLT) Print measured value TYPE * TYPE * WRITE(5,98) TOTPWR FORMAT(' : Total power of laser . ',F5.0,' mW') Check power level to see if in range and COO 0000 100 99 105 61 give warning if not in range IF (TOTPWR.LT.10.0) THEN TYPE * TYPE *,'******** CHECK CONECTIONSI! ********' CALL BEL(3) ELSE IF (TOTPWR.LT.900.0) THEN TYPE * TYPE *,I******** povgg T00 LOWll ********r CALL BEL(3) ELSE IF (TOTPWR.GT.1017.0) THEN TYPE * TYPE *,'******** POWER T00 HIGHll ********' CALL BEL(3) END IF Sample voltage of laser through sample TYPE * TYPE *,'Measure laser beam power through test chamber : ' TYPE * CALL FSAM(VAL,VOLT,SCHAN,NCHAN,RATE,N,MODE) Convert voltage to milliwatts of power using known constants CALL CALPOW(TNKPWR ,VOLT) CONTINUE Write results on screen TYPE * WRITE(5,100) TOTPVR,TNKPWR FORMAT (' ;',/, ' ; Total Laser Power - ',F5.0,' mW',/, ' 3 Initial Chamber Power - ',F5.0,' mW',/,' ;') TYPE 99 FORMAT(' Do you want to change the power (Y,N) ? '8) CALL WTCHAR(ANS) IF (ANS.EQ.'Y'.OR.ANS.EQ.'y') GOTO 5 IF (ANS.NE.'N'.AND.ANS.NE.'n') GOTO 7 Shut off circulating pump CALL DTOA(1,0.0) Print sampling screen CALL LSCRL(5,8) WRITE(5,105) ESC FORMAT('+',A1,'[12;lH') VRITE(5,100) TOTPWR,TNKPVR VRITE(1,100) TOTPVR,TNKPVR TYPE *,' Time(min) Voltage(volts) Power(mW)', +' Concentration(mG/L)' 000 110 62 TYPE *,' ..................................... . ------- ' TYPE * Initialize sampling TIME(I) . 0. CONC(1) a 0. POVER . TNKPVR MODE 3 0 RATE . N/DVELL RATE 2 RATE*1.1 SAVEl . 0.0 SAVE2 . 0.0 I s 1 TYPE 123 READ(5,*) Start drive motor CALL TNKFRQ(FREQ,MOTOR) Start Circulating Pump CALL DTOA(1,3.1) CALL LSCRL(5,8) VRITE(5,110)I,TIME(I),VOLT,POVER,CONC(I) VRITE(3,110)I,TIME(I),VOLT,POVER,CONC(I) FORMAT(' ',I4,2X,F7.2,7X,F6.3,6X,F7.2,8X,F7.3) CALL STRTIM DO 10 1:2,NUM CALL MARK(45,DWELL,2,IDS) Sample from channel 0 CALL FSAM(VAL,VOLT,SCHAN,NCHAN,RATE,N,MODE) Check for errors and correct IF (VOLT.EQ.SAVE1) THEN VOLT - 2.0*SAVE1-SAVE2 END IF SAVE2 . SAVEl SAVEl - VOLT Up date time TIME(I) . (I-1)*DVELL/60. Convert voltage to milliwatts of power using known constants CALL CALPOW(POVER,VOLT) Calculate Power Ratio RATIO - POWER/TNKPVR Calculate Concentration IF (RATIO.GT.0.0) THEN 000 63 CONC(I) s LOGIO(RATIO)*(CONST) ELSE CONC(I) = 0.0 END IF Print results on the screen and in temporary storage file WRITE(3,110)I,TIME(I),VOLT,POWER,CONC(I) VRITE(5,110)I,TIME(I),VOLT,POWER,CONC(I) CALL WAITFR(45,IDS) 10 CONTINUE Stop Circulating Pump CALL DTOA(1,0.0) CALL BEL(3) WRITE(5,112) ESC 112 FORMAT('+',A1,'[1;24r') WRITE(5,113) ESC 113 FORMAT('+',A1,'[24;IH') CALL STPTIM('SAMPLE’) CALL STRTIM CALL CLOSE(3) Stop tank CALL TNKFRQ(0.,MOTOR) TYPE * TYPE *,'***** CALCULATING CONSTANTS FOR THEORETICAL FIT *****' TYPE * Calculate parameters for ' theoretical fit CALL FIT(NUM,TIME,CONC,B) BGNCNC - 3(1)+B(4) FNLCNC . B(1)+B(3)*TIME(NUM)+B(4) VRITE(1,117) BGNCNC,B(2),B(3),FNLCNC VRITE(5,117) BGNCNC,B(2),8(3),FNLCNC 117 FORMAT(' ; Initial Concentration - ',F5.1,' mG/L',/, + ' ; Transfer Rate - ',F5.3,' 1/min',/, + 4' ; Production Rate - ',F5.3,' mG/(L*min)',/, + ' 3 Final Concentration - ',F5.1,' mG/L',/,' 3') CALL STPTIM(' FIT ’) RATE I 2. MODE . l RNSCNC 3 000 GOO GOO GOO 15 16 17 118 130 20 64 CONTINUE TYPE * TYPE *,'Are you going to rinse the cloth (Y,N)? ' CALL VTCHAR(ANS) Start Circulating Pump CALL DTOA(1,3.1) IF (ANS.E0.'Y'.OR.ANS.E0.'y') THEN CONTINUE Sample voltage of laser through tank of rinse water TYPE * TYPE*,'Measure laser beam power through tank and rinse water:' TYPE * CALL FSAM(VAL,VOLT,SCHAN,NCHAN,RATE,N,MODE) Convert voltage to milliwatts of power using known constants CALL CALPOW(POWER,VOLT) RATIO . POWER/TNKPWR Calculate concentration of lint in rinse water IF (RATIO.GT.0.0) THEN RNSCNC . LOGIO(RATIO)*(CONST) ELSE RNSCNC . 0.0 END IF CONTINUE Print Results TYPE * - WRITE(5,118) RNSCNC FORMAT (' ; Rinse Concentration - ',F5.1,' mG/L') TYPE 130 FORMAT(I' Acceptable (Y/N) ? '5) CALL WTCHAR(ANS) IF (ANS.EQ.'N'.0R.ANS.EQ.'n') GOTO 16 IF (ANS.NE.'Y'.AND.ANS.NE.'y') GOTO 17 ELSE IF (ANS.NE.’N'.AND.ANS.NE.'n') THEN GOTO 15 END IF CONTINUE Sample voltage of laser through tank of drain water 000 119 25 107 120 30 121 31 65 TYPE * TYPE *,'Measure laser beam power through tank and drain water:' TYPE * CALL FSAM(VAL,VOLT,SCHAN,NCHAN,RATE,N,MODE) Convert voltage to milliwatts of power using known constants CALL CALPOV(POWER,VOLT) RATIO . POWER/TNKPWR Calculate concentration of lint in drain water IF (RATIO.GT.0.0) THEN DRNCNC . LOGIO(RATIO)*(CONST) ELSE DRNCNC . 0.0 END IF Print Results TYPE * VRITE(5,119) RNSCNC,DRNCNC FORMAT (' ; Rinse Concentration - ',F5.1,' mG/L',/, ' ; Drain Concentration - ',F5.1,' mG/L',/, I ;I) CONTINUE TYPE 130 CALL WTCHAR(ANS) IF (ANS.EQ.'N’.OR.ANS.EQ.'n') GOTO 20 IF (ANS.NE.'Y'.AND.ANS.NE.'y') GOTO 25 Stop Circulating Pump CALL DTOA(1,0.0) TYPE * WRITE(1,119) RNSCNC,DRNCNC WRITE(1,107) FORMAT(' ; Time (min) Lint Concentration (mG/L)',/, '; ---------------- ———',/,' ") D0 30 I-l,NUM VRITE(1,120)J,TIME(I),CONC(I) FORMAT('C',Il,2615.7) CONTINUE J - J+1 CALL MODPLT(NUM,TIME,CONC,B) DO 31 I'lgNU" VRITE(1,121)J,TIME(I),CONC(I) FORMAT('C',Il,2G15.7) CONTINUE XMAX . TIME(NUM) 66 CALL CFILE(FILEN,XMAX,0) CALL BEL(l) TYPE 123 123 FORMAT(' Hit “RETURN" to Continue ..... 'S) READ(5,*) CALL EXIT END SUBROUTINE CALPOV(POWER,V2) C--- Calibration for power meter ----- Cl . 0.95E-04 C2 . 0.011 C3 - 97.1 C4 - -0.770 IF (V2.GT.0.0) THEN V1 - (V2-C4)/C3 POVER : (V1-C2)/C1 ELSE POWER 3 0.0 END IF RETURN END c ...................................................................... C SUBROUTINE STRTIM STRTIM STPTIM are subroutines to keep track of elapsed time between events. STRTIM is called to start timing and STPTIM is called to stop timing and to print elapsed time. David J. Fanson 29-MAY-87 0000000000 --------- Variables in Common TIM --------- 0 INTEGER ITIME1(8) COMMON ITIM/ ITIMEl C Get first time CALL GETTIM(ITIME1) 67 RETURN END C ...................................................................... C SUBROUTINE STPTIM(NAME) C c ...................................................................... C --------- Variables in Common TIM --------- INTEGER ITIME1(8) COMMON ITIM/ ITIMEl C --------- Local Other Variables --------- CHARACTER*6 NAME INTEGER MIN,DIF(8),ITIME2(8) REAL SEC C Get second time CALL GETTIM(ITIME2) C Find difference between C first and second time C calls DO 10 I-1,7 DIF(I) . ITIME2(I)-ITIME1(I) 10 CONTINUE C Calculate time C difference in minutes C and seconds SEC . DIF(6)*1.+DIF(7)/60. MIN - DIF(4)*60+DIF(5) IF (SEC.LT.0.0) THEN SEC . SEC+60.0 MIN - MIN-1 END IF C Vrite routine name and C time WRITE(5,100) NAME,MIN,SEC 100 FORMAT(/' Total CPU time used in ',A6,' routine a ’,I3, ' minutes ',F6.3,' seconds') RETURN END C _________________________________________________ _ ---- C SUBROUTINE LSCRL(IU,N) c ...................................................................... 00000000000000 000000000000000 0 68 LSCRL causes the unit number device (IU) to scroll "N" lines. Written by: David J. Fanson Michigan State University Last Modification: July 16, 1987 INTEGER IU,N Write "N” blank lines to unit "10” DO 10 I-1,N WRITE(IU,*) 10 CONTINUE RETURN END SUBROUTINE SCROLL(TOP,BOTTOM) SCROLL sets the scrolling region on the terminal. Written by: David J. Fanson Michigan State University Last Modification: July 16, 1987 --------- local other variables -------—- BYTE ESC INTEGER TOP,BOTTOM ESC . "033 Set scrolling region WRITE(S , 100) ESC,TOP, BOTTOM 69 100 FORMAT('+’,A1,'[’,IZ,';’,IZ,’r’) RETURN END c ...................................................................... C SUBROUTINE CSCRN C c ...................................................................... C C CSCRN clears the whole srceen. C C ...................................................................... C C Written by: C David J. Fanson C Michigan State University C C Last Modification: C July 16, 1987 C C ...................................................................... C --------- local other variables --------- BYTE ESC ESC . "033 C Clear screen WRITE(5,110) ESC 110 FORMAT('+',A1,'[2J') RETURN END c ...................................................................... C . SUBROUTINE BEL(N) C c ...................................................................... C C BEL will immediately beep the bell of the calling terminal. C C Peter J. McKinney 7-aug-1986 C C ...................................................................... BYTE BELL 1 define BEL as type BYTE BELL-'7'O ! octal code for the bell character (BEL) C Ring bell N times C Write to user's terminal 70 C (logical unit 5 by default) C as a character WRITE(5,100)BELL 100 FORMAT('+',A1) 10 CONTINUE END c ....................................................................... C SUBROUTINE WTCHAR(CHAR) C c ....................................................................... C . C WTCHAR returns the the first character in the keyboard buffer C in BYTE form. If no character is available program execution C is suspended until a character is entered. C C Written by Peter J. McKinney -— Michigan State University C C Reference: Engineering Computer Facility -- M.S.U. C C Last Modification: August 11, 1986 C c ................................ - ................ INTEGER DSW l directive status word CHARACTER*1 CHAR 1 character input LOGICAL*1 158(4) ° 1 I/O status block as byte array INTEGER IOST(2) 1 I/O status block as integer array INTEGER PRAMIN(6) 1 device dependant paramater array BYTE CHARS(2) 1 buffer count call result EQUIVALENCE (IOST(1),ISB(1)) 1 integer byte versions of I/O status CALL GETADR(PRAMIN(1),CHAR) 1 get address of WTCHAR PRAMIN(Z) . 1 1 set to read one character CALL WTQIO("1010.0R."20,5,1,,IOST,PRAMIN,DSW) 1 read all bits with no echo IF((DSW.NE.1).OR.(IOST(1).NE.1)) CALL ERRORS(ISB,DSW) 100 ' RETURN END c ....................................................................... C SUBROUTINE ERRORS(ISB,DSW) C c ....................................................................... C C ERRORS returns the values of the Device Status Word and the C I/O Status Block if an error occurs durring an I/O operation. C c ....................................................................... INTEGER DSW LOGICAL*1 ISB(4) ! I/O status block as a byte array WRITE(5,100)ISB(1),DSW 100 FORMAT(/,' ISB . ',I8,/,' DSW = ',I6) CALL EXIT END 0000000000000000000000000 00000 Subroutine IPOKE and Function IPEEK PEEK POKE routines written for the DEC PDP11/73 running under RSX—llM-PLUS. These functions subroutines appear the same to FORTRAN as the RT-ll calls of the same name. KCOM is a common area set up by RSX that resides directly over the device registers (addresses 160000 to 177777 octal). The following routines set up arrays (IREG IREGB) that coincide with this comman area. By reading from and writing to this array, the the program is actually PEEKing and POKEing the coincident device registers. Routines written by Peter J. Mckinney -- Michigan State University Adapted from routines by Dr. C. Radcliffe -- Michigan State University Last Modification: July 10, 1986 Define word array residing 72 c over device registers by c declaring common to KCOM COMMON IKCOM/ IREG('70000’0:'77776'0) C POKE by placing value in array 0 (divide ICSR by 2 since we are C using a word array and C addresses increment by bytes) IREG(ICSR/2) = IVAL RETURN END c ....... __ ............................................. I FUNCTION IPEEK(ICSR) C ....................................................................... C This function returns the contents of a device register at ICSR c ....................................................................... C Define word array residing C over device registers by C declaring common to KCOM COMMON /KCOM/ IREG('7000O'O:'77776'O) C PEEK by reading value from C array. (divide ICSR by 2 since C we are using a word array and C addresses increment by bytes) IPEEK a IREG(ICSR/2) RETURN END c********************************************************************** C SUBROUTINE TNKFRQ(FREQ,MOTOR) C C********************************************************************** TNKFRQ sets the speed of the DC motor that opperates the linear tank. The speed is entered to the computer as a frequency then converted to a voltage using channel 0 of the DEC AXVll-C D/A converter. ********************************************************************** Program written by: David J. Fanson Michigan State University (ICICICICICICICDCICOCD 73 C C Last modification: August 14, 1987 C C********************************************************************** C C Variable List: C C FREQ Frequency C MOTOR Motor number C 1 . Linear Tank C 2 . Washing Machine C RPM Desired RPM of drive system C VOLTS Voltage produced by D/A board C C********************************************************************** C RPM 3 FREQ*60. VOLTS a 0.0 1 C Determine if desired speed is 1- c within the motor's range 1 C IF (MOTOR.EQ.1) THEN IF (RPM.LE.190.0.AND.RPM.GE.0.0) THEN C C Calculate output voltage using C constants found from a 3rd C order polynomial fit of a set C of calibration points CALL CAL1(VOLTS,RPM) ELSE TYPE *,' ’ TYPE *,'Speed out of range 1' ENDIF ELSE IF (MOTOR.EQ.2) THEN C Determine if desired speed is C within the motor's range C IF (RPM.LE.360.0.AND.RPM.GE.0.0) THEN C C Calculate output voltage using C constants found from a 3rd order C polynomial fit of a set of C calibration points C RPM . RPM*8.056 RPM . RPM*12.66 CALL CAL2(VOLTS,RPM) ELSE TYPE * TYPE *,'Speed out of range !' ENDIF ELSE TYPE * 74 TYPE *,'Invalid Motor ' ENDIF C Set channel 0 of the BIA board c to disired voltage CALL DTOA(0,VOLTS) RETURN END Cir********************************************************************* C SUBROUTINE CAL1(Y,X) C c********************************************************************** C 000000000000000000 C Subroutine CALl is 3rd order polynomial fit to a set of C calibration test points C C********************************************************************** Written by: David J. Fanson Last Modification: September 1986 ********************************************************************** Variables: C1-C4 Constants for fit X Independent variable Y Dependent variable ********************************************************************** Set constant to values found in polynomial fit IF (X.EQ.0.0) THEN Y'an ELSE Cl . 0.168121677E-07 CZ --0.279793403E-05 C3 . 0.482184552E-01 C4 . 0.195198685E+00 C Solve for the value of the C dependent varible using the C current value of the C independent variable Y . X**3*C1 + x**2*cz + X*C3 + C4 ENDIF RETURN END 75 C C********************************************************************** C SUBROUTINE CAL2(Y,X) ******i*************************************************************** Subroutine CAL2 is 3rd order polynomial fit to a set of calibration test points Written by: David J. Fanson Last Modification: April 1987 ********************************************************************** Variables: C1-C4 Constants for fit X Independent variable Y Dependent variable ********************************************************************** Set constant to values found in polynomial fit C1C1C1C5C1CIC1C)C1CICICICICDCICOCDCD(563(16363 IF (X.EQ.0.0) THEN Y a 0.0 ELSE C1 3 000 C2 - 0.243787728E-07 C3 - 0.455234572E—02 C4 - 0.198532641E+00 C Solve for the value of the C dependent varible using the C current value of the C independent variable C Y - X**3*C1 + X**2*C2 + X*C3 + C4 ENDIF RETURN END c ...................................................................... C . SUBROUTINE DTOA(CHNL,VOLT) C . c ...................................................................... 000 0000000000000000000000 000 0000 00 76 DTOA is a generic subroutine that sets channels 0 or 1 of the DEC AXVll-C D/A convertor to a selected voltage. Written by: David J. Fanson Michigan State University Last Modification: August 14, 1987 Variable List: ADRESS Octal address of register for D/A channels CHNL D/A channel to be altered VALUE Integer value to be POKEd into the address VOLT Desired voltage of output INTEGER ADRESS,CHNL,VALUE Check for errors in input and exit if found IF (CHNL.NE.0.AND.CHNL.NE.1) THEN TYPE *,'******* Illegal channel *******' GOTO 15 ELSEIF (VOLT.GT.10.0.AND.VOLT.LT.0.0) THEN TYPE *,'******* Voltage out of bounds *******' GOTO 15 ELSE Set adress and POKE value from input for channels 0 or 1 IF (CHNL.EQ.O) THEN Address of channel 0 ADRESS - "170404 ELSE Address of channel 1 ADRESS . "170406 ENDIF Calculate the POKE value from the desired voltage using a linear calibration VALUE . VOLT*409.6+O.5 Make sure VALUE is within range IF (VALUE.GT.4095) VALUE a 4095 IF (VALUE.LT.0) VALUE . 0 POKE the calculated value into the 77 C selected address CALL IPOKE(ADRESS,VALUE) ENDIF 15 CONTINUE RETURN END CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC C) SUBROUTINE FSAM(VOLT,VLTAVE,SCHAN,NCMAN,RATE,NSAMPL,MODE) This subroutine is a basic sampling routine. The starting channel, the number of channels to be sampled, the sampling rate, and the number of samples to be taken are passed in. An array of integer values in the range of 0-4095 corresponding to 0.0 — 10.0 volts DC. _‘u Arguments: 1‘11. Input- SCHAN Starting channel of A/D board NCHAN Number of channels sampled RATE Sampling rate NSAMPL Number of samples taken MODE Sampling starting mode (0 . instant start,1 . wait) Output- VOLT Array of voltages from sampling VLTAVE Average of voltages ********************************************************************** Linking procedure: Ribbit S TKB TKB) filename - filename, FSAM, LNTUTL TKB) @[1,54]KCOM ********************************************************************** Written by: David J. Fanson Michigan State University Last modified on August 26, 1987 ********4************************************************************** Variable list: ADSWP Routine to do A/D sweep BUF Buffer array that hold sampled data (ICICICICICICICICICICICI(3636363C3C§C§C§C§C1C1C5C1C1C1C3C1C3(ICICDCDCDCE(3(1(5(3(3(§(1(1 78 C IND Error code C C*********************************************************************** C C ----- VARIABLE DECLARATIONS ----------------------------------------- C INTEGER BUF(256),SCHAN REAL VOLT(NSAMPL),RATE CHARACTER ANS C ---------------------- Start A/D sweep ----------------------------- SUH.OO IF (MODE.EO.1) THEN 20 TYPE 100 100 FORMAT(' Hit "S” to start sampling : '3) CALL WTCHAR(ANS) IF(ANS.NE.'S'.AND.ANS.NE.'s') GOTO 20 CALL BEL(1) Typg *,r************** SAMPLING ************r ELSE IF (MODE.NE.0) THEN TYPE *,'***** Illegal mode ******' GOTO 50 END IF CALL ADSVP(BUF,RATE,SCHAN,NCHAN,NSAMPL,IND) IF (IND.NE.0) THEN TYPE 110,IND 110 FORMAT(' ***** ERROR ',Il,’ IN SAMPLING ROUTINE *****') CALL BEL(Z) ELSE DO 40 Isl,NSAMPL CALL ADCAL(VOLT(I),BUF(I)) SUM . VOLT(I)+SUM 40 CONTINUE IF (SUM.E0.0.) THEN VLTAVE - 0. ELSE VLTAVE a SUM/NSAMPL END IF END IF SO CONTINUE RETURN END 0000000 0000000000000000000000000000000000000 79 ADCAL is a subroutine to convert integer value into a voltage Written by: David J. Fanson SUBROUTINE ADCAL(Y,X) INTEGER X A : 1./409.6 Y8X*A RETURN END ADSWP is FORTRAN subroutine that uses the A/D converter (AXVll-C) and Programable Real-Time Clock (KWVll-C) to collect a buffer of integer sample data. The subroutine is written to allow the first sample at each time step to be started by the Programable Clock and sub subsequent samples to be taken individually by the A/D converter. Written by: David J. Fanson Michigan State University Last Modification: February 22, 1987 Variable List: BUF Array of integers from sampling DBR ADDRESS of Data Buffer Register CSR ADDRESS of Control/Status Register ERR Error flag FREQ Desired sampling frequency N Number of samples per channel NCHAN Number of channels to sample PERIOD Sampling period 00000000000 00 000 '0000 10 20 30 80 SCHAN Starting channel TOTAL Total number of samples VALUE POKE value for CSR Declare variables INTEGER CSR,DBR,VALUE,SCHAN,NCHAN,N,ERR,SAMPLE,TOTAL,SET INTEGER BUF(256) K a 0 ERR . 0 TOTAL . N*NCHAN Set addresses of registers CSR . "170400 DBR . "170402 E: 3?? Find value of starting channel and move it over 8 places SET . (SCHAN*2**8).0R."40 Enter starting channel CALL IPOKE(CSR,SET) Start clock CALL CSTART(FREQ) Top of sampling loop CONTINUE IF (ERR.EQ.O.AND.K.LE.TOTAL) THEN K=K+1 . CONTINUE Check to see if sample is complete VALUE . IPEEK(CSR) IF ((VALUE.AND."200).EQ."200) THEN If complete then set BUF equal to the value of the DBR BUF(K) n IPEEK(DBR) Gather rest of channels using function SAMPLE DO 30 I-SCHAN+1,SCHAN+NCHAN-1 KIK+1 BUF(K) . SAMPLE(I) CONTINUE Set A/D to starting channel CALL IPOKE(CSR,SET) Wait for sampling intervule CALL CWAIT(ERR) ELSE GOTO 20 END IF 00000000000000000000000000000000000000 81 GOTO 10 END IF Shut off clock CALL CSTOP RETURN CSTART and CSTOP are FORTRAN subroutines that start and stop the KWV11-C Programmable Real-Time Clock. Subroutine CSTART sets up the Clock to generate a DONE bit at a user specified rate. if“ “i—‘I- - ,. __ Written By : David J. Fanson Michigan State University Last Modification : January 29, 1987 Variable List : BPR ADDRESS of Buffer/Preset Register CSR ADDRESS of Control/Status Register FREQ Overflow Frequncy RATE Clock Rate SET Clock Rate Bit Settings STORE Real Clock Count VALUE Clock Count Declare variables INTEGER BPR,CSR,SET,VALUE REAL FREQ,RATE,STORE Set constants CSR . "170420 BPR . "170422 Values for maximium Clock Rate 00 0000 82 RATE 3 IOB+O6 SET a "12 Calculate Clock Count using maximium Clock Rate STORE . RATE/FREQ 10 CONTINUE IF Clock Count value is greater than 77777 OCTAL THAN lower Clock Rate until it is IF (STORE.GT."77777) THEN STORE . STORE/10.0 RATE . RATE/10.0 SET - SET+”10 GOTO 10 END IF Calculate Integer Clock Count VALUE s -1*(STORE+O.5) Calculate actual Overflow frequency FREQ . -1.*RATE/VALUE POKE the Clock Count into BPR CALL IPOKE(BPR,VALUE) POKE Clock Setting into CSR CALL IPOKE(CSR,SET) Flip the GO Bit to start the Clock CALL IPOKE(CSR,(SET.OR."1)) RETURN CSTOP is a subroutine to stop the KWVll-C Programmable Real-Time Clock. INTEGER CSR CSR - ”170420 Set CSR to zero CALL IPOKE(CSR,0) RETURN END 000000000000000000000000000000000 00 000 000 83 CWAIT is a FORTRAN subroutine that waits for the DONE bit on the KWVll-C Programmable Real-Time Clock to be set. Written By : David J. Fanson Michigan State University Last Modification : January 29, 1987 Variable List : BIT1 Check for bit 1 BIT7 Check for bit 7 BIT12 Check for bit 12 CSR ADDRESS of Control/Status Register ERR Error Flag VALUE Test value Declare variables INTEGER BIT1,BIT7,BIT12,CSR,ERR,VALUE Set constants ERRIIO CSR . "170420 , PEEK value of CSR VALUE . IPEEK(CSR) Find status of Bits 1 and 12 BIT1 . VALUE.AND."1 BIT12 - VALUE.AND.”10000 IF Bit 1 is not set Error Flag equals 1 ' (Clock not started) IF (BIT1.NE."1) THEN ERR . 1 IF Bit 12 is set Error Flag equals 2 (Flag Overrun) ELSE IF (BIT12.EQ."1000O) THEN ERR . 2 000 000 00000000000000000000000000000 84 ELSE 10 CONTINUE Find status of Bit 7 BIT7 s IPEEK(CSR).AND."200 IF Bit 7 is set IF (BIT7.EQ."200) THEN THEN Clear Bit 7 and exit VALUE 3 IPEEK(CSR).AND."177577 CALL IPOKE(CSR,VALUE) ELSE Continue Loop until Bit 7 is set ELSE GOTO 10 END IF END IF RETURN END SAMPLE is a FORTRAN function that uses the A/D converter (AXVll-C) to collect a single sample from a specified A/D channel. Written By : David J. Fanson Michigan State University Last Modification : February 6, 1987 Variable List : DBR ADDRESS of Data Buffer Register CSR ADDRESS of Control/Status Register CHNL A/D channel to be sampled VALUE POKE value for CSR 85 c Declare variables INTEGER CHNL,CSR,DBR,VALUE c Set addresses of registers CSR . "170400 DBR . "170402 C Find value of channel and C move it over 8 places VALUE : CHNL.AND."17 VALUE . VALUE*2**8 c POKE channel into CSR CALL IPOKE(CSR,VALUE) C Start Sample VALUE . VALUE.OR."1 CALL IPOKE(CSR,VALUE) 10 CONTINUE Check to see if sample is complete 00 VALUE 2 IPEEK(CSR) IF ((VALUE.AND.”200).EQ."200) THEN If complete then set SAMPLE equal to the value of the DBR 0000 SAMPLE - IPEEK(DBR) ELSE GOTO 10 END IF C Clear the CSR CALL IPOKE(CSR,0) RETURN END c ...................................................................... SUBROUTINE FIT(N,X,Y,A,R) c ........... _ _ ......... -_ ................. REAL X(N),Y(N),A(4) REAL E,DX,R INTEGER I,N,M,MM,MAX E - 2.0 - MAX - 10 DX . X(N)-X(N-l) CALL 0FFSET(N,X,Y,A) C TYPE *,N,A C DO 10 3300595009001 A2 - 0.3 86 A(2) - A2 I . 0 30 CONTINUE 1814*]. MM a M M - (-LOG(E/100)/(A(2)*DX))+0.5 IF (M.GT.N) THEN M a (-LOG(E/100)/(A2*DX))+0.5 MM a M ENDIF C TYPE *,M,X(M) CALL LINREG(M,N,X,Y,A,R) CALL NLREG(N,X,Y,A,R) C TYPE *,R,A IF (M.NE.MM.AND.I.LT.MAX) GOTO 30 C TYPE *,E,R,A(l),A(2),A(3) 10 CONTINUE RETURN END c ...................................................................... SUBROUTINE 0FFSET(N,X,Y,A) c ........................................................ REAL X(N),Y(N),A(4) REAL MAXSLP,SLP INTEGER I,N,MAX HAXSLP . 000 0x - X(N)-X(N-1) C Find maximum slope of C data and it's location SLP - (Y(I+l)-Y(I))/DX IF (SLP.GT.MAXSLP) THEN MAX 2 I MAXSLP - SLP 87 END IF 10 CONTINUE C Calculate offset A(4) - X(MAX)—Y(MAX)/MAXSLP C Correct data with C respect to offset 00 20 I-MAX,N X(I-MAX+2) a X(I)-A(4) Y(I-MAX+2) a Y(I) 20 CONTINUE Update number of data points because of data correction 000 N a N-MAX+2 RETURN END c__-------------------_-s .............................................. SUBROUTINE LINREG(M,N,X,Y,A,R) c ...................................................................... REAL X(N),Y(N),A(4) REAL SUMX,SUMY,SUMX2,SUMXY,SR,ST,R INTEGER I,M,N,NN SUMX a 0.0 SUMY a 0.0 SUMXZ n 0.0 SUMXY - 0.0 DO 10 I-M,N SUMX - SUMX+X(I) SUMY - SUMY+Y(I) SUMXZ - SUMX2+X(I)**2 SUMXY - SUMXY+X(I)*Y(I) 10 CONTINUE NN . N-M+1 A(3) - (NN*SUMXY-SUMX*SUMY)/(NN*SUMX2-SUMX**2) A(l) - SUMY/NN-A(3)*SUMX/NN SR I 000 ST - 0.0 00 20 I-M,N SR . SR+(Y(I)-A(1)-A(3)*X(I))**2 ST - ST+(Y(I)-SUMY/NN)**2 20 CONTINUE c ..... c _____ 00000 10 20 00000 88 R a SQRT((ST-SR)/ST) RETURN END SUBROUTINE NLREG(M,XI,YI,A,R) REAL XI(M),YI(M),A(4) REAL D,E,F,Z,SUMZZ,SUMZD,SUMY,DA,YY,ST,SR,R INTEGER I,M,MAX F(I) s A(1)*(1-EXP(-A(2)*XI(I)))+A(3)*XI(I) E a 0.01 MAX . 100 I . 0 Perform nonlinear regression using Gauss-Newton Method to solve for A(2) in the function F CONTINUE I I 1+1 SUMZZ - 0.0 D0 20 I'l," ' Z - A(l)*XI(I)*EXP(-A(2)*XI(I)) D - YI(I)-F(I) SUMZZ - SUM22+Z**2 SUMZD - SUHZD+Z*D CONTINUE DA 8 SUMZD/SUMZZ A(2) - A(2)+DA Stop iterating after DA becomes less than E percent of A(2) or MAX iterations have been completed IF (ABS(DA*100.0/A(2)).GE.E.AND.I.LE.MAX) GOTO 10 Calculate correlation 0000000 00000000000 30 40 89 coefficient R SUMY I 000 DO 30 I21," SUMY - SUNY+YI(I) CONTINUE YY - SUMY/M ST - 0.0 SR 8 000 D0 60 Isl," ST 2 ST+(YI(I)-YY)**2 SR 2 SR+(YI(I)-F(I))**2 CONTINUE R . SQRT(abs(1-SR/ST)) If subroutine stoped because of reaching maximum number of iterations, create a value for R that is impossible and can be used as a flag. IF (I.EQ.MAX) R 3 1.111111 RETURN END Vritten by: David J. Fanson Michigan State University Last Modification: August 24, 1987 INTEGER DVELL,NUH REAL TOTAL,FREO,LENGTB CHARACTER ANS CHARACTER*10 MOTION CHARACTER*4O TEXT,LOAD,BLADE,OPER,TITLE TITLE . ’ LINT SAMPLING AND ANALYSIS ' 10 210 20 120 130 40 50 140 60 150 90 CONTINUE TYPE * TYPE *,'Enter Run Discription for Data File ((40 char) : ACCEPT 210,TEXT FORMAT(A40) TYPE * TYPE *,'Enter Name of Test Operator ((40 char) : ' ACCEPT 210,0PER CONTINUE TYPE * WRITE(5,120) TEXT,OPER FORMAT(' ;',/, ' ; Run Description - ’,AAO,/, ' 3 Data Collected By - ',AAO) TYPE 130 FORMAT(l' Acceptable (Y/N) ? '3) CALL VTCHAR(ANS) IP (ANS.EO.'N'.OR.ANS.EO.'n') GOTO 10 IP (ANS.NE.'Y'.AND.ANS.NE.'y') GOTO 20 VRITE(1,120) TEXT,OPER Enter sampling conditions CONTINUE TYPE * TYPE *,'Enter Load Discription ((40 char) : ' ACCEPT 210,LOAD TYPE *,'Enter Agitator Discription (<40 char) : ' ACCEPT 210,8LADE CONTINUE TYPE * VRITE(S,140) LOAD,BLADE FORMAT(' ;',/, ' Cloth Load - ',A40,/, ' Agitator - ',AAO) TYPE 130 CALL WTCHAR(ANS) IP (ANS.EO.'N'.OR.ANS.EO.'n') GOTO 40 IF (ANS.NE.'Y'.AND.ANS.NE.'y') GOTO 50 VRITE(1 , 140) LOAD, BLADE Set tank speed CONTINUE TYPE * _ TYPE 150 FORMAT(' Enter Tank Frequency (<3 Hz) : 'S) READ(5,*) PREO Check speed IF (FREO.GT.3.0) THEN TYPE * TYPE *,’****** FREQUENCY TOO HIGH!! ******' CALL BEL(3) GOTO 60 91 END IF TYPE * TYPE 160 160 FORMAT(' Enter Stroke Length (degrees) : 'S) READ(5,*) LENGTH C LENGTH . 10.0 MOTION . 'Symmetric ' C TYPE 180 C 180 FORMAT(I' Enter D/A sampling dwell time (sec) : 'S) C READ(5,*) DUELL DVELL . 12 TYPE 190 190 FORMAT(/' Enter total time to Sample (min) : 'S) READ(5,*) TOTAL NUM a (TOTAL*60./DVELL)+1.5 TOTAL 3 (NUM-l)*DVELL/60. 9O CONTINUE TYPE * WRITE(5,170) DVELL, FREQ, NUM, LENGTH, TOTAL, MOTION 170 FORMAT ' 3',/, + ' ; Dvell Time - ',14, ' sec ' + 'Oscillation Rate - ' ,FS. 2, ' Hz',/, + ' ; Samples - ’,Ié, ' + 'Stroke Length - ',F5.1,' degrees', /, + ' ; Total Time - ' ,PA. 1, ' min ', + 'Agitator Motion - ',AlO) TYPE 130 CALL WTCHAR(ANS) IF (ANS.E0.'N'.OR.ANS.EQ.'n') GOTO 60 IF (ANS.NE.'Y'.AND.ANS.NE.’y’) GOTO 90 URITE(1,170) DVELL,FREO,NUM,LENGTH,TOTAL,MOTION CALL CSCRN CALL SCROLL(19,24) CALL VTITLE(IU,TITLE) WRITE(5,120) TEXT,OPER WRITE(5,140) LOAD,BLADE WRITE(5,170) DVELL,FREO,NUM,LENGTH,TOTAL,MOTION CALL LSCRL(5,8) RETURN END c ............... __ ....................................... C SUBROUTINE VTITLE(IU,TITLE) e_-. 'n. _ - 000000000000000 0 100 110 120 130 92 Written by: David J. Fanson Michigan State University Last Modification: July 16, 1987 -—-----------------_-------------—-----------—-----------------—------ ---- local other variables --------- CHARACTER*40 TITLE INTEGER IU BYTE ESC ESC I "033 Vrite title in large print VRITE(S,100) TITLE FORMAT('+',A40) VRITE(5,110) ESC FORMAT('+',A1,' 3') WRITE(5,120) TITLE FORMAT(' ',AAO) WRITE(5,130) ESC FORMAT(’+',A1,' 4') RETURN END CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC 0 00000000000000 SUBROUTINE OPILE(PNAME,TIX,TIY) This subroutine setsup data files to put the collected data into, and supplies pdl file with axies and title information. Last modified on 3/10/86 by Dan Budny BLANK Used to setup file names CB Used to check for upper case letters in FNAME CM Used to check for upper case letters in PNAME DAT Date FNAME Name of file PDLNAM Name of PDL file POS Used to check length of file name TIM Time of day CHARACTER*15 FNAME,PDLNAM,BLANK 93 CHARACTER*40 TIX,TIY CHARACTER DAT(9),TIM(8) CHARACTER*1 CH INTEGER POS BYTE CB EOUIVALENCE (CB,CH) C --------------------------- Put DAT after filename ---------------- 101 TYPE 107 107 FORMAT(lX,'Enter a filename for the data : ',S) READ(5,'(A)',ERR-101) FNAME BLANK . ’ ' IPOS . INDEX(FNAME,'.') ILEN . INDEX(FNAME,' ') - 1 IMAstEN(FNAME) IF (ILEN .E0. 0) GOTO 101 IF (IPOS .E0. 0) THEN IF (ILEN .GE. (IMAX-4)) THEN FNAME((IMAX-3):IMAX) . '.DAT' ELSE FNAME(ILEN+1:ILEN+4) . '.DAT' FNAME(ILEN+5:IMAX) . BLANK(ILEN+S:IMAX) ENDIF ELSE IF (IPOS .GE. (IMAX-3)) THEN FNAME((IMAX-3):IMAX) . '.DAT' ELSE FNAME(IPOS:IPOS+3) . '.DAT' FNAME(IPOS+6:IMAX) . BLANK(IPOS+4:IMAX) ENDIF ENDIF C ------------- Make sure all characters in FNAME are UPPER CASE ------ ILEN:INDEX(FNAME,’.')-1 DO 110 I-1,ILEN CB-PNAME(I:I) IF ((CB .GE. 97) .AND. (CB .LE. 122)) THEN CB-CB .AND. 95 . ‘.u l' ‘f ENDIF FNAHE(I:I)ICH 110 CONTINUE TYPE* TYPE*,'Filename- ',FNAME C ---------- Generate a PDL file with the same name as the data file -- POS-INDEX(FNAME,’.') PDLNAM(1:POS)=FNAME(1:POS) PDLNAM(POS+1:POS+3)-’PDL' OPEN(UNIT-2,NAME-PDLNAM,FORM-'FORMATTED',TYPE-‘NEV’) URITE(2,220) FNAME 220 FORMAT(lX,'; Automatic PDL file for ',A15) C ----- Setup axes and title information C Get time and date CALL DATE(DAT) CALL TIME(TIM) URITE(2,*) 'DTO’ 94 VRITE(2,*) 'YAO.4,3.S,O.5' URITE(2,*) ’XAO.5,5.0,0.5’ VRITE(2,*) 'CPT10.12,0.125,0.0S,0.0S' VRITE(2,*) ’CPT2.12,.125,.OS,.05' VRITE(2,*) 'CPTL.1,.l,.OS,.Ol' VRITE(2,*) ’CPAT.1,.1,.OS,.OS' VRITE(2,*) 'FMXL(F7.0)' VRITE(2,*) 'FMYL(F8.1)’ VRITE(2,*) 'FMXU(I3)' VRITE(2,*) 'NA2,2,2,2' VRITE(2,*) 'CNT12,0,0' VRITE(2,*) 'CNT22,0,0' VRITE(2,*) 'CNAT2,0,0' URITE(2,*) 'CNTL2,0,0' 222 VRITE(2,112) TIX 112 FORMAT(lX,'TIXL',A40) VRITE(2,113) TIY 113 FORMAT(lX,'TIYL',A40) VRITE(2,225) FNAME,(DAT(I),I-1,9) 225 FORMAT(lX,'TIT1 ',A15,2X,9A1) C C Open the data file and note the date and time C OPEN(UNIT.1,NAME:FNAME,FORMs'FORMATTED',TYPEa'NEV') VRITE(1,115) FNAME,(DAT(I),I-1,9),(TIM(J),J-1,B) 115 FORMAT (' ;'/, + ' ; Data File - ',AlS,/, ' Test Date - ',9A1,/, ; Test Time - ',8A1) ... ‘0 I RETURN END SUBROUTINE CPILE(FILEN,XMAX,J) 'C SUBROUTINE CFILE(FILEN,XMAX,YMAX,J) C C ------------------------- Create plotting files ------------------------ C . CHARACTER FILEN*15 VRITE(2,130) FILEN ! set F phase in MULPLT 130 FORMAT(IX,'FN',A15) VRITE(2,134) J ! J is data tag 134 FORMAT(lX,'TGC',Il) VRITE(2,*) 'FT1,150' VRITE(2,135) 13S FORMAT(lX,'SS') J8J+1 VRITE(2,140) FILEN ! set F phase in MULPLT 140 FORMAT(IX,'FN',A15) VRITE(2,144) J ! J is data tag 144 FORMAT(IX,'TGC',Il) VRITE(2,*) 'FT2,1SO' VRITE(2,145) 145 FORMAT(lX,'SS') C ------------------ Set pdl information ------------------------------ 153 155 C155 159 160 165 95 VRITE(2,1S3) FORMAT(lX,’GP') URITE(2,155) FORMAT(lX,'MMY10.0,150.') FORMAT(lX,'MMY10.0,',F7.2) YTIC a VRITE(2,159) FORMAT(lX,'TMYL30.0,5.0,1,0.0') VRITE(2,160)XMAX FORMAT(IX,’MMX10.0,',F7.3) VRITE(2,165) FORMAT(lX,'TMXL5.0,1.0,1,0.0') VRITE(2,*) 'GOl’ CLOSE(UNIT-Z) CLOSE(UNIT-l) RETURN END .....‘0. Set G phase in MULPLT Set Y axis limits Set tick marks on YL Set X axis limits Set tick marks on XL Include the command to plot the graph on the screen Close data files APPENDIX B EXPERIMENTAL DATA 96 97 RUN - 1 Frequency - 2.00 Stroke Length - 100 200- 253‘ {160- ” < E o . g 1 E1201 .5 j ............................................................... 2 so« ............................. c 4 ............. 8 4 ----- 5 . :3 4o« .5 i _l 4 .. 13' o v v i l t v 1' v I f r v r ] v v I v I y 1 , fi' t ‘fi 0 5 1° 15 20 25 so Time (minutes) RUN - 2 Frequency - 2.25 Stroke Length - 100 2001 ’33 {160‘ 03 E O ..‘én £120: ........................................................ .s : ................................................ :é: ao~ ................. C 1 8 : 8 . 3 40« = 1 ’3 i . r,“ o fi't'r'rfir"rfi1"rfffi'l 'fi 0 5 1o 15 20 25 30 Time (minutes) Lint Concentration (milligrams/liter) Lint Concentration (milligrams/liter) 804 q .3. O l A A A 98 RUN - 3 Frequency - 2.50 Stroke Length - 100 O... ....... ooooooooooo ...... ............. .oo .............. O ...... .....I... ''''' ....- I .0 .0 o-o0 ...... ... O O 1 L A J .b o L I l V V I U I V 5 1O ..13..:.20. .25. ..30 Time (minutes) RUN — 4 Frequency - 2.75 Stroke Length - 100 It 1 V’ 1' 4‘ U U l U V 47 I If V U I I ‘ I I T47 AVTA’ If 4'? F" 1 Y I 5 10 15 20 25 30 Time (minutes) Lint Concentration (milligrams/liter) Lint Concentration (milligrams/liter) .p. O l 99 RUN - 5 Frequency - 2.75 Stroke Length — 100 000 ........... ........ .aa' a... ..................... a. ...... .......... .a O aaaaaaa ca 0 A ‘ I T r r l I I I T I I I I I I T r I ‘I' I T I T I I I I If‘l 10 15 20 25 30 Time (minutes) RUN - 6 Frequency - 2.50 Stroke Length - 100 ..- .0... 0.00....- l O . 0 o .Oaggo..o.' . .00....IOOIOI . ..ooa . ......O .090. o. ....- a0... ..acol' 0' .00. .Ol .... .... ...a’ I I I I I T I I j ' I I I I r I I r I r T T I I j 5 1o 15 20 25 30 Time (minutes) 4o- Lint Concentration (milligrams/liter) Lint Concentration (milligrams/liter) IOO RUN - 7 Frequency - 2.25 Stroke Length - 100 I... ..................... ..... .......... ........... a a ........... ......... no e'° ......... a C 0" I ea. ........ r I I I T I I i f I I I ' I III I I T# I If II‘I ' 5 ' 1o 15 20 25 so Time (minutes) d RUN - 8 Frequency - 2.00 Stroke Length — 100 ..0' .II.."" .........I0.0IAOOO'.. ......I.. '0'... I... . a. ...... ...... '0'. 0...... .0. .00. I . ... at. 0' .0 O I I I I I 1 T I I 1 I I r I I l I I I I I ' 5 ' ' ' ' 1o 15 20 25 so Time (minutes) 101 RUN - 9 Frequency - 1.50 Stroke Length - 140 200- E 1 $150: E O 1 g. . 31201 c 1 ............................ .9. 1 .......................................... ‘é ao« .............................. E ‘ ............. 8 4 .... 5 1 3 404 .5 ‘ A a 0 . V r I I I v ff v I I r u 1 I r v r v T I r , t T ‘ t ‘ ‘_' Time (minutes) RUN - 10 Frequency - 1.75 Stroke Length - 140 200- :5 1 $160: E 4 O E 1 3120-: H c 1 ............................................ .2 ‘ ........................... E 801 .......................... *5 r ............... 3 ..... 5 r f 404 c: t .3 1 ofifirg 0 5 1° 15 20 25 so Time (minutes) 102 RUN - 11 Frequency - 2.00 Stroke Length - 140 200- ’5‘ 1 “$150: E O .5» 1 ........................... E120: ................................................. .5 j ........................... g 80" ..... c 4 8 1 S r j 4o~ c 4 .3 4.. o'. .T:.fir,evs.fir:::,.:r.,...ei O 5 1O 15 20 25 30 Time (minutes) RUN - 12 Frequency - 2.25 Stroke Length - 140 200- A 1 E 1 {160-1 '5 ‘ E d O 5 1 ........................................................ E1201 ...................................... c ‘ ............. .2 < jg ao< C 1 8 1 5 1 3 40« c 4 '3 1 o tifvrrfw.rrjt..:,r51....rvj O 5 10 15 20 25 30 Time (minutes) Lint Concentration (milligrams/liter) Lint Concentration (milligrams/liter) 103 RUN - 13 Frequency — 2.25 Stroke Length - 140 oooooooooooo ...... ..... ....... no. ........... .. a... .' ..ee nee a". .00. .0 e“ .e .e 80- 4o~ I r I I 1 IT I I I I I I I I j I I I I I I T I T l o ' ' 5 1o 15 20 25 so Time (minutes) RUN - 14 Frequency - 2.00 Stroke Length - 140 ‘ a: 0 fl N O D O A A# l A J A I A J J 1 .p O 1 A A o I ITIFTI IT'TTfi—I I I TITI’I ITITTITTII 0 5 10 15 20 25 30 Time (minutes) Lint Concentration (milligrams/liter) Lint Concentration (milligrams/liter) 200 - A A ._A a, O I —L N O l j, on O . n A A A A A .h C l A L A A A O a O .p. O A l A. l A A A l A A O 104 RUN - 15 Frequency - 1.75 Stroke Length - 140 .......... .................. """"""""""""""" ...... I... .............. ........... .- cccccc o" e o 0" .o-' o 10' .- 0 IO. .0. e .- C ...... I 1’ If I ‘I ‘r’ ‘r" I’ I I’ If I I I 10 15 20 ' '25 ' so Time (minutes) q til-4 4 RUN - 16 Frequency - 1.50 Stroke Length - 140 O .0. u I. a. I... .0 ooeeoefi.‘.""'... ecooo'e¢"uo‘....... 3 ......II.. .0. '00. none. ...0‘... .0... e' 00" I 000' e-“ .0 .0 .e .0. Tf'rTTfi'rtT‘I '1'r5""' 5 1o {3 20 25 ' ' '30 Time (minutes) Lint Concentration (milligrams/liter) A A A ._n N O 1 -F on O O L A l A L A J A J O Lint Concentration (milligrams/liter) 105 RUN - 17 Frequency - 1.00 Stroke Length - 180 .. o O "C 0' 'I I ' o It .IeoUOIIODOQO.'OIenloO' C l'loe ' ° " I I no... II 0 ' e 0‘ II can't-O o .0. .. .... ....IIO IOOOuoOO .0000 ...-D I...OC .000. 1.. O... 5 ' t r135 "5 '15 ' ' ' '20 ' 25 r ' 50 Time (minutes) RUN - 18 Frequency - 1.25 Stroke Length - 180 r l I I I II I I I I I I ' 5 ' ' ' 1o 15 20 ' ' 25 ' ' 50 Time (minutes) Lint Concentration (milligrams/liter) A A A d on N C) O A l A A A A 4. O 1 A A Lint Concentration (milligrams/liter) A A O A O RUN - (JP-i 106 19 Frequency - 1.50 Stroke Length - 180 0". eeeeeeeeeeeee ............... 00' ........ .000 00" 00" ooooo .......... 0.. 00' .00 .00 0“ ..... .... a ..... .0 .0 .C I II” If 414' I I I IIII' I ‘I’ I I I I I I' ’II’ I I I I' 'I_ __1 10 15 20 25 30 Time (minutes) RUN - 20 Frequency - 1.50 Stroke Length - 180 I II ‘I III I I I I I I 5 .............. 0 ...................... 0 0. 0' 000° ooooooooo ........... 0 .00 ..... 000000 O I 0 I .00! .0 00000 .0 0 .000 I I' I I I I I I I I ' II I II ‘I '7 I I I I 1 10 15 20 25 30 Time (minutes) Lint Concentration (milligrams/liter) Lint Concentration (milligrams/liter) 4o~ 107 RUN - 21 Frequency - 1.25 Stroke Length - 180 00- ooooooo .............. eeeeee ...... 00 oooooooooooooooo 00000 .......... ........ cccccc 0" .0 00000000 .0 00" .- 0 0 .0 .00 0 a 80‘ I I III I II I I r I II I I I III I I I I 1 10 15 20 25 30 Time (minutes) 01-4 RUN - 22 Frequency - 1.00 Stroke Length - 180 .............. 00000000 00000000 ..... 00000000000 00000000000 000000000000 0000 ......... ..00 0". eeeee 00. ..00 I... 0000000 .- 0 0 0 0 00' I .0 Time (minutes) 108 RUN - 23 Frequency - 1.75 Stroke Length - 120 200- ’E.‘ 1 § 1504 E . O .53 :8: 120: ..S: l ................... g 80‘ ............................................................ c ‘ .................... l- 8 4 .............. 5 1 ......... 3 40‘ c 1 .3 4 OJ, rjjv tjr r1 -x.1 v.1 r1 Time (minutes) RUN ‘ 24 Frequency - 2.25 Stroke Length - 120 200- E 'i 160 - ” 1 E 1 O E.” . E 120‘: ......................................................... c ‘ .......................... .3 1 ................... g 30. ................. c 1 ...... 8 i 5 1 :3 405 .. C 4 . '2 1 '- 1 . O T I I v I t l I v t fT T T III I v t v v 1 . , t j Time (minutes) 109 RUN - 25 Frequency - 2.00 Stroke Length - 120 2001 3:? {160‘ V3 1 E 1 O .2”? 1 Tim? ............... .5 1 .......................................................................... E 804 ................... .05 + ....... 8 1 ..... g 1 3 4o« .5 ‘ .' _J ‘ u 1,‘ o. .r. .,..11.111fir.111..11fi O 5 10 15 20 25 30 Time (minutes) RUN - 26 Frequency - 2.25 Stroke Length - 120 200- ’g 1 {160-1 03 4 E 1 O .2 1 ...................................................... 3120: ...................................... § ; ....................... :9; oo- = 1 8 1 g 1 fj 401 , c «i . :1 1. o '. 111.111.11fi1H111, 11.,1fr11 O 5 1O 15 20 25 30 Time (minutes) _— 110 RUN - 27 Frequency - 2.00 Stroke Length - 120 2001 ’g 1 i 160:1 E D g. 1 E ””7 .5 : ........................................................................... 212‘ ao- ............................ c 1 ........... 3 1 S 1 if 404 c 1 '1: .- o 9 1 ,fi111r1111,1111,111.,1rfi1, O 5 1O 15 20 25 30 Time (minutes) RUN - 28 Frequency - 1.75 Stroke Length - 120 2001 J :5 1 {1501 I) 1 E 1 D 326.» 1 ,=-_-, 1 £120; .5 3 .......... Lé 80+ ............................................................. c ‘ ......................... o ........... u I ......... g 1 j 40« c 4 :1 4 _. o.'—' fofTI' ‘1‘ 1' r'I' r1 0 5 1O 15 20 25 30 Time (minutes) 111 RUN - 29 Frequency - 1.25 Stroke Length - 160 200- :5 I €160: E 1 O 3' 1 E120: .5 I E 80‘ ............. c ‘ .................................................... 3 1 ..................................... 5 1 ........................ 2 4o- ....... .5 ‘ -J 4 o'1111151111,f11111511,...+.1111fl 0 5 10 15 20 25 30 Time (minutes) RUN - 30 Frequency - 1.50 Stroke Length - 160 200- :5 1 {1601 a 4 E 1 O .169 1 .5 : .................................................................... E 80- ................................ c 1 ............. 8 1 .... 5 1 3 4o: .5 ‘ _J 1 1 o.111.rr11,r.111.111.111.1fi1111 0 5 10 15 20 25 50 Time (minutes) Lint Concentration (milligrams/liter) Lint Concentration (milligrams/liter) 112 RUN - 31 Frequency - 1.75 Stroke Length - 160 200 1 120 ‘ a} o AAAAAAA A A A . A A .3- O 1 A A A I II r I I I I r I I I I TIIII I I I Time (minutes) RUN — 32 Frequency - 2.00 Stroke Length - 160 5 1o 15 20 H 25 Time (minutes) I rj 30 30 Lint Concentration (milligrams/liter) Lint Concentration (milligrams/liter) on O n 113 RUN - 33 Frequency - 2.00 Stroke Length - 160 eeeeee .0 000000000000000000 00000 ..... 0" ' .0 aaaaa ....... ............. .0 D .. .0 0". 000000 I 0". .0 .0 .I II I I I I I I I I I r I T I I I I I I I r I I 5 1o 15 20 25 " so Time (minutes) RUN - 34 Frequency — 1.75 Stroke Length - 160 e 0... .0 .0 I III If I I r I I II T I1 I I T I I I I r I I I I I 5 10 15 20 25 30 Time (minutes) 4- O 1 Lint Concentration (milligrams/liter) 200 1 .—0 03 O l -. m N O O L- A l, A A A 41 A L L 4. o 1 Lint Concentration (milligrams/liter) A L 114 RUN - 35 Frequency - 1.50 Stroke Length - 160 00000 .............. 00000000000 0000000000000 00000000 000000000 00 ..00 ...... ........ .0 .0 000000 0“ .0 .0 .0 00000 .0 .0 0 .0 0 0 .0 .0 0‘. I I I I I I I I I l I I I I I I I I I I I I I I j 5 10 15 20 25 30 Time (minutes) RUN - 36 Frequency - 2.25 Stroke Length - 160 ..... ........ .... ...... ,,,,,, 00000000000 .... 00000 000000 00' ...... 0000 00" 00.. C 0... D .0 0'. .0 0 .0 0"I 0". 00' I I I I I I I l I I I I I II 1 I I I 5 ' ' ' ' 1o '13' 20 25 so Time (minutes) Lint Concentration (milligrams/liter) Lint Concentration (milligrams/liter) 2001 150‘ 120‘ i 80-i 200 , i 1604 1202 115 RUN - 37 Frequency - 1.25 Stroke Length - 160 I Ir’ 41' 1’ I I I ‘I If I I I I I I ‘II I] If I IIII’I I I II I I I 5 10 15 20 25 30 Time (minutes) RUN - 38 Frequency - 2.25 Stroke Length - 160 000 0" ........ 00 .I0‘ 000' ..00 000‘ 0 00000 ......... ....... 00' ...... .000 ...... 00 00' ... 0". 0 ..... 0". 0O. 00000 .0 .0 0'. 0'. 0.. 0" u C '. I 0 Q .0 0 ' I I IV I U I7 I I I l I If I I 1’ II I I I 4‘ 20 25 3O 0" d o —. C” Time (minutes) Lint Concentration (milligrams/liter) Lint Concentration (milligrams/liter) 200 g 160 4 120* 4 200 1 4 160 5 1204 I L O O 1 .0- O l J A L A A 116 RUN - 39 Frequency - 1.00 Stroke Length - 200 .03. 00000000000000000000 0000000000000 ............. 0000000 0000000 0000000 000000 0000000000 0000 000000 00000 000' 000. 0" 00' 0 .0 .0 0 0 0 .0 Time (minutes) RUN - 40 Frequency - 1.25 Stroke Length - 200 I I I I I I I I I T I I I I l I I I I I I I 10 15 20 25 30 Time (minutes) (II-4 117 RUN - 41 Frequency - 1.50 Stroke Length - 200 2005 g 4 {1605 m . E 4 D g. . 5”": ................................................. ,3 ........................................... g 80- ...................... c ‘ ..... 3 r S J j 405 .5 ‘ __. 4 o rt....1,,.5....5.T,..rtrreTTj 0 5 10 15 20 25 30 Time (minutes) RUN - 42 Frequency - 1.75 Stroke Length - 200 200- 3 + ............. €150“ .................... «n i ......... E ................. -o 1 ......... 2' i ................... 3120: .................... c i ....... 5% T g 80+ c «i 8 i S e ,- 3 4o«- c . :1 +. or5....tr....fij,..flff.,..rrTv, 0 5 10 15 20 25 30 Time (minutes) Lint Concentration (milligrams/liter) Lint Concentration (milligrams/liter) 200 - 160 - 120‘ an O A 4 A l .0. O 1 118 RUN - 43 Frequency - 1.75 Stroke Length - 200 ... 0.. ..... 0 00000 0 O ..... I 00 '''''' 0 00° ........ .0 0". 000000 ..... ...... ....... 00" .- .0 0". .0 00" .0 .. .0 0" O .0 .0 0 I I .0 ' 0 D 0 .I 0 0 200 1 d -. m N O: O O O A l, 40 A, A l 204,41, A l A A .5 O A l A I I I I I II t I I II I I I II I I I' I I I III" I I I I 1 5 10 15 2O 25 30 Time (minutes) RUN - 44 Frequency - 1.50 Stroke Length - 200 0 0000000000 ..... 00000000 ............ '. 00000000 .0 .000 0000000000 000 00' .. 00" 00‘ .0 0000000 .0 00’ .0 .. .. 0.. 0" 0 0 0'. .0 O .0 0.. 0 .0 .0 ' I I I I r I I I I ' I I I I I I I I I I I I I I I 5 1O 15 20 25 30 Time (minutes) 119 RUN - 45 Frequency - 1.25 Stroke Length - 200 200 - 150 < 120~ .0 ...... 00000000000000 000000000 000000 00000000000000 00000000000 0'. 0000 ....... ......... 000 00" 00' .0 .0 000000 .I .0 .0 0" I .0 00’ 0 0" .0 Lint Concentration (milligrams/liter) I I I i I I I I I I I I I I I I I I I 10 15 2o 25 ' so Time (minutes) O 01—4 RUN - 46 Frequency - 1.00 Stroke Length - 200 _0 0’ O 1 Lint Concentration (milligrams/liter) I o I I II I I I I I I TII I I I I I I I— ] I I I 0 5 1O 15 20 25 30 Time (minutes) i Lint Concentration (milligrams/liter) Lint Concentration (milligrams/liter) 200 '1 O) O 1 120‘ an O L A A l A A .5. o l L A A I O 120 RUN - 47 Frequency - 1.75 Stroke Length - 180 ...... .0 ......... 0o. ......... O ....... oooooooo nnnnnnn I ...... 0000 .... ----- I ...... O 0000 .e 00' .- '0'. ..... I... 0". .0 .- .0 D .- .I o I .0 o o 40« I' 17 II, I7 l I II I IV I I I I I i I IV I I l I I I' I I I’ I I 1fi4‘_1 5 10 15 20 25 30 Time (minutes) RUN - 48 Frequency - 1.75 Stroke Length - 180 I ...... Q. .000 .I .0. 00° .C 0" ........ .... a... .0. 0" an. .0. .I one. .0 .e ....... e ...... I .0 ..... O 0". ... 9". .I .- 0'. .o .0 .0 0" .. I ' I I I II I I I I I I» I I l I' II I I ' ' 5 ' ' ' 10 '13 20 25 so Time (minutes) ll Lint Concentration (milligrams/liter) Lint Concentration (milligrams/liter) 121 RUN - 49 Frequency - 2.00 Stroke Length - 180 .o l.- I. 1". n'. ........ I 0000 .. ..... I .I .- ....... no 0" o" ...... '0 I .C ....... o. .C C ..... 0" a... ..... .- .0 0" .0 .a .e ... .C .O In C .0 o .. 80‘ A O 1 A A A i I I I I I I I I I ' III I I I I I I I II I I I I I I 15 20 25 30 Time (minutes) :3 0'1 ... o RUN - 50 Frequency - 2.00 Stroke Length - 180 200 - L A ... O: O 1 A J. A A 120‘ ................................ O O ..4 401 .‘ 4 . I I ' I I I I I I I I I I I I I I T I I II I I o ' 5 ' 1o 15 20 25 so Time (minutes) Lint Concentration (milligrams/liter) Lint Concentration (milligrams/liter) 200 - 160 -* 120* 122 RUN - 51 Frequency - 2.50 Stroke Length - 140 ...... C ..... 50" as O 1 4o~ 10 15 20 25 Time (minutes) a... RUN - 52 Frequency - 2.50 Stroke Length - 140 ..... ....... ......... 0 ......... 0 ....... ...... a ...... ...... ..... .- a... 3" 0|. 0'. ,e 0". 0". C 0". .. 0'. e... ,0 I e" 0 0 .0 .0 e If I I I I I I I I I' I II I II' If ‘I' I I I I ‘I I I I I I I I I l I I I I I444 I I' I III, I' I I II I 5 1o 15 20 25 Time (minutes) ‘so Lint Concentration (milligrams/liter) Lint Concentration (milligrams/liter) 40‘ 123 RUN - 53 Frequency - 2.50 Stroke Length — 120 .00 0 .......... 0 00' 0" 0 .......... 0 00' 000. .II' 200 - 160 < 120‘ A an O l 5 1o 15 20 25 Time (minutes) RUN - 54 Frequency — 2.75 Stroke Length - 120 .0 00000 .00 00' 000' ........ 0 0000 00000 .0 .0 .0 0 00000 I... 000000 U .0. 0" 0" ..... I 00000000 C 0". .00. .00' 0 ...... '0 0". .0 .0 I' ‘I I II I I III I I ' If I I47 If ' I47 I I I I I' I I ‘II II] 30 1O 15 20 25 Time (minutes) (II-i ‘I I If I I If I I I I I I I I I I ‘r’ II I '7 I '50 200 - __. O: O 1 120-1 a O I J A A .b O 1.. Lint Concentration (milligrams/liter) A L 200 q -‘I 05 O A, A A 41, A, A ; ._b N O l A A an O l A A A A 404 Lint Concentration (milligrams/liter) 124 RUN - 55 Frequency - 2.75 Stroke Length - 120 0 ......... .0 0000000000 00'. ......... 0'. 0000000 05" ..... O ....... I ..... .3 .- ....... p. ......... O ...... I C ... 0". .0. I 0 l 0". I I al‘ I' If I I [I I’ I I III I I’ I I I I I If I I I III ‘II_II_1 1o '15 20 25 so Time (minutes) (11-1 RUN - 56 Frequency - 2.50 Stroke Length - 120 00. 000000 .00. 0000 000000 00000 O 00.. ...... 000' .00. ...... .0 .......... 0 O 000. 0". 00" eeeee I ..0. .00 00000 05' .0 0". I 0 0 0" .0 .0 .0 0" I IIIIII ’1 .1 l I I I I l I I' I I l I I I I ‘I ' 5 r ' ' '10 15 20 25 so Time (minutes) "1111111111111111117111is