as, MHHWIMIIHWmilling» This is to certify that the thesis entitled FLY-ASH RESISTIVITY - ITS MEASUREMENT AND CORRECT INTERPRETATION presented by Arun V. Someshwar has been accepted towards fulfillment of the requirements for M.S. Chemical Engineering degree in £55,. (,0 (Midfield w) Major professor Date M 0-7 639 I OVERDUE FINES ARE 25¢ PER DAY PER ITEM Return to book drop to remove this checkout from your record. FLY-ASH RESISTIVITY--ITS MEASUREMENT AND CORRECT INTERPRETATION BY Arun V. Someshwar A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemical Engineering 1979 ABSTRACT FLY-ASH RESISTIVITY--ITS MEASUREMENT AND CORRECT INTERPRETATION BY Arun V. Someshwar Existing literature on the problem of high elec- trical resistivity of coal fly-ash is surveyed. A 'Point- to-Plane' resistivity probe is fabricated and the two techniques for the measurement of resistivity with this probe are employed. Experiments are conducted over a couple of months' duration at two power plant sites. An attempt is made to resolve some of the prevailing ambig- uities concerning the appropriateness of the reported critical ash resistivity. The phenomenon of 'incipient' sparking is postulated to help overcome the non- uniqueness of the experimentally observed 'critical' sparkover point. The ’Point-Plane' technique involving the measurement of 'Clean Plate' and 'Dust-laden Plate' voltage-current characteristics is found to be more accurate and representative than the 'Disc-Plane' technique in which no corona current is used during measurement. Finally a concise manual for the operation of the SRI resistivity probe is included. ACKNOWLEDGMENTS The author wishes to express his sincere appre- ciation and thankfulness for the advice and assistance obtained from Mr. P. N. Shukla (the author's co-worker in many of the experiments undertaken) and Dr. Bruce W. Wilkinson (the author's major professor and constant guide). Expressions of gratitude are also extended to Mr. Leo Szafranski and Mr. Don Childs of the Engineering Shop; to the authorities of Power Plant 65 at Michigan State University and the Lansing Board of Water and Light, Lansing, Michigan; and last but not the least, to Dr. M. L. Davis of Civil Engineering. ii TABLE OF CONTENTS LIST OF TABLES O O O O O O O O O O 0 LIST OF FIGURES . . . . . . . . . . . Chapter I. INTRODUCTION . . . . . . . . . II. REVIEW OF THEORY AND PREVIOUS WORK . . Conduction Mechanisms . . . . . Factors Affecting the Measurement of Resistivity . . . . . . . Techniques of Measuring Resistivity III. EXPERIMENTATION . . . . . . . . Discussion of Experimental Technique Recording of Data and Calculation of Resistivity . . . . . . . Tabulation of Results . . . . . IV. SUMMARY OF RESULTS . . . . . . . V. SOURCES OF ERROR AND RECOMMENDATIONS . VI. CONCLUSIONS . . . . . . . . . . REFERENCES 0 O O O O O C O O O O O O APPENDICES O O O O O O O O O O O O O I-A. I-B. I-Co DRAWING OF THE SRI POINT-PLANE RESISTIVITY PROBE AND PHOTOGRAPHS OF PROBE AND OTHER EQUIPMENT 0 O O C O O C O O O EQUIPMENT UTILIZED . . . . . . . SCHEMATIC DIAGRAM FOR PROBE SYSTEM . . iii Page vi 15 16 25 26 34 36 43 47 50 51 54 55 59 61 Page I-D. OPERATING INSTRUCTIONS . . . . . . . 63 II-A. FORTRAN LISTING . . . . . . . . . 71 75 II-B. EXPERIMENTAL DATA . . . . . . . . . iv LIST OF TABLES Table Page l-A. Phase I results by the 'D.P.' technique . . 37 1-3. Phase II results by the 'D.P.' technique . . 38 2. Phase II results by the 'P.P.' technique . . 4O 3. Phase III results by the 'P.P.' technique . 42 4. Percentage deviation from mean . . . . . 45 Figure 1. LIST OF FIGURES Simple illustration of conventional single- stage, dry electrode, parallel-plate electrostatic precipitator with suspended corona wires . . . . . . . . . . . Voltage vs. current for a precipitator with 23 cm plate spacing and 0.29 cm corona wire. Typical resistivity data as a function of reciprocal absolute temperature . . . . Point-to-plane resistivity probe . . . . Plot showing the voltage-current charac- teristics of clean and dirty plates . . .Explanation of reversal in AV vs. pb rela- tionShip I O O O O O O O I O O 0 Variation of resistivity with AV . . . . (a) Point-to-plane (SRI) resistivity probe . (b) Close-up of point and plane . . . . (c) High and low voltage supplies, ammeter . Schematic diagram for probe system . . . vi Page 10 12 19 21 32 33 57 57 58 62 I . INTRODUCTION Fly-ash is the generic term given to that portion of coal which is incombustible and which is light enough to be entrained in the hot combustion gases. It may constitute as much as 5 to 12% of most coals and as such the removal of fly-ash from stack gases before the latter are vented off into the atmosphere is a major environ- mental problem. Ironically enough, as the thirst of this materialistic world continues unabated, coal is burned in ever-increasing amounts, and the significance of 'high-efficiency' gas cleaning equipment has to keep pace with this trend. Today, in terms of the volume of gas cleaned and the mass of particles collected, the 'Electro- static Precipitator' (ESP) has the widest application and this is expected to continue at least into the near future. To fulfill present day 'air-quality' require- ments the ESP is expected to meet efficiency levels of 99 plus %, i.e., over 99% of the particulates in the ash- laden gas stream is to be efficiently collected and rendered harmless from polluting the environment. Quite naturally, to maintain such high quality standards, the factors affecting the precipitator efficiency are of particular interest. Experience and, to some extent, indirect theo- retical evidence have shown that of the many factors influencing this efficiency the fly-ash, particulate electrical 'resistivity' is one of the most common causes of poor electrical precipitation.1 Many researchers working with coal fly-ash have measured its resistivity and reported it in literature. The techniques and the measuring devices vary. As described later, owing to the poor reproducibility of precipitator fly-ash character— istics under laboratory conditions, 'in-situ' resistivity measurements become imperative. Among the many resis- tivity devices2 commonly employed for 'in-situ' resis- tivity measurement are (a) Point-to-plane probe, (b) Cyclone resistivity probes, (c) Kevatron analyser, and (d) Lurgi resistivity device, etc. Of these, it is claimed by many2 (at least in the U.S.) that the 'Point- to-Plane' resistivity apparatus is best suited for repre- sentative and convenient 'in-situ' resistivity measure- ments. Two techniques are presently employed with the 'Point-to-Plane' probe. In either technique varying voltages are impressed upon an electrostatically deposited fly-ash layer and the voltage-current readings recorded. It has been observed that the measured resis- tivity of a compacted layer of high-resistivity particu- lates decreases as the magnitude of the measuring elec- tric field increases and the reason for this is postulated8 as arising from the particulate nature of the layer. High electric fields are generated in the vicinity of the contact spots between the particles. This reduces the bulk resistivity of the bed. Also in addition to the flow of charges through the particles themselves charges can now be transferred across the small air-gaps near the contact spots, thus increasing their effective area and reducing the effective resis- tivity of the layer.8 The value of the resistivity most commonly quoted is the one calculated just prior to 'total sparkover,‘ i.e. the point of electrical break- down of the dust layer whose resistivity is being measured. The primary objective of this research work was to fabricate a 'point-to-plane' resistivity probe and to apply the two existing techniques* toward measuring the resistivity of power—plant fly-ash. The Michigan State University Power Plant 65 was the first experimental site and later the experiments were extended to the Eckert Power Plant of the Lansing Board of Water and Light, Lansing, Michigan. It was intended to obtain reproducible and dependable results with the Southern Research Institute (SRI) probe and subsequently to pre- pare an Operating manual for future use. * Viz., the 'Point-Plane' (P.P.) technique and the 'Disc-Plane' (D.P.) technique. As the research progressed certain additional objectives came to light. In spite of both the tech- niques being capable of yielding the resistivity, it was felt that one of them was far superior, both in theory and practice, over the other. Also, it was realized that the common practice of reporting the ash resistivity just prior to sparkover was, in reality, beset with some ambiguities both due to the very nature of the sparking phenomenon and the experimenter's biased judgment. Consequently, the electrical phenomena at and around sparkover was felt in need of investigation and a cri— terion yielding a more dependable resistivity backed up both by theory and experimental evidence was to be looked for. Given that fly-ash resistivity is a complex function of such varied prOperties as coal composition, gas temperature, humidity, applied electric field, etc., the purpose of this research work is less to explore the existence of new theories or methods of its measurement and more to sieve the abundant but scattered existing information on resistivity measurement and to deve10p a reproducible algorithm of reporting representative ash resistivity in a working precipitator. II. REVIEW OF THEORY AND PREVIOUS WORK The field of electrostatic precipitation is at least as old as this century and there exists an exten- sive amount of literature on the problem of fly-ash resistivity alone. A brief description of the operation of an ESP and the undesirable phenomena arising due to high ash resistivities is given below. A typical Electrostatic Precipitator consists of parallel plates spaced 8 to 12 inches apart and between which are suspended corona wires at suitable intervals (Figure 1). Very high voltages of the order of 60-80 kv are impressed between the wire and plate electrodes. A unipolar, stable, self-maintaining gas discharge (corona) between the emitting wire electrode and the receiving plate electrode is initiated due to such high field intensities. In theory both negative and positive corona are possible but in practice the negative corona is pre- dominantly used as (a) it is stable and (b) it permits about twice the magnitude of the allowable 'collection' voltage (prior to an electrical breakdown) obtainable o o o 7 With a pOSitive corona. ngle- llel-plate electro- pdd Hot combustion gases laden with fly ash pass in between the parallel plates and the particulates are charged by either of two mechanisms: (a) field charging, i.e. charging by an ionic current in an electric field, (b) diffusion charging wherein as the particle size gets smaller and smaller the phenomenon of charging shifts to one of ionic diffusion.* In either case the charged particles drift toward the collection electrode, give up their charge on deposition and thus constitute the corona current. As the particulate layer builds up the total corona current now flows through the previously collected dust layer. This establishes an electric field (E) in the dust layer pr0portional to the corona current density (j) and the particulate resistivity (3) as given by E = 3‘3 (1) As the resistivity is increased, keeping the current density constant, the electric field in the dust layer increases proportionately (Equation 1). When the electrical field exceeds the field strength for electri— cal breakdown or corona initiation within the dust layer one of two phenomena, depending on the ash resistivity, *i.e., the particles are no more charged by virtue of lying in the path of ions traveling along lines of electric intensity but by ions diffusing haphazardly in the gas stream. is likely to occur. If the resistivity is moderately high (~ 1011 ohm-cm) this breakdown may propagate across the entire air gap and thus cause a spark. This sparking occurs at lower current densities and, unlike some suggestions,2 it occurs at an applied voltage lower than that required for sparking between clean electrodes. As early as 1918 Walcott,3 in his studies on the effects of solid and powdered dielectrics on the sparkover voltage, had observed that a layer of porous or fibrous material on the plate electrode lowers the sparking voltage by up to 50% for the negative corona but has little effect on positive corona. Frank4 in 1933 confirmed Walcott's observations and established for the first time a quanti- tative relationship between dust resistivity and reduc- tion of sparkover voltage.* On the other hand if the 12 resistivity is very high (~ 10 ohm-cm) breakdown will occur at a voltage too low to propagate a spark across *1! the inter-electrode space. The dust layer will be * In many of the experiments conducted, however, the sparkover point (observed as a sudden surge in the corona current on a current meter) corresponded to an applied voltage greater than that required for sparking between clean electrodes. As pointed out later (under Discussion) among other factors this could be attributed to an inability to record the first few sparks that occur at a lower voltage. The 'incipient' spark point (shown in Figure 2) whose existence and characteristics are to be discussed shortly provides a possible explanation for this 'strange' behavior. ** In other words at lower resistivities (~ 1011 ohm-cm) the field breakdown strength of the air gap is continuously broken down electrically and will emit ions of an Opposite polarity from that produced by the corona into the interelectrode region, thus initiating the phenomenon of 'back corona'2 (Figure 2). In either of the above two cases (excessive sparking or back corona formation), the allowable collection voltage and there— fore the efficiency of collection in the ESP is appre- ciably lowered due to the high resistive dust. From the above discussion it is then abundantly clear that the accurate measurement of the fly-ash resistivity as seen by the ESP is an important factor in estimating the performance of a precipitator. In the following paragraphs a brief outline of the two types of conduction mechanisms and the various factors that affect them in experimental practice is given. For a more detailed and systematic description refer to the Southern Research Institute (SRI) report entitled "Techniques for Measuring Fly Ash Resistivity."2 Conduction Mechanisms When a plot of resistivity on a logarithmic scale vs. the inverse of the absolute temperature is made a likely to be reached before that of the dust la er and at higher resistivities of the dust layer (~ 10 2 ohm-cm) this trend is reversed and the layer breaks down first. However it is inconceivable of the two phenomena exist— ing independently and a certain amount of overlap of sparkover and back corona is to be expected. 10 so . . . i or SO . Spark paintq Apparent f1- spark 40 _ porng 30 - """" /7R;::""'- . Incipient Corona current density (nano-amperes per sq. cm) spark point 20 - p = 2 x 1011 . ohm-cm lo - . 200°C) or in the absence of any reactive or condensable material (H20, 803, etc.) volume conduction prevails and laboratory measurements are justified. However if the temperature is low or if an adsorbed surface layer does exist, as is normally the case, in situ measurements become imperative. In addition, it is believed2 that in spite Of being able tO duplicate effluent gas stream compositions in the labor- atory, the time expended in collecting, cooling and transporting a sample from the field tO the laboratory is more than likely Sufficient to promote chemical changes in the surface properties Of the ash. Several resistivity measuring devices exist differing fundamentally in the method Of sample collec- tion, mode Of sample deposition in the measuring cell, compaction of the collected bed, intensity of field applied and method Of maintaining thermal equilibrium. A more detailed description Of these different measuring instruments is given by G. B. Nichols.2 In the author's Opinion, as well as in the Opinion of many, the 'Point- Plane' resistivity probe, used in the U.S. since the early 1940s,7 comes the closest in reproducing the resistivity as seen by a precipitator. 18 Figure 4 corresponds to the 'Point-to-Plane' 11 This Southern resistivity probe used by the author. Research Institute (SRI) probe was fabricated in the Michigan State University (MSU) Engineering Shop from the SRI drawings and its details Of construction, main- tenance, precautionary measures and step by step Opera- tion are listed in Appendix I. The probe is inserted into a dust laden gas stream and allowed to reach thermal equilibrium. A high voltage of the order Of 8-20 kv is applied between the corona point and the grounded plate, the point plane spacing being anywhere from k to 1%". At such high fields a corona current is initiated and the particulates in the dust-laden stream flowing through the point plane gap are charged either by the ions or free electrons emanating from the corona in a manner similar to that in an ESP. The charged particulates drift towards the grounded plane due to the applied field and transfer their charge to the electrode by either Of the two con- duction mechanisms described before. Two major tech- niques have been employed at present to measure the resistivity Of the precipitated dust layer. In the 'Point-Plane' technique, to start with, the 'VOltage-Current' relationship for the 'clean' plate (the case with no particulate deposition) is Obtained. 19 W/f‘ Dial indicator High voltage 0 connection l’i I‘AL‘A‘L in. Ax'l.‘ LX'. .‘ " Movable shaft Stationary point Grounded ring Figure 4.--POint-to-plane resistivity probe.11 20 Next a dust layer is collected over a suitable interval Of time and then the 'dirty' (or dust-laden) plate V-I characteristics are Obtained. Typical V-I curves for the 'clean' and 'dirty' plate conditions are shown in Figure 5. AV corresponds tO the voltage drop across the dust layer for a certain corona current I. The clean and dirty plate 'spark' points shown are those Observed by experiment. To Obtain the dust thickness 2 a sliding disc of the same cross-sectional area as the grounded plate is lowered onto the dust layer and the thickness care- fully measured by a sensitive screw micrometer. In the 'Disc-Plane' technique no 'clean' or 'dirty' plate char- acteristics are necessary. After collection Of a suit— able dust thickness the disc is lowered to measure the dust thickness and increasing voltages are then applied to the layer. The current-voltage characteristics Obtained are recorded until the dust layer breaks down electrically and sparking occurs. Given a dust layer Of thickness 2, across which is applied a voltage drop AV and through which passes a corona current Of magnitude I, the bulk dust resis- tivity pb is given by _ AV x A 0b I x z where A is the area Of the collection surface. I(uA) + 21 Total 4'0 7' sparkover +— 30 6 h..- “' Clean plate 3.2 E” characteristics 2.8 ”' __ Dirty plate character- 2.4 P“ istics b—I- 2.0 P. Observed sparkover ,— v . o max1mum 1.6 - ('incipient' __ sparkover) 1'2 ET Region t. beyond 0 8 ’/2 inflexion . -——.' (Avmax) o 0.4 ”“ o f— o ° "’JJIJJILJIIJIJL 7 9 11 13 15 17 19 V(kV) -> Figure 5.--Plot showing the voltage-current character- istics of clean and dirty plates (data corresponding to 4/10/78). 22 As pointed out in the Introduction, pb is a function Of the voltage drop across the dust layer, AV, and due tO the particulate nature Of this dust layer,8 decreases con- tinuously with increasing AV. In both the techniques it is common practice to report the resistivity just prior to sparking. From experience it is known that it is exactly at such a condition when resistivity causes the most problems in a working precipitator. Listed below are some Of the advantages and disadvantages2 of using a Point-Plane resistivity measurement device for in-situ measure- ments. Advantages l. The mechanism Of particulate collection is very similar to the one Observed in an Electrostatic Precipitator. 2. The dust-gas and dust-electrode interfaces are the same as those found in an ESP. 3. The electric field and corona current densities during measurement are comparable to those in a precipitator. 4. Flue gas properties are maintained. 5. Two techniques can be simultaneously employed providing for double checking. 23 6. The dust layer is virtually undisturbed during measurement. Disadvantages 1. Dust layer thickness measurement is diffi- cult and liable to be inaccurate.* Also there is no means of checking the integrity Of the dust layer during measurement. 2. Very high voltages are required for collec- tion. 3. Time for each experimental run is tOO long (> 1 hr) 4. Particle size distribution Of ash layer deposited is not representative of that in the flue gas. 5. Sample size is tOO small and due tO other irregular experimental conditions consider- able scatter in results is to be expected. Measurements in a series Of runs have there- fore tO be made to yield some average value. 6. Experienced personnel are required to conduct the tests. 7. Carbon in ash can obstruct proper measurements especially in those by the 'Disc-Plane' method. * Check 'Sources of Error.‘ 24 Measures to minimize some Of the above dis- advantages are discussed under 'Sources Of Error and Recommendations' and also in the 'Step-by-Step' Operat- ing guideline laid out in Appendix I-D. A computer program designed to search for a unique point at which resistivity may be reported (thus eliminating the depen- dance Of the results on the experimenter's keen Observa- tion) is included in Appendix II-A. III. EXPERIMENTATION The entire experimental program was carried out in three distinct phases. Phase I: Location: Power Plant 65, Michigan State Uni- versity; inlet port to ESP, Boiler NO. 3. Duration: January and February of 1978. Special Comments: Only the 'Disc-Plane' method was used as the author and his colleague were unaware of the 'Point-Plane' technique during this time. Phase II: Location: Eckert Plant Of the Board of Water and Light, Lansing, Michigan; inlet port of ESP, Boiler Nos. 4 and 6. 4 Duration: April and May Of 1978. Special Comments: Both the techniques were used during the first half but later the 'Disc-Plane' tech- nique was dropped owing to a number of reasons listed under the topic Of 'Discussion of Experimental Technique.‘ Phase III: Location: Power Plant 65, Michigan State Uni- versity; inlet port to ESP, Boiler NO. 3 and 'model ESP inlet.‘ 25 26 Duration: August and September 1978. Special Comments: Only the 'Point-Plane' technique was employed. Discussion Of Experimental Technique Appendix I-C contains a schematic diagram of the experimental set up (Figure 9) and Appendix I-D includes a step-by-step guideline for the experimental procedure. In brief, a high voltage (8-16 kv) is applied across the point-plane gap in the resistivity probe. As in an ESP, the particulates in the gas stream are charged and pre- cipitated onto the ground plane. Once a substantial layer Of dust (.5 ~ 1 mm) accumulates, the collection is stopped. Next the Voltage-Current characteristics (with the dust layer behaving as a dielectric) for both the 'Point-Plane' (P.P.) and 'Disc—Plane' (D.P.) techniques are recorded till a spark is Observed on the current meter. The collection voltage, time of collection, temperature Of dust layer (initial and final), and other essential information (date, point-plane distance, etc.) are also recorded. The dust layer thickness is measured from the initial and final micrometer dial readings. Having Obtained the required data for the measure- ment of resistivity by both the techniques we now arrive at the crux Of this research endeavor viz., the correct 27 interpretation Of this data towards reporting a rele- vant, reproducible and unique fly-ash resistivity. As mentioned earlier the resistivity reported by the 'D.P.' technique is given simply as H4< 54y PB X where V and i are the last values of the applied voltage and measured current, respectively, that could be regis- tered before 'sparking.‘ The current here is not from a corona point but due to a flow Of electrons through a bed Of resistive particulates under the influence Of an applied field. In the 'P.P.' technique the voltage-current (V-I) relationships are Obtained and plotted as in Figure 5, both for the 'clean' plate and the 'dirty' or dust-coated plate conditions. The AV for a finite cur- rent value i (Figure 5) corresponds to that part Of the total applied voltage which is applied across the dust layer. As in the 'D.P.' technique the resistivity of the bed is given by ___Avx T Ba» ‘31: Again in common practice, the resistivity reported is calculated from those values Of AV and i that correspond to the 'spark' point for the 'dirty' plate experiment. 28 In the author's viewpoint, the 'P.P.' technique is superior to the 'D.P.' technique for the following reasons: 1. In the 'D.P.' technique the resistivity measurements are taken after the disc has been lowered onto the uncompacted dust layer. This could disturb the integrity of the bed and in cases Of high thickness well dislodge a portion of it.1 McLean has reported that resistivity could change due to rearrangement Of particles in a layer when pressure is first applied.9 2. There is a possibility that all Of the disc may not touch the ash layer resulting in air gaps1 and hence premature sparking. 3. In the 'P.P. ' method a spark travels across an air gap and then through the accumulated dust layer just as in a working precipitator. The air gap is absent in the 'D.P.' method. 4. A very sensitive current measuring device is required in the 'D.P.' method (~ nano-ampere scale). This was not available and instead a microammeter was used. Thus large layer thicknesses are required for appreciable leakage currents which in turn would neces- sitate much longer collection times. 5. In many cases when back corona might have set in before the sparkover phenomenon is visually 29 Observed on the microammeter some knowledge Of the V-I characteristics prior to 'total' sparkover leads to the definition Of an 'incipient' sparkover point (explained later in this section). The latter is useful in pro- viding a 'break-down threshold' and therefore a positive safety margin when reporting the resistivity. 6. Carbon in the ash, especially for thin dust layers, can seriously hamper the measurements by the 'D.P.' technique. As a substantial number of sets Of V-I charac- teristics were Obtained by the 'P.P.' method and the 'clean' and 'dirty' plate curves drawn on graph papers an unusual trend was noticed. Whereas it was always known that the bulk resistivity pb varied inversely with AV (the voltage applied across the dust layer), in many instances a reversal of this trend was Observed at a point close tO the visually determined sparkover point (refer to Appendix I-D for experimental procedure). This point shall be called the point Of 'incipient' sparkover as shall be made clear from the explanation to follow. The V-I data was then fitted with second order polynomials by the least squares normal equations method (on the computer). The choice Of the parabolic equation was made due tO two reasons: (1) its relative insensitivity to the data points close to sparkover 30 which is desirable* and more important (2) the form for the voltage current relationship** for a point-to-plane corona is suggested by many to be parabolic.10 As Observed with the preliminary plots over half the V—I data exhibited the inflexion point though some not so strongly as others. Possible Explanation When the voltage across a dust layer is grad— ually increased a point is reached when all conditions are favorable for a spark to jump right across the corona air gap and through the dust layer. This spark followed by a few other 'early' or 'incipient' sparks is not sufficient to cause an appreciable Surge in the corona current so as to be identified on the current meter as a 'sparking condition.‘ In the process a few holes are burned through the dust layer. These holes now act as back corona points and emit ions Opposite * A third or higher-order polynomial is extremely sensitive in this region and predicts a rapidly decreas- ing applied voltage with increasing corona currents. This is Obviously impossible! **The suggested relationship is of the form I = V2 + BV + C where I is the corona current and V is the applied voltage. However, the form used in this work (an inadvertent slip) was of the form V = 12 + BI + C. It was later confirmed using the above sug- gested form that this interchange Of variables does not affect the results appreciably. 31 in nature to those arriving from the negative corona point (see Figure 6). Thus the voltage drop, AV, exist- ing across the dust layer prior to sparkover, is reduced by an Opposing field set up by the back corona points. The resistivity however continues to decrease as the corona current is all the while increasing with increas- ing applied voltage. From the above discussion two important conclu- sions may be drawn. Firstly, the 'Point-Plane' tech- nique is a more accurate method of measuring resistivity and, secondly, in the light Of the possible coexistence Of back corona and sparkover prior to the visual identi- fication Of the sparkover point, it is more appropriate to report the resistivity at the condition of 'incipient' sparkover which then provides a convenient 'threshold breakdown' region and a point Of uniqueness for a given set Of data. Also, reporting the resistivity at such a point avoids the inevitable complicacies that arise once sparking has been well established. Figure 7 is a com- puter plot of ob vs. AV for the set Of data represented in Figure 5. The fortran program is written so as to identify the approximate region Of the inflexion point and then concentrate in that region to pin down the exact point Of inflexion. The program is listed in Appendix II-A. Listed in Appendix II-B are the 'clean' 32 Holes burned through dust layer due point P—Ké (a) 'Incipient' sparkover conditions V - Voltage applied \ J .i ‘1‘: “\‘ ix. fl. AV‘ AV - Voltage V across dust layer (b) Magnified picture of one 'hole' net net (-) - (+) - Figure 6.--Explanation Of reversal in AV vs. tionship. -:__ Grounded plate (B'-—E+)2 effective voltage drop across dust layer field intensity dust thickness negative ion or electron positive ions pb rela- Resistivity x 1011 Ohm-cm, pb 3.4 33 T 1 i l n T I __ __ P d -r—- Region below " 'incipient' sparkover '— Incipient sparkover :: F pb ~ 1.55 x 1011 Ohm-cm __ l I 1 1 l 1 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4 Voltage drop across dust layer, AV, kv Figure 7.--Variation of resistivity with AV (from data corresponding to Figure 5). 34 and 'dirty' plate recorded data for both Eckert and M.S.U. plants. Tables l-A and l-B comprise the data recorded and the resistivities calculated by the 'Disc- Plane' technique in Phase I and II respectively. Tables 2 and 3 contain a summary Of the calculated resistivi- ties and other pertinent information using the 'Point- Plane' technique for Phase II and III, respectively. Recording Of Data and Calculation Of Resistivity Shown below is an example data sheet (with data Obtained on 4/10/78) resembling the one used in Phase II and III experiments.1 SAMPLE DATA SHEET . Date: 4/10/78 6. Port Location: Inlet to ESP 1 2. Test NO.: 7 7. Point-plane distance: 3. Personnel: Mr. X 1%" 4. Plant: Eckert Station 8. Probe orientation: Vertical 5. Boiler NO.: 4 9. Boiler Condition: Full load 'POINT-PLANE' TECHNIQUE 10. Initial Temperature (Ti): 285°F 11. Initial Dial Reading: 8.39 12. Final Dial Reading: 8.34 13. Difference: .05 mm 14. 15. 16. 17. 19. 20. 21. 22. 23. 35 Voltage (kv): 7.5 8.0 9.3 11.4 12.8 14.5 16.1 >17.0 Current (uA): 0.15 0.3 0.45 1.15 1.65 2.25 3.6 Spark Collection Data Collection Voltage: 16.1 - 16.5 kv Initial Corona Current: 2.8 uA Start: 11:59 a.m. 18. Stop: 12:14 p.m. Collection time: 15 minutes Dirty Plate Characteristics Voltage (kv): 10.4 12.0 14.2 14.6 15.5 16.5 >17.0 Current (uA): 0.45 0.7 1.3 1.45 1.85 2.2 Spark Final Dial Reading: 7.89 Dust Thickness: 0.45 to 0.50 mm Final Temperature (TF): 286°F 24. 'DISC-PLANE' TECHNIQUE Voltage (V) - 740 1000 1100 > Current (pA) - 0.1 0.13 0.15 Spark 25. 26. 27. Auxiliary Data Type Of Coal: MAPCO Percentage Sulphur: .85 Special Observations and Comments: Technique II inaccurate - Current meter resolution poor - Dust thickness insufficient - Premature sparking. 36 Tabulation Of Results Disc-Plane Technique _AV A 10 pb’TXIXIO AV = voltage drop across dust layer during measurement (in kv) i = corona current during measurement (in uA) 11 oh = bulk ash resistivity (in 10 Ohm-cm) A = area Of cross-section Of collection plate (2 5.0 cm2) 1 thickness Of dust layer (in mm) Additional notations: Ti = initial temperature Of grounded plate (in °F) TF = final temperature Of grounded plate (in °F) VC = collection voltage (in kv) t = time Of collection (in minutes) (see Tables l-A and l-B) Point-Plane Technique ~9¥x§x101° pb - J. 2 Notations: P.P. Gap = Point-to-plane gap in inches Ti = initial temperature Of grounded plate in °F T = final temperature Of grounded plate in °F F 37 TABLE 1-A.--Phase I results by the 'D.P.' technique. Collection Measurement Date pb t v '1'. 'r 9. AV 1 c 1 f 1/ 5/78 - 1.7 208 - .37 *1.5 0.6 3.38 1/10/78 48 7.0 223 231 .33 *0.5 0.8 0.95 1/10/78 - 6.4 222 - .39 *0.75 0.17 5.66 1/11/78 60 7.1 225 227 .55 0.5 0.3 1.52 0.7 0.4 1.59 *0.75 0.6 1.14 1/13 - 220 220 .51 0.5 0.19 2.58 1.0 0.55 1.65 *1.3 2.2 0.58 1/14/78 - 6.8 165 - .50 .25 0.6 0.42 .50 1.8 0.28 * .75 4.6 9:16 1/15/78 - 6.6 180 - .36 .25 0.2 1.74 .50 1.0 .70 * .75 1.8 _.5_8 1/16/78 - 6.6 225 - .19 .25 1 .64 * .50 4.8 .QEZ 1/16/78 - 6.7 237 - .67 .50 0.2 1.87 .75 0.39 1.44 1.0 0.85 .88 *1.3 2.40 .fifll 1/27/78 83 7.6 217 — .60 1.0 0.3 2.78 1.2 0.6 1.67 1.29 0.8 1.25 *1.32 1.0 1;1_ 1/30/78 - 8.0 208 - .38 .62 .25 3.26 .80 .50 2.11 *1.0 1.5 ng' 2/ 5/78 92 8.7 200 203 .59 .55 .85 .55 .77 1.6 .41 * .82 2.5 _fifii 2/ 5/78 - 8.3 200 200 .48 .42 .65 .67 .50 1.3 .40 * .61 2.2 :22_ 2/12/78 - 7.5 225 - .44 .92 0.25 4.14 *1.3 0.5 2.92 2/19/78 98 8.4 221 226 .46 * .4 0.4 1.09 2/19/78 - 8.1 223 - .70 1.0 0.5 1.41 *1.4 1.0 ‘;22 * Values corresponding to last measurement before sparkover. 38 .mflgmfimhfimmg Hl> MO “mm UTflHOUOH “.me um @OUMHBHMO hflflurflflmflmmm m5.o m.o H .1. . . . O q no N OO.H cm.H >< Om omm nHm 0 OH 5H H\O~\ om. H . . . u O N v «m av. >< mm H can «mm o a nxm \ om.m on. H OH. OOH omm 4.5 OH OH\O~\O 4O. >< . m.o N.o H . . OO m NN.H mo.H >< Ha OOm man a H OH Onxvmxv Ov.m H.o H mm.H omm mom a.mH OH OH\NH\O Om. >< . OH. MH. OH. H . . no a H.H o.H we. >< OO OOH OOH H OH OH OH\OH\O . OH.o H.o H . . «H 5 0O. we. >< Om OOH omm N OH OH OH\OH\¢ . OO.H Om. OH. OH. H . . HO NH.H o.H Om. mO. >< om omm mmm H OH OH OH\O \v . m.H OO.H Om.H om. om. cm. H . . mm H». HO. HO. mm. He. Om. >< HO mmm omm m OH OH OH\O \v . o.~ OO.H om. OO. mm. H . . mm OH.H Oo.H OO. OO. Os. >4 Ov mom mom O OH OH Oa\v \O . ma. om. OH. H . . mm H 0.4 o.m NO.H >< an m mam «Hm 0 OH Om Oa\m \O n O o o a q mucmsmusmmms Hu> a a a > p mama 653538. .dd. 0:» an muHOmOH HH Ommnmll.m.H mqmda < ll 39 Collection voltage in kv duration Of collection in minutes dust layer thickness in mm Observed sparkover voltage in kv incipient sparkover voltage in kv 11 resistivity at V8' in 1 x 10 Ohm-cm resistivity at incipient sparkover voltage VS in l x 1011 Ohm-cm corona current at Vs' in uA corona current at VS in uA Tables 2 and 3) 40 .Om>ummno mcflxummm OzO OOO.H OHH.H HO.H H.H HO.OH OO.OH O.O OO. OH OHH OHH H OO\OH\O HOO.H OOH. OO.H O.H HH.O OH.O O.OH OO. OH OOH OHH H OuxOHxv HOO.H HHO.H O.O O.O OH.OH. OH.OH O.H OH.H . OHH OHH H OHxOH\O OH.H HOO. HO.H H.H OO.H OH.O O.O OH. OH OOH OOH H OO\OH\O HHO.H OH.H OO.H H.H HH.O OH.O O.H HO. OH OOH OOH H OO\OH\O OOH.H OOO. OH.H O.H OH.O HH.O O.H OH.H OH OOH OOH H OO\OH\O OOO.H OHO.H OH. O.H H.HH H.HH O.HH OH.H OH OHH OOH «H OO\HH\O OHO.H OOO.H OO.H H.H «O.OH OH.OH H.OH OH. OH OOH OOH mH OH\OH\O OOO.H OHH.H OO.H O.H OO.OH O.OH H.OH OH. OH OOH OOH mH OO\OH\O OOO. OOH. OH.H O.H «H.OH OO.OH O.OH OH.H HH HOH HOH mH OH\O \O OH.H O HH.H OO.H HO.OH O H.OH OO. OH OOH OOH «H Opr \O HO. O OO.H OO.H HO.OH O H.OH HO. OH OOH OOH OH OnxO \O OOO.H HHO.H HH.H OO.H HO.OH OH.OH O.OH OH. OH OOH HOH OH OH\O \O OOO.H OOO.H OH. OH. OH.OH. HO.OH O.OH OH.H OH OOH OHH mH OO\H \O nO .nO H .H O> .m> o> H u He He wwww OOOO .3358”. find. on» an muHsmOH HH Ommsmln.m wands 41 .cw>uwmno mus QHSOOOHHOHOH Q m> >< on» GO cowxwamsw o: #6:» OOHHQEH i OHH.H OHH.H HH. O.H 0.0H OH.OH O.HH HH.H OH HOH OOH mH OH\OH\O HH.H HH.H OH.H OH.H OO.O. OO.O 0.0 HO. O OOH OOH H OH\OH\O OOH. OOH. O.H O.H OO.OHO OO.OH 0.0 OO.H OH OOH OOH H OH\OH\O OHO. HHH. HH.H O.H O0.0H O0.0H 0.0 HO. OH HOH OOH H OH\OH\O OOO. HOH. HO.H 0.0 OH.HH O.HH 0.0 OH.H OH HOH HOH H OH\OH\O OHO. OOH. HH.H O.H HH.HH H.HH O.O HH.H OH OOH HOH H OHxOHxO OHH.H OHH.H HH.H O.H HO.O 0.0 0.0 HO. OH OHH HOH H OH\HH\O HOO.H HOO.H OO.H OO.H O.O; O.O O.O HO. HH HHH OOH H OH\HH\O HOO.H HOO.H OH.H OH.H OH.O. OH.O O.H HH. OH OOH OOH H OH\OH\O OOO. HOO. OO.H H.H OO.OH HO.OH O.O HH.H OH OOH HOH H OH\OH\O HHH. OHO. OO.H O.H HH.OH O0.0H 0.0 OH.H a OOH OOH H OH\OH\O OHH.H OHH.H O.O O.O OH.OH. OH.OH O.O OH. OH OOH OHH H OH\OH\O HOH.H HOH.H OO.H OO.H OO.O. OO.O O.H OO. OH OHH OHH H OH\HH\O OHO. HOH. OH.H OH.H OH.O HO.O O.O OH. OH OHH OHH H OH\HH\O O .OO H . H O> . O> o> H » O.H H.H WWW BOO .OmscHucoonu.H OHOOO 42 H.OMHOOOm mo OHOEEOm. OOmv mcowmeMCH .Oustmwv. oz .OOHHOOOO mm: >< m> no OH sOHuOOHMOH 02W OHO.H OHO. OH.HH OO.HH 0.0H OH. : HOH HOH H OH\H \O OH. OOO. H.O 0.0 OO.HH O.HH 0.0 OO. n OHH OHH H OH\H \O OOH.H HHO. OO.H H.H O0.0 H0.0 O.O OO. . OOH OOH H OHxH \O OHH.H OOO. OH.H H.O OH.OH O.HH 0.0 OO. O OHH OHH H OH\H \O OHO.H OHO.H O.H O.H HH.O; HH.O O.H OH. n O.OHH OHH H OH\OH\O OOO.H OOH. OO.H O.H OH.HH 0.0H 0.0H OO. OO OOH OOH mH OH\OH\O OO0.0 OHH.O O.H O.H HO.On O.O O.O HH. . HOH HOH H OH\OH\O HHO.H H0.0 HO.H OH.H O0.0 0.0H O.O HH. HH 0.00H OOH H OH\OH\O OH.HH OH.OH OO. OO. OH.On HH.O 0.0 OH. HH HOH HOH H OH\OH\O H0.0 OOH.H OH.H O.H O0.0 O0.0 0.0 OH. HH HOH HOH H OH\HH\O OO.OH OO.OH HO. HO. O.OO 0.0 0.0 OH. u OOH HOH H OH\HH\O OH.HH HOO.H OO. OO. O0.0 0.0 OO.H OH. n 0.00H OOH H OH\HH\O OO0.0 OO0.0 O.H O.H O.OHO 0.0H 0.0 OO. O OOH OHH H OH\OH\O .nO nO .H H m> .m> o> O. O Ha Ha mOO OHOO .OOOHOOOOH ..O.O. On» On OHHOOOH HHH OOOOO--.H OHOOH IV. SUMMARY OF RESULTS Owing to the strong dependance Of resistivity on temperature and sulphur content (both Of which varied over the duration of the experiments), it is not possible to Obtain a representative value Of the ash resistivity without stating the relevant conditions. Four different categories Of coal sulphur content were Observed centered around the average values Of 0.5%, 1.1%, 1.98% and 2.2%. The coal used at the MSU site was always low in sulphur (< .75%). In Phase I the temperatures ranged from 200 to 240°F. In Phase II three distinct temperature zones were Observed, 280-300°F, 300-320°F and 340-360°F. Finally in Phase III the resistivity measurements were taken at two locations; one at the ESP inlet (340-360°F) and the other at the 'model* ESP' inlet (220-255°F) . From the very outset the purpose Of this work was not so much to verify the dependance Of ash-resistivity on temperature or sulphur content (or any other factor) as to establish (if possible) the reproducibility Of the measuring probe and to interpret the data Obtained * A model ESP was fabricated for conducting cer- tain 'conditioning' and related experiments by the author's colleague. 43 44 towards gaining further insight into the resistivity prob- lem. The reproducibility, here, refers to how close the calculated values (Of resistivity) for two or more con- secutive experiments (on the same day) compare with each other. Listed in Table 4 are the percentage deviations from the arithmetic average Of resistivities estimated in a single day. As the table suggests the deviations are within normal allowances for experimental uncertainties. However when the resistivity results were plotted against temperature considerable scatter was observed (not shown here) and in the absence of proper sulphur content information (day to day) no definite regions (Of conductivity) were observed. A more systematic effort (varying one parameter at a time like temperature, or coal sulphur content, etc.) than the one undertaken in this work is essential to Obtain any dependable variation of resistivity with some parameter. Regarding the insight gained from the interpre- tation Of the results by the 'Point-Plane' technique the possible existence Of an 'incipient sparkover' point was suggested from an Observed inflexion in the ob vs. AV (voltage across dust layer) relationship. Of the Phase II experiments 71% showed the inflexion with 51% depict- * ing a 'definite' inflexion in the pb - AV plot. * By definite is meant that an appreciable (> 20% or so) difference between the corona currents 45 TABLE 4.--Percentage deviation from mean. Phase Technique Date (3611:? c511) % Deviation I 0.9. 1/10/78 3.30 x 1011 -71.0, +71.0 I 9.9. 1/16/78 .34 x 1011 -19.4, +19.4 I 9.9. 2/ 5/78 .28 x 1011 - 3.5, + 3.5 I D.P. 2/19/78 1.04 x 1011 - 4.8, + 4.8 II 9.9. 4/ 5/78 .45 x 1011 -48.9, +48.9 11 9.9. 4/10/78 7.56 x 1011 - 5.5, + 5.5 II 0.9. 4/24/78 3.33 x 1011 -14.1, +14.1 II 0.9. 4/25/78 1.19 x 1011 -71.0, +71.0 II 9.9. 4/ 5/78 1.04 x 1011 -22.1, +22.1 II 9.9. 4/10/78 2.00 x 1011 -23.0, +23.0 II 9.9. 4/24/78 1.36 x 1011 -1l.8, +11.8 II 9.9. 4/25/78 1.58 x 1011 - 6.1, + 6.1 II 9.9. 4/27/78 1.60 x 1011 -41.4, +41.4 II 9.9. 5/15/78 .95 x 1011 -24.5, +24.5 II 9.9 5/16/78 1.22 x 1011 -29.8, +29.8 II 9.9. 5/17/78 2.32 x 1011 - 5.9, + 5.9 II 9.9. 5/19/78 .66 x 1011 - 2.0, + 2.0 II 9.9. 5/24/78 .65 x 1011 -43.7, +43.7 II 9.9. 5/26/78 2.48 x 1011 -30.5, +30.5 III 9.9. 8/17/78 9.69 x 1011 -31.0, -12.1, +43.0 III 9.9. 8/18/78 10.14 x 1011 -25.9, +25.9 III 9.9. 8/20/78 3.18 x 1011 -40.0, +40.0 III 9.9. 9/ 2/78 .99 x 1011 -32.9, -27.7, +60.6 46 Sixty-nine percent of the Phase III experiments exhibited this reversal phenomenon with 54% showing noticeable evi- dence. The reason why all of the experiments did not possess a characteristic 'incipient' sparkover point is not very clear at this point. Among other explanations it could be suggested that the nature Of collection Of the dust layer and the care with which the V-I data are recorded (i.e. care in avoiding any sparking) may decide as to whether the Observed sparking point would precede or succeed any possible coexistence of sparking and back corona. Given below are some crude estimates Of the ash resistivities Obtained from an arithmetic-averaging Of the values calculated. Temperature Average resistivity Phase range °F Ohm-cm I (0—9) 200-240 1.292 x 1011 II (D-P) 280-360 2.81 x 1011 II (9-9) 280-300 1.41 x ioii 300-320 1.40 x 1011 340-360 1.30 x 10 III (9-9) 220-255 1.05 x 10}: 340-360 7.82 x 10 at 'Observed' sparkover and 'incipient' sparkover was Observed. V. SOURCES OF ERROR AND RECOMMENDATIONS Among the many factors that could not be held constant over the entire period of 'in-situ' resis- tivity measurements (such as sulphur content, tempera- ture, quality of coal burned, particle size distribution, humidity, gas flow rate, etc.) only the first two varied enough tO be of significance in the calculation of resistivity. The coal sulphur content varied from as much as 0.5 to 2.5% but generally remained around 1%. Temperatures varied considerably in both the sites. At the Eckert Plant the range was from 280 to 360°F. At the M.S.U. plant two temperature ranges existed; one at the precipitator inlet (340-360°F) and the other at the 'race-track unit' inlet (225-250°F). Another source of error could be involved in the measurement Of the dust-layer thickness. During the recording Of the 'Clean-Plate Characteristics' it is unavoidable for some ash to be collected in spite Of the probe orientation being unfavorable for such a deposition. With certain precautionary measures (like partly covering up the slots in the coverplate, or pull— ing the probe closer to the precipitator wall, etc.) the 47 48 dust collected could be limited to about .05 mm which then results in a maximum error of about 10% (for a .5 mm dust layer thickness) in the calculated resistiv- ity. The clean-plate deposition tends to decrease the slope Of the clean-plate curve which in turn reduces the AV and therefore the resistivity. In the measurement Of the corona current (espec- ially at low current values), the resolution and sensi- tivity Of the microammeter was found to be highly inade- quate and it is suggested that one use a pico- or nano— ammeter in all future work. The resolution Of the H.V. supply was found to be insufficient. An H.V. supply with a better resolution (.25 kv) and a greater range of available voltage (0—30 kv) is recommended. Among other recommendations may be included: (1) A pump attachment on the probe supplying compressed air to clean up the accumulated dust layer (between experiments). In this way a large amount of experimental time can be reduced in that the coverplate need not be dismantled and the probe allowed tO reattain thermal equilibrium prior to the next experiment. (2) The point- to-plane distance could be made continuously adjustable * so that 'Optimum' collection fields may be set up for 1: 'Optimum' in terms Of substantial build up Of dust layer (> 1.0 mm) in the shortest possible time 49 each experiment. (3) It would be helpful if the micro- meter dial needle were calibrated as a function of temperature so that the exact error due tO a change in temperature during an experiment may be gauged. (< 30 min) and with little or no disturbance Of the dust layer due to sparking during collection. This is judged from a few trial runs. VI. CONCLUSIONS With the aid of the Southern Research Institute drawings a 'Point-to-Plane' resistivity probe was fabri- cated at the M.S.U. Engineering ShOp. Its functions were tested with a series of resistivity measurements conducted over a couple Of months at two different power plants burning coal. A manual for convenient Operation was put together. The probe was found to yield quite reproducible results (see Summary of Results) and was also found suitable for 'in-situ' resistivity measurements. From the resistivity data recorded a rough estimate Of the fly-ash resistivity at each site was Obtained. For a number Of reasons the 'Point-Plane' tech- nique was found to be more accurate and representative than the 'Disc-Plane' technique. Also the value Of this technique was strengthened by the discovery that a 'threshold breakdown' voltage range may exist for the sparkover voltage. A computer program was written to facilitate accurate data fitting and to focus on the point of 'incipient' sparkover. 50 REFERENCES 51 10. REFERENCES P. N. Shukla, A. V. Someshwar, and B. W. Wilkinson. "Reliable Fly Ash Resistivity Measurement and the Criteria for Determination and Docu- mentation," presented at the National Conference on 'Qpality Assurance of Environ- mental Measurement,T Nov. (1978). G. B. Nichols, "Techniques for Measuring Fly Ash Resistivity," Rep. no. EPA-650/2-74-079, prepared by U. S. Environmental Protection Agency, August (1974). E. R. Walcott, "Effects of Dielectrics on the Spark- ing Voltage," Phys. Rev., XII (Oct. 1918). S. Frank, "Spark Discharges in Air-Dust Mixtures," Z. Physik. 81:323 (1933). R. E. Bickelhaupt, "Electrical Volume Conduction in Fly Ash," J. Air Poll. Control Assoc. 24: 251 (1974). R. E. Bickelhaupt, "Surface Resistivity and the Chemical Composition Of Fly Ash," J. Air Poll. Control Assoc. 35:148 (1975). Harry J. White, "Resistivity Problems in Electro- static Precipitation," J. Air Poll. Control Assoc. 24:313 (1974). K. J. McLean and Prof. R. M. Huey, "Influence Of Electric Field on the Resistivity of a Particulate Layer," Proc. IEE, Vol. 121, NO. 1, January (1974). T—T' K. J. McLean, "Factors Affecting the Resistivity of a Particulate Layer in Electrostatic Precip- itators," J. of Air Poll. Control Assoc., L 26, NO. 9, Sept. (1976). J. D. CObine, "Gaseous Conductors - Theory and Engineering Applications," McGraw-Hill, New York, pp. 261 (1941). 52 ll. 12. 13. 53 G. B. Nichols, "Methods and Apparatus for Conduct- ing Resistivity Measurement," Rep. no. EPA- 68-02-1083, prepared by U. S. Environ- mental Protection Agency, March (1975). H. J. White, "Electro-static Precipitation Of Fly- ash," J. Air Poll. ControlyAssoc. 21:115, Feb. (1977). E. C. Potter, "Electro-static Precipitation Tech- nology," J._Air. Poll. Control Assoc. 28: 40. Jan. (1978). APPENDICES 54 APPENDIX I-A DRAWING OF THE SRI POINT-PLANE RESISTIVITY PROBE AND PHOTOGRAPHS OF PROBE AND OTHER EQUIPMENT 55 o? O . OHON... ......1. >Ootfinq Q ,Itk {1335 3Q!- Oele €3.33 H. 8.: 1:33.. {SSE-II. I. c H _ ‘§~L..I an“. '5». Lglec .5200 ha:\o:. It: ’30:: a Z. I: . \ . “F3: to.) * hon-t O.H. out} .2 ...H .... L§§lea>t\ Ra o 0:“ tbsnuéei\3 Hon ...s C $3.96!,» is LL sfi‘uu .h lb‘{»‘.5\ O‘HOOIIQ 3?). 0:! I {00.x able): at 2‘05» .964 O!.u§\n >8.) Onstage... 3': .vsal.» 0»! {.26 ...-.2530.“ R1 OO. bot-i I! OH 9!. 52‘ 8.3232 43‘ 9.6 u'.‘ ...» 5...! st; ...-nun d u «To. x... O \0 OD! 93.. ...-use.“ DU. 0‘... Co's... .t 1.: PHOQ OOOIOO. ‘3‘ Cut. $.31 0.81.: 0.? )HHV Din. l\\—. u...’ ‘35 a «x u.:‘ . Q Lost-...: Q 2 ....Q s H? In ‘9?) Pt E at.» H: H HO.O. \ ..-. . u /...... “at“ K: ‘wew T - . . .. u“. .NNN Nos SIN 5.93.!- \ MUM-Cid . ...- 7.. .... 1...-.. W . . “Hugs? if. ...“.l .H-. .I ....V ... ... a- ,. IQO .m .1 A2 , 1 . c . /. y 945-711711/7/7/fl . , , ... . \ , vy :3 a”. .. . gnu]. -..‘x ,‘z’H ."-“/‘/'/«/ i ' H. E .4. I—. - - - . . . .l h 0. coal: \IIOlilw QQJIQNK L1 .OQLUNH © 13:30 uh 56 a V , a..---i—--o Figure 8(b).--Close-up of point and plane. 58 Figure 8(c).--High and low voltage supplies, ammeter. APPENDIX I-B EQUIPMENT UTILIZED 59 APPENDIX I-B EQUIPMENT UTILIZED High Voltage Supply: 'M-S—A Electrostatic Sampler' 115 volts, 60 cycle 0-20 kv, FS - 150 DA Microammeter: 'Hickok' [D.C.] range -15 uA to +15 uA Temperature Measuring Device: (a) 'Honeywell' Strip Chart Recorder (calibrated for iron-constantan thermocouple (b) 'Mini-mite' (iron-constantan thermocouple) Low Voltage Supply: 'Sorensen' H. V. Supply (1003-200) Ranges: 0-0.3 kv, 0-3.0 kv, 0-1.0 kv Resistivity Probe, cover plate, spare corona points, wooden supports, iron clamp with heavy base plate, cylindrical flanges, etc. Brush, piece Of cloth, leather and cloth gloves, tool box, spare fuses, ear plugs, etc. 60 APPENDIX I-C SCHEMATIC DIAGRAM FOR PROBE SYSTEM11 61 Screw 1: micrometer High (low) - ’ ‘ voltage ‘7‘ Dial supply ;’ indicator @kmete: 0 - 20 kv + '— -— —. —- I .1. ‘3" Probe I Sliding disc I Corona == == I I Grounded plane \\\\\. I ll Temp. meas. Micro- ammeter -15 to +15 uA I Protection = circuit 11 Figure 9.--Schematic diagram for probe system. 62 APPENDIX I-D OPERATING INSTRUCTIONS 63 APPENDIX I-D OPERATING INSTRUCTIONS Preparation 1. At the start Of the day's experiment check to ensure that the stem, point and plane Of the resis- tivity probe are clean. Unscrew the sliding disc grad- ually until the micrometer dial guage reads zero. After a few more turns pull the stem gently all the way back. With a light anti-clockwise tug lock the stem in this position. 2. Put on the desired coverplate with the slots or holes away from the gas flow.* 3. Screw on the high voltage cable to the rear end Of the probe and insert the microammeter lead into the BNC (Figure 8) on the probe surface. Plug in the *Sometimes due to poor gas flow conditions it might be desirable to position the holes facing the gas flow to hasten collection. Also the choice Of the point- plane (P.P.) gap rests with the experimenter. If a few pilot runs indicate that the collection rate is tOO small a smaller P-P gap may be tried. On the other hand if excessive sparking is Observed during collection and measurement a larger P-P gap should be used. It was Observed from these experiments that the selection Of the P-P gap does not affect the ash resistivity. It only affects the collection characteristics Of the dust layer. 64 65 thermocouple male joint into the female one adjacent to the current junction. 4. Connect the HV supply to the nearest avail- able 220 V single phase supply line. 5. Check to make sure the outer surface Of probe has been correctly grounded by a long, several stranded, COpper wire. 6. If flanging mechanism or support for probe is needed prepare the same for proper probe positioning. 7. Carefully Open cap Of test port. Note that substantial ash could have accumulated on the inside of cap. Also check for pressure at port. If pressure is positive exercise extreme care to avoid contact with $02 and other Obnoxious gases and dust in effluent stream. 8. Insert probe to a depth of about 1% to 2' inside the ESP inlet duct or such that the boundary regions are avoided. Rotate the probe to yield the best possible gas flow pattern in terms of the fastest collec- tion and least entrainment.* Mark this orientation. 9. Make sure that the gap between the port and the probe is well sealed to avoid dilution or alteration Of gas composition or temperature at the point-plane region. * This is determined from one or two trial runs. 66 10. Rotate the probe by 90° to the position in (8) and allow the temperature Of the plane tO rise. Check this temperature every five minutes until two consecutive temperatures are the same. This 'thermal stabilization' period runs from 20 to 30 minutes. 11. Once temperature has stabilized record it as Ti. Now unlock the stem, push it in gently until it moves no further; then screw on in clockwise direction until sliding disc contacts the plane and the dial needle stops rotatingfik Record the micrometer dial read- ing. 12. Unscrew the sliding disc and lock it in the raised position as before. The clean plate V-I char- acteristics can now be measured. Measurements by_the 'Point- PlaneT’Technique 13. Switch on the high-voltage power supply. Adjust for any zero-error in volt-meter reading. 14. With coverplate Openings perpendicular to gas flow direction as in (10) raise the voltage very gradually until the ammeter registers a current. Next increase the applied voltage in steps of 0.5-1.0 kv and record the voltage-current relationship. 1: Rotate very gradually when approaching the plane or the dust layer. Get a feel for when the dust layer is 'approached.‘ 67 15. When ammeter needle takes an abrupt jump note the voltage and current just prior to jump and then lower the voltage back to zero. Shut off the H-V source supply. 16. Check the temperature once again. Next lower the sliding disc onto the plane and record dial reading. If necessary* some Of the coverplate holes should be covered to avoid dust collection during 'clean' plate V-I measurements. 17. Rotate the probe to position the holes in the path Of the gas stream as in (8).. 18. Switch on the H.V. supply. Gradually increase the voltage until the microammeter needle registers a sudden jump. Lower the voltage by about 1-2 kv and record the applied voltage and initial current. 19. Collection has begun. The current falls ** rapidly and then remains almost steady at some low * If the reading is significantly different (by > .05 mm) from the one recorded earlier then appreciable dust has collected during V-I measurements or, alterna- tively, the temperature has changed considerably. In either case this is undesirable. **If the current does not register a rapid or steady fall then one Of the following could be the cause: (1) Re-entrainment Of deposited fly ash is tOO signifi- cant. In such a situation re-orient the slots, or change the slot pattern (to one with lesser flow area) or invert the cover plate with holes away from the gas flow (to reduce gas flow rate). (2) Point-plane gap may be tOO large. (3) The applied voltage may not be high enough. 68 value (0.5-1.5 uA). From previous trial runs Obtain the approximate collection time required to deposit a dust layer Of 0.5-1.0 mm thickness (about 20-40 minutes, in our case). Once the current stays constant at some low value for 10 to 15 minutes rotate the probe carefully to position as in (10) without switching off the power supply. This helps to prevent any re-entrainment Of dust particles were the power supply to be turned Off. 20. Now lower the voltage knob to zero and check the temperature reading. 21. Next increase the voltage gradually and record the V-I characteristics for the 'Dirty' plate condition. Record at steps of about 0.5 uA until the first spark is noticed on the microammeter. Avoid excessive sparking so as to preserve the dust layer integrity for measurements by the 'D.P.' technique. Lower the voltage to zero and shut Off supply. Dust Layer Measurement 22. Unlock the stem and lower the disc. As the disc approaches the dust layer rotate the micrometer knob very gradually. The sensitive needle Of the micrometer gauge quits from moving as soon as the disc touches the ash layer. Sometimes a slight compaction of the layer is unavoidable. Record the dial reading and also the temperature. Now the 'D.P.' technique may be applied. 69 Measurement by the 'Disc- Plane' Technique 23. Replace the H.V. supply connection by the low voltage (0-3 kv) supply cable. Turn on the L.V. supply. 24. Raise the voltage very gradually in steps Of 100 V. Note that due tO (a) poor sensitivity of ammeter used (uA range), (b) thin dust layer collected and (c) inherent nature Of this technique, sparking occurs at low voltages (.75"l.5 kv) and rather rapidly. At least two to three points of V—I data must be taken for some reliability. Sparking may be intermittent in which case prOper judgment is to be made.* Once spark- ing occurs record the V-I values prior to sparking, lower the voltage knob and shut Off power supply. Clean-up Procedure 25. Remove the probe from the port very care- fully with a pair of leather gloves and place it hori- zontally on the two wooden supports. Replace the port cap. DO not pull back the stem yet. Unscrew the cover- plate immediately. Unscrew the stem very slightly and check to see if the visual approximation of the dust *If the microammeter needle is unsteady but nevertheless does not surge ahead appreciably this could be due to (a) voltage supply fluctuations, (b) incipient sparks too few to cause appreciable current rise or (c) rearrangement Of the field in the dust layer, etc. 70 layer is compatible with the measured value. Clean the stem, point and plane thoroughly with a brush. Also clean the coverplate and the probe interior. Unscrew and E222 pull back the stem all the way and lock it. Reassemble the cover plate. Replace the L.V. supply cable with the H.V. supply cable. Reinsert the probe into the port for another run as before. At the end Of the day's experiments clean up the probe assembly, screw on the port cap securely, and unplug the main power supply. Cover all equipment with a piece Of cloth or plastic sheet and return all acces- sories to their proper place. Also Obtain the type Of coal burned, sulphur content, boiler load, etc. from the proper sources . APPENDIX II-A FORTRAN LISTING (Computer program using example data of Phase III experiments) 71 L E 9 1 9 5 9 .I U 3 9.9 9 2 9 F: E. 9 5 .I 9 z 9 3 ON N 99 5 5. 9 9 2 5"" N s 9.9, 9 Oz 9 3 5 TY| ...G 9932 .91 Z 2 9 "L IR" 0 D 9 9. O 9 ~ .9... EOE OAIP Ni“ 00 9 5 2 I 1P5 9. OF— E 9 9 9 9 9 z 9 c R 0" 9.9 7R 9 6. U ~ 9 9 0 ET EOE! X 3 90 91 3 2 9 C I FHL .99 FL L 9.912 7 9 O 2 L. D 2 FTC ‘0 E 993 9 75 3 9 5 .... E V LNET 0 C1 9 9 9 9 9 3 C V 0.9. Pan-ls N99 55 U 5 9 N O COL Ru 0 953 91. 2 5 0 I 9 on YLLD 91V.» 0 9 9 3 3 T X ESI TCP 0 .9 9 9 95 0 9 .9 T 3 ”R" R E R 99.9 9 5 O 9 O 3 E 9 TEO I 9VH :9 9 233 99 0 3 9 S 9 TN USTT T 99.3 0 9 2 Q 5 9 EV. SR E P99 9 I 95 9 3 9 0 s I RNL D F i 9 5.5 9 0 9 0 m 0 m 9.9 RAW N 00 99 Z I 90 9 9 Q 2 V m AK I Z 91 959. 0 0 3 9 E 9 .9 CES TR .992 [.96 9 .. .9 9 m 0 I 9 9 “PO NINE E '3 9 31 9. 9 3 9 = C I X 9 :- flN-lm STOP. 91 9 9.9“.9 9 0 9 7 E m I 9 I. 7 3 T X 91991 9 99 9 9 I 9 999*" L RT .5 300 2990 9 E 7 a R W ”s 99 O CRT: CROA 9R £9ch 99 9 9° ... Q 9 U 9... E A x s 7 F .HFRE E Z 9 9 923 9 9 3 9 C 5 0 99 I 3 ... ... 9 9E FY ECOH 99: 9 91 90 0 9 2 7 C L N 9 R N 9 0 95RE CASIA NT 9937 9 9099 9 9 .Q I. 9 9 l. ... 9 992 E .9 EAH LEETA .919 63 9 9 7 Z 3 O 1 66 X I N 9 9 L9 m ) TU T R UTL F 9.91 9 937 1 .9 9 9 NM I 99 Y 0 X 09 H 5L) ETL 0Y0 3C 9 92 97 9 3 9 S E II R X L F 0 C U .... UA3E BSALUT 99:97 9 99.96 9 0 9 6 E L TT T 9 0 9 I 00 "w X EVER NUVA IETR 97 ... 9 90 9 9 5 S 9 TT ‘9: 99.! F O 9 T99 L u 2‘ U0 NR'TU 935 1.9.7 9 7 7 3 R 9! II 9' H1 9.... \l O 3 o E RE N TIEIAF993Z 91 95 9.. 5 9 5C5 FF I ha 9 O Z F .. V 0 9 0 G 90 ENFVTDTS: 9 h 926 9 3 5 R...2 9 5 JJ .9 9 0)) L ...I S 9 ix 9 LHE as 019.9 9 91 O I C 9 I 999999.993 c .0 0 ‘R I!” No.91“ LL 3. E 31.9 R500 TTIU ATRJKI...O 95.9 5170 9 9 6 0000 N040 TT NDN O9 9 9 9 00 S 92 0 In. H1599 SLR 9 T RNRSH 92 92 9 9 9 9 7 L. 5 KKKK I: 9: AA 9! 9 I9 9 9 9 CC R11 9 L9 11/ 872 OSU‘RKTT‘, O 5955 9.. 5 3 9 9 9 9 C8 00 9.911 cl)”, "N E: O 9 RC N U TV|9 TIC PR U73 9 9 95 9 9 3 9 1111.9 PJNO 9.9. .. 9.. ALL LL 9 9 T199 X I9 0001 9H9 L5 5:. 91.» 70 99 3 9 9 N ...—rm... E A EENNJZJ U 9 9 9 9 N“ E 97. 6 T99 ITTu Y... X F.Xol.l 55.9.99“ 9 97 9 9 9 U (9|... 91 99799.... 99.9 TT9999 9 9 9 09.99.7979 00 H99... 99 9.92 x TN59 9HLSI0H I5U 9 67 96 6 5 “I 0 9 9 9 9 N5555 AACO’5’ E49499 RR NJ... 9 9 N 9 E00 9 EME.TEETETE(.CI 9 5 1 2 3 93N9999I) 99 777 LNNJFJ YYYY BNN F193 50 II L560 3E OGIT T 99X 37 9 3 l3..:JJJJ)UT 999 P.... 9999 L9999 .. 99 A29 C: 99 99F .. TRICUE RNAE IZlL 9 98 9 8 I105 90 9 9 9 9ZCA0012 NNI5Q A1999.) LAA P 9 IV NP 1N))9 UTZLR RIHRTFE’JE 59 5 I c TZIKIIZZ 9: 9 9 9 9N7 9.9.. H123“ 09.919 O J. 3 "LEE. 9911'“- EU 9“ L IALRXIU //6 9 E T3 9: (ETESUUFTTIATT’Y 9X3“ NR 9 9 9 9 0N0 999 9 9.... 9.: E99 I. 9L 9 T QXV ET EU .9 [DC 90 5A 9HEYXYXF ELGG:ER1299X( 9 9r. 0JJJJ NLS X 9X1L 90101 9 0E5 E9 0 UNSEVCNLIINNOI I IZTE 9999999963T 9 9 9ILI 9 9 92 911 91N(99(( EN 5999::9 0X .. 9 .9159 IN )TN NTERI OCTE II T T 9220) 9 9 9 91 ‘99...) C0005995: =0: RAAAEPESSICUILZE175ISI 2 9 9 9V TLZN A THI 9A3AALU III 020N0000T99UCCC0 EEZ(ZJK _.LG .. : .. ..UU 0A9T199 .. U199 A... (CCEL E AIEU EEDCISTO9TTCC 0306 90(1111TNL...01299 T l NllllNN3GB A TSDN TOSHPVVEEGE ENZRN TRHNERN3 9“ 9 2 99C ICIOODU 9 IITATOOKIIZHJBII: TNTAULITlsz .. L: 9R 90 RR9RI SITAPGEI 900.5 9 Z 01N0000HLV(99(100(99NNN3.9 5T 9 9 9 9TTHLLNRNN1 TNNCICTEVCIF .. UIOTU E SON-9 7 TT/ 9 9 5 :AAAARCE‘CC EE IRI J AJJJJNNOLLIOIR INIRE I0LE99995 S NICC LOSOE IX h 5‘7 2 9 ZOCEEC: FFFO 9 FFRCROOIOUC‘C‘OC AARFROO‘ORCRO 9(sERFFE N 9 ..9 N LTIRR 0L 9 00 9 9 0 UKRRRRFCVYIUIIIFFPUUAUL‘IA‘CC CCP PFUPCPFIDEVFUITTR T520. A 9 3 6 3 6 6 6 L A m 9HxN NOS 0 E FF 5 0 V 00 0 0 50 0 1 5 F 5 v5.9" STU 1 0 II 2 Q E 53 9 U 0' 1 IN? I I I 10 4. 9. 1 9. 9.. CCCCCCCCCCC C CC C C C 72 73 O E 5 3 F N 9 V ) A X 3 I T 5 9 s l n 9 I I R R 5 I9 0 T E u 5 F N V F I o N \l 9 R 0 K o I x \l A. P R I I U 9! l. A X 9 1 N U N D. E 5 9 ( G ) 0 S L 0 ET 2 0 N I I F ..K EN 9 T A 9 X T N RE “ 9 I 2 E A I 5 TN)F ) 9‘ 9 L V L", 9 N ol 5 o F T F COHX 9‘ E E N s C V Vaul‘o nu R 9 W I U L 01 T E X 0 T E K T 9 9 P 7 U E N 0 R 5 ) P 9 C H S I ) A ) 9 H U 2 O T S O F I P H1 ) .I 9 U P o 9 $09.91 I X E 5 s m m 9 CE I L H F A U N S 9 9 .... T 9 H A N 0 C T 9X 9 0 X N . U I I \93 s 9 0 3 N I E o 9 X T N 9 T ) T ) 9 C G R Z )) E K S Y (Z L N 9 )5 I G A A 9 )) L P I T X 9 U ( XE 2 9 X N T 9 N7. F S R I L5 5 E )IU (2 E I I. V. C (9| N V E VTEF E R LRQ V... L 0 0 T 9 XV I = T 1R0 9 R 9 OTI 0 E F R V I ) L 9 ) C TA 9X ) CAN 2 ) 9 N E V 3 ...) T H A 5TR9 F N NHH 9 1X 0 I 7. F I 9 01 A I. R IS 9 9 0 ( 9 C 3 ) (9 3 0 T I 9 ( S A Tune) 5 NSE 3 L V 9 1 00R 5 99 0 CV T V H E P 9 Y V ONT 3 9 93 N60 N I P 6 ( 9 N C R S... R 9 R0 T ) K S 9 0 .F 0 S O . Is ... 9) VI A ) NI. N I9 19 E5 T 0 I E C CE 0 R G 9E H H TN E ) 0 A RF ROE T R 9 0 1R 6 R S N S 9 H ( AAD S H K .9 V 9 9 9 0 Elm C ) T 9 1 U E I SEX U 0 (UI T E 9) 9 I H VX G 7. E D 2 V c R K SR5 S R HOT 0 L I) 5 \I F N LF LS CA L N 9 0 9L 0 = R E 9 9 9 LEA)V K E (K 5 J L/ 0 0 . E 9 ) SGR M A I 6 0E T E ) A TN)2 9 ) E NI ._) A 9 9 0) m C) 02 ) A ( ) 10 T H P AKH 9 X H SLY. 9P )TN(I )SL K S NL 9 9 1 )T I V P 1) U 9 0 A )( S DC(.¥ C ( OAH3 1 CK005( KNO 99 N 9 9 9 05 9 SUN 9 L29 (V 00 G H H0 9 91...... 2 KE GHI(G. I 9VR8A 90C)A 0) 9K H 91 (F 9 N E.) 9 E 9) VL 99! I (R 9 9H( 9 9 9( RLANN DINA( QRNJ: IKK( K)K( 9 9 o IORII x 011 .E 5 9 \I x X x VAT 9x I I V EOE Io N(p 9 9 E 9 9) T 9 A K N 0X) (iAER 97.000 .. 9 )0)90 01 ADC—51.300090 00 5N0 :0 NN NSR IA KTS 951C! AK: _. 9: 252 91H 9U... 9.9110EII 2 92V2... 9.9 R01... 3.9919901 E019 HOE I SOINE (R __LBIECN..I 9DER( ._ 1 J __ 1(( T1ECC9 . 1(1 9T (V(E(L1n-T1 P . 10R1(1(SZ( UZ(1 2U TESIL: UISOI 9A __ UIIJDC: UEAI 9J(L TVCG RE FF T UAZPLVGVE 9.... IPC 9 : T TU T)N T 0 N UHUSANN 8:. V NDH NI)NP )K A ) TA :0 90 RGNNTAT..LB=E: 9.. 09.99 90 :A: OS)TATA0TA/ITAT 1TI OTANII1A flI..CINC0I0JIO=I 0 UL NHKOCTCINIINHNVUI)0VFKzNOCTC FTEHNNNH NN/TNHN3ZNT R GET T9 :NZP ITIT5(N 9T 6 9:3:7 9 INPCI E9\ICIRILC I: L(PSC( EN R(INIR IR/NYARII INDBEu NICNKVA (VDN(I AIKNNV K J RC5 : FORFN .. ..RORELO(FEFSER:FORRI0 :ERCRO R0 CROR : ORONUC IN: 0.. ICOFINCFNO:((00I0(N0HO( FFVCCIDCPFPOAOV UIVRFCIGIWCGRRPFFF FF ICPFFKCPCESU CIKCKFSOIPICISUD‘ACRFUAKOFDA F C 6 6 DE 0 FOE 0 12 O“ 0 0 0 0 D 5 0 EH 0 0 0 0 0 5 0 0 2 I30 .9 b.» 5 6 0 9 0 0 0 1 T 1 3 2 k 5 5 6 7 I Q. 9. 1.9. 1 9. 9. Q. 9. U .9. 9. 0 C C C C 1 CC C C C C 74 o u «a u o- » ?H :2 to 30!”? “35.099551 mnucuua mum to». 2 N6 L §§§5fi2fi85L*9a‘h§Bu.NRou) , a a O 3|! K ‘ 020 Z" O” )s k! 1. “HA1 6 i“ § \I 90:: ‘ 3° KL R” B" ‘R '0 ’NL' 0‘” "JO N900 El ( Q 0‘“ OHCR NTN‘ 1 J‘ 9 EQONN ILO..1..I :OJE UER 9 9 TLI’ . I . OKHCU N ONHHN UOSHH H o U‘NN 1K ‘OOR osuooooas : 75R TC! RU RKEQRBRz9=iTU N 1‘ NTDBCNNN NH HJNTD OFF(CENU‘I(=O=UOU¢OEN CIII‘RESBDZNOJSDSZCRE 080 no G 9. 1 9 O 1 1 APPENDIX II-B EXPERIMENTAL DATA 75 Eckert (Phase II) V-I Data w 00 Q o QP \c \‘PQ QQ A NN QM w OM WP NAG. QQ VG. QI- 0‘? ONO Q Q'- ZVNP 0 Q Qwoe A“?! NQ QQQO NAOQ>NP vxmm QQ UNom N0 ZVI- 0 Q Q ~m~o M0 A>OQ QP €Q¢N QQ V's IN cho WP VN Q0 Q Q avxs 0% ‘DNW \P A“ 0(- N Q NQOO us VAQP MQ >nm~ ‘0 QNFQ 0Q AU 0M Vs GOP. QQ VPQP U\ NQOQtAQ \nmm ~m Aw-N Os MNCO PC QV Q0 Q Q {UPQ QC VPKJNT \ Q (ro «m QAOO QQ AOQO mw ¢NQQ QQ PV€O\MN ZJ m In '4 m M 0m") 0 I— a: N5 «BIO II II InIIBMIInO‘ C II II ‘IOI 90 GO IaI-IIINII H 00 DVD dlvfln I-Ifl III-I Odo-0001406! O m m m " m m on .4 Ca ON: “‘05 In H00 to: NM ON ”IN MU‘II I; I'DON IIBONO" III II II II INI I00: 0 OIIIBIIIII 30% 00 ON 00 “ML-I") @Mflfl «dfinwv-mmmN o m mm m m N mm N N a e N “\N ”I? on mm on IO I'HD Iowan I In IlaNmmmomgmo MOII II II II II NIOIINIIIN IIIIIIIIIIIII 13:05 0k om 00 km ngNOdmnOd mSONakmNONOH co m N N mm In 9 III N b n a (a 0 In JON Nn NM 0:0:mN No m” Icoooomkuo ooaanaNaoaN _III II II IIIII II II NIU‘IIQIIII IQIIIIIIIIII Preom no no ooo~s3 mm s: «NddOdQNOO ododoooNomod >Imm damn mNoNNcchONNmN ImNIooo nomwmvmnan on no IOIdMNeHmeonmnmNmamomNma A‘IIQIII IIIIIIIIII II II '40-'IIIIIII‘5IIIIIIIIOIIII rthéflfloa OJOBNJOHNn um um dNfifiOOONOONJOOOHNJOHNJNfl .- '*< mNmmm mmmm m mg a: ammo Ne n F4JCNIOOW mmdeOHNMQNOmOInON JBIKOImNOMIOomaBOOBmONGO fiIIOIII IIIIIIIIIIIIIIOIII InIIIIIIIOIIIIIIIIIIIII mthdahn knoahnoQONooknddun fiHdOOONdOUflNOOOflNMOflNnNO 3.: m m In In a no ~ I s m n I o to .cumoooaN uNo:moa:aomNo:okhhu0INImmoonomomdmkmomamumo .JIIIIII IIIIIIIIIIIIIIIIIIIIOINIIIIIIIIIIIIIIIIIIIII 0.: QIDI-OO‘DNN KNOOIDNOOIDWONNU‘QIDNO‘HHHIIONJNHNJU‘deNDONkI-IONIDO V- In Om '0 NIB mm own I o o 0 3 fl .HNNmnmsoomNmnoocnanNk:kdmwkkJoIced accommnomcaawedo: D:IIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIII I IIIIIIII ’OdOOOBU‘BOB‘.IDGIN-MdOOIDPIQQIDdOOWdCNnOHNMO‘IIONNBONNBIDNB EDIma m an no pm 0 I . o'NNhnaNkngNmasNIoowmmfld cmmtoIWmoWoocmNMchmomNmmmm 2‘IIIIIIIIIIOIIIIIIIIIIIIIIIIIII°IIIIIIIIIIIIIIIIIIIII bmaoemam-ma0amak°me~ooooaoam909d°0«weadodmooamaOomeme ( 0 M on In |D O '9 mom 5 mm mm m mNIevyuzunI'm::oow‘oNdodsDdvudoaooammodommfl‘r‘ 90‘! HF- NMNQU‘ JMQ‘N II I-IIII IIIIIII IIIIIIIII6IIII I’I I‘IIIIIIII omcoopoflaapm IIIDCNOIINCNOIOCOOIBOQ BINGU‘OJ 000m3091 CID m.o°lm°:.l L-b_}1 4&11bN1T&-EN}N} 8/1' 8/1 8/1 8/1 8/1 3/1 8/2 8/ 8/3 9/ 9/ 9/ 9/3 "IC'fiiifl'fiflfflmfl'TL'IWflJflflmiflr