. 3:..fgf f j i; mmm 4:39 0‘} 'ICJ~ I : 0 it a: Sim: (In: Af- r3 .9 0 #0 n. no. .. " tr LIBRARY Michigan State University ,n, ,m 1' . Y ‘ x N '4'“. ? Q -r 1?? . u ‘ M" ‘ ~ g 5:“? f” U ‘ . 4 “w is” (“-1 "v '4 3 ‘4." m" i ‘ L Q A33T3ACT A STUDY OF THE VAXIABLES THAT AFFECT HEAT PENETdATION RATES IN GLASS JARS by Edward D. Schmidt Heating rate tests were conducted using 16, 26, 32, b8 and 64 ounce glass Jars filled with water plus small (3/8" diameter), medium (1/2" diameter) and large (B/h" diameter) size marbles heated in water and in steam plus air mixtures at heating medium temperatures of 1650?, 1800? and 1950? (initial temperature 95°F) to determine the effect of: Jar size, liquid vs. liquid plus solid particles, particle size, heating medium temperature and heating medium on the heating rate of the slowest heating zone in the container. Temperatures were measured using thermocouples located in the Jars and recorded using a temperature recording potentiometer. The time vs. temperature data were plotted on semilog paper and f and J values determined. It was found that: (l) the heating rates were independent of Jar size and dependent upon the ratio of the surface area of the Jar to the volume of the jar, (2) there was no detect- able difference in heating rate due to differences in marble siZe, (3) the jars containing water plus marbles heated faster than the Jars of water, (4) the Jars heated faster as the heating medium temperature increased, and (5) the jars heated faster in the water bath at 1650? but faster in the steam air mixtures at 1950?. Edward D. Schmidt I-Iajor Proferor Date V\var‘ ‘EJ1_ \QKOS" A STUDY OF THE VARIA9L35 THAT AFEECT HTAF FEES RATION RAPES IN GLASS JARS BY v“! ddward Daniel Schmidt Submitted to the School for Advanced Graduate Studies of Michigan State University in partial fulfillment of the requirements for the degree of MASTEK OF SCIENCE Department of Food Science 1955 ACKNOWLEDGEMENTS The author wishes to eXpress his sincere thanks and appreciation to the following persons: Dr. I. J. Pflug, his major professor, who provided inSpiration, timely guidance and constructive criticism dur- ing his graduate work. Dr. B. S. Schweigert, Chairman of the Food Science Department, for his interest in the authors program. Dr. F. H. Buelow and Dr. R. C. Nicholas, members of authors committee, for their consideration and interest in their review of this manuscript. Mr. Ronald Fisher for his suggestions and help in set- ting up the experimental apparatus. Mr. Carlos Borrero, Bruce Eder, David Gurevitz and Jeshaihu KOppleman for their stimulating suggestions through- out the project investigation. Mrs. Helen Erlandson for proofreading this manuscript. ii 8. TABLE OF INTRODUCTION . . . . . . LITERATURE REVIEW . . . ‘J XFERIMENTAL PROCEDURE . I a. Container preparation. . CONTENT S b. Preparing temperature sensing elements C. Heating mediums . . (l) Aater bath . . . (2? gteau d. feasiig procedure. . e. Sv:iwdticn of data . HESULTS . . . . . . . . a. Discussion of Results plus air mixtures. (1) Effect of heating medium medium temperature . . . . . . . . . . . . . . . and . . (2) Water vs. water plus marbles (3) Effect of jar siz (4) Effect of marble SUMMARY . . . . . . . . CONCLUSIONS . . . . . . APPENDIX . . . . . . . . a. Nomenclature . . . . -- ~1fi.—:r 7‘7 ”1'1 KILILLZILIVCILLJ o o o o o o o e . size . . . . . . . . . . iii heating 0\ N H09 0\ \O k») O\ 12‘ 42* L? 41‘ 11‘ (D _. '— J 0 LIST OF fABLEfi Faye Jar Specifications. . . . . . . . . . . . . . . 7 Marble fill data . . . . . . . . . . . . . . . 3 Narble Specifications . . . . . . . . . . . . . 9 Number of marbles per jar . . . . . . . . . . . 9 Steam pressures, air flows and rotameter settings. . . . . . . . . . . . . . . . . . . . l? Summary of the heating rate results in terms of f-values for the several conditions . . . . . . Ew Jummary of the heating rate results in terms of the heating lag factor j . . . . . . . . . . . 2; fie-arrangement of the f-value value data to make possible comparisons of the f-values for three sizes of marbles. . . . . . . . . . . . . . . . 29 Average surface conductance of the transducers in the two heating mediums at the three temper- atures. (These are the averages of surface con- ductances determined using the copper and aluminum cylinders in the heating mediums at the reSpective temperatures; the curves were broken, the first f value (fl) and the second f value (f2) were treated separately. . . . . . 29 iv LIST OF TABLES (continued) Table Page 10. Results of the statistical analysis of the KL) 0 f value data for jars containing marbles . . . . ll. Calculated ratios, f-value (steam plus air) / f-value (water) and ratios of surface area / volume for the five jar sizes. . . . . . . . . . 31 Figure Page 1. measuring the temperature at the cold point OfléOZojarSoooooo00000.0... l]. 2. Setup for measuring temperatures in jars. . . 12 3. D agram of the water bath system used . . . . lb 4. Diagram of the laboratory retort system used . . . . . . . . . . . . . . . . . . . . 1 5., Determining f and j factors . . . . . . . . . 20 o. ‘The relation of f-value and heating medium temperature for 16 oz. jars heated in a non- agitated water bath and steam plus air mixtures . . . . . . . . . . . . . . . . . . 32 7. The relation of f-value and heating medium temperature for 26 oz jars heated in a non- agitated water bath and in steam plus air mixtures . . . . . . . . . . . . . . . . . . 33 8. The relation of f-value and heating medium temperature for #8 and 6b oz jars heated in a non-agitated water bath and in steam plus air mixtures . . . . . . . . . . . . . . . . 31L 9. The relation of f-value and surface to volume ratio for jars heated in steam plus air I ’f' V ‘5 afllX.bl"€':.. o o o o o o o o o o o o o o o o o o LIST OF FIGURES (continued) Figure Page 10. The relation of f-value and surface to volume ratio for Jars heated in a non- agitated water bath 0 o o o o o o o o o o o o 38 vii INTRODUCTION In the food industry the general practice when design- ing a heat process is to measure the rate of heating of the product in the container under the actual commercial heat processing conditions. The slowest heating container is customarily used for the heat process design. While this method of measuring and designing the thermal process is ideal there are certain situations where this is not possible. Under these conditions the data for a similar product, con- tainer and heating conditions are used for process design. Ihe precision of the Judgements depends on the general know- ledge of the heating process, the more knowledge there is available the better will be heat processing Judgement. There is a great deal known about the heating of liquid and solid type food products, however relatively little is known about particulate foods. This study was directed toward obtaining data relating thermal process variables of particu- late foods in a model type study. The Specific objectives were to determine the effect of three sizes of glass marbles in a water marble system, in five sizes of glass containers when heated in a water bath and in steam-air mixtures at 165°, 180° and 195°?. LITERATURE BEVIEW Ball and Olson (1957) outlined the procedure for making heat penetration tests. A discussion of their recommenda- tions follows. Thermocouples were recommended for measuring temperatures, copper constantan being satisfactory for the temperature range from 40 to 3250?. The Optimum size thermo- couple is 20 - 26 gage, a heavier wire will tend to conduct heat from the hot Junction causing erroneous temperature readings. Thermocouples should be held in cans by receptacles that do not proJect into the sides of the cans. For a glass Jar, the Jar cover is punctured in the center to receive the thermocouple assembly. Thermocouples should be connected to a potentiometer for measuring temperatures obtained. Enough thermocouples should be used to provide a complete cycle of readings every 2 or 3 minutes, depending upon the product being tested. That the data be collected and plotted on semi-logarithmic paper and heat penetration data be evaluated using f and J factors was recommended by Ball and Olson (1957). The rate of heat penetration is affected by the temperature gradient between container and heating medium; the rate becomes slower as the temperature difference decreases with the product temperature asymptotically approaching the retort temperatures (Hersom and Hulland, 1963). (0 3 The vicinity of the slowest heating point in Jars was located by Pflug and Nicholas (1961) and was found to be 10% of the height of fill measuring from the bottom of the Jars. Pflug, Blaisdell and Nicholas (1965), working with 16 ounce Jars packed with fresh cucumber pickles, found that the slowest heating zone was near the geometric center in the case of conduction heating Spears in 50° Brix syrup but moved toward the bottom of the Jar for spears and slices in a 30° Brix syrup. The volume of headSpace of a container is important and: some provision is usually made for positive control. (Joslyn and Reid, 1963). The container fill requirement under the U. 8. Food and Drug Act for products with Standard of Identity and the general requirement by the U. S. D. A. - A. M. S. is 90% of the total capacity. Blaisdell (1963) used cOpper and aluminum cylinders as transducers to determine the surface conductance coefficients of water and steam plus air mixtures; the determinations of the film coefficients (h) were made by relating the film coefficients to the basic conduction heat transfer equations. Varying the heating medium temperatures, Pflug and Nicholas (1961) found that the slope of the heating curves decreases with an increase in heating medium temperature. Pflug and Blaisdell (1961) list the factors which may con- tribute to the_increase rate of heating as: (l) the increase in thermal diffusity and decrease in viscosity of the water with an increase in heating medium temperature, affecting both the water inside and outside the Jar, (2) the increase L; in the convective flow inside the Jar produced by the initially larger temperature difference and (3) an increase in the convection heat transfer coefficient with an increase in steam air temperature. Pflug and Nicholas (1961) found that (l) a steam plus air mixture at zero velocity is a less efficient heating medium than a water bath, water spray or saturated steam when heating glass containers and (2) steam plus air mixtures vary in efficiency according to the percent of steam present, increasing with increasing percentages of steam. Investigating the effect of velocity of steam plus air mixtures, Pflug and Blaisdell (1961) found that with an increased velocity, faster heating occurred. Mixtures of steam plus air have been used commercially in processing glass containers in retorts, but during the past twenty years this has lost favor completely because of its uncontrollability in retorts of commercial size (Hersom and Hulland, 1963). The use of surface to volume ratios to predict the f parameter was suggested by Nicholas and Pflug (1961). An increase in fill ratio (ratio of product weight to fluid ounce capacity) and reduced surface to volume ratio caused nearly significant increases in J, but decreases in f (Pflug, Blaisdell and Nicholas, 1965). Blaisdell (1963) listed the following causes for vari- ation in f and J. (l) the introduction of container capaci- tance causing an increase in f and J, (2) an increase in surface resistance producing an increase in f but reduction 5 in J, and (3) an increase in f and decrease in J due to thermo- couple capacitance. The heating characteristics of two different size and shaped cucumber products, Spears and slices, were studied by Pflug, Blaisdell and Nicholas (1965) and they found that at the slowest heating zone the cucumber received significantly lower lethality values. Hersom and Hulland (1963) postulate that any substance which retards convection currents decreases heat transfer. In liquids, heat transfer takes place prima- rily by convection. According to Hersom and Hulland (1963), with solids packed in liquids, the ratio of solids to liquids will effect the heating rate. The presence of channels which permit convection currents facilitates the transfer of heat. EKPEdIMENTAL Psocsouas Container Preparation Five sizes of glass containers were evaluated in this series of eXperiments. These sizes selected are in common use in the food industry and represent a range of sizes that must be dealt with in processing a number of different types of food products. Jar Specifications are shown in Table l. The data in Table 1 includes specifications interpreted from the manu- facturers code stamped on the bottoms of the Jars. To simulate the effect of particles on the rate of heat- ing of water in glass containers, the five sizes of Jars in Table l were filled with three sizes of marbles; small (3/8" diameter), medium (1/2" diameter), and large size marbles (3/4" diameter). In Table 2 are shown data for the weight of marbles and water in the different size Jars; in Table 3 are shown marble Specifications and in Table u are shown the number of marbles per Jar for each of the three sizes of marbles. .CH onw.3 .CH nmm.m .No em.wc .mo on.uu eons .sa nwe.: .sa onm.s .No Hm.m: .No mm.ma <-namm mmsxooam . qr‘ \I § “‘0 has _N oCH moom (1:3 .qd nah.» .No mm.mm .No mm.ma «Idlouwr mmaum Host; .No NA .Am Hardy CH 00.: :« oo.m MO mm.mw No as.aa mczxooam .NO om manasaofimfiomgn H Ill-Ill.“ hemp, -H pew .SH mma.m .QH 05.: .No em.na .No mH.m Jena zcsxooam .No ma ampmsmao sneaxm: nerds: zpflomamo soapam>o How mumsm mo unwam. ouoo mamazaowassm: headpommzcm; , .msm H.0Nma “was: spas emHHHc .ps .maw ©.mmmm .msm H.3mmm .mfiw H.Nmmm mamuom .me 0.5mm .me 0.035 .wEm o.mmm amass mo .pz .mem ©.mo:m .mem H.mmmm .mem H.emom moanaws mo .pz .No no .wam o.msma , pmpwz spas emaadc .92 .mam o.oosm .msm o.amsm .mae o.omsm mampoe .mew o.mom .msm o.msm .msm o.mmm swam: so .93 .maw o.HowH .mem o.moma .mem o.amma mmappea co .uz .No we .msm m.mem amps; spas emaadc .p: .maw m.mmma I I I I m.mmma mampOH .maw o.mmm I I I I o.nmm amps: mo .pz .msw m.mmaa I I I I n.0mma mwaanms no .pz .No mm .mse 2.05s popes some eoaaae .pe .mem s.smma .mse 3.5mma .mem s.mosa masses .mew o.esm .mse o.mmm .msm o.mam amps: co .pg .maw :.Hmm .mEm 3.:ooa .mam 2.0moa mmapame mo .93 .No mm .msm o.mse amps: nude emaaae .93 .msm mmm .msm ems .mam com maouoe .mEm wam .wsw :ma .mEm Hma amps: mo .pz .mam own .mam mmm .wew mmm moanams no .px .No ma mmapacz mwamq mmHnawa enabm: meanam: HHQSW mmam and spam Haas mange: \0 TABLE 3 Narble Specifications Average Max. Diam. in. mm. Avg. Wt. Avg. Density Large Marble .723 18.36 7.68 gms, 2.396 gm/cc Medium Marble .505 12.83 2.73 gms. 2.472 gm/cc Small Marble .388 9.86 1.18 gms. 2.356 gm/cc TABLE 4 Number of Marbles Per Jar JAR SIZE ' LARGE FEDIUM SMALL 16 oz. 74 217 535 26 oz. 128 389 932 32 oz. 151 --- 1068 48 oz. 235 698 1642 64 oz. 351 942 2293 The data from the marble fill tests were to be compared with data collected where the Jars were filled with water and heated. Four sizes of water filled Jars were heated: 16 02., 26 02., 48 02., and 64 oz. Preparing the Temperature Sensing_Elements Temperatures were measured by thermocouples made at the end of duplex 24 gauge capper constantan thermocouple wire 13 introduced into the tOp of the Jars through Ecklund packing glands. A description of the assembly, designed to keep leakage to a minimum and found in Figures 1 and 1, falls 9; (l) the thermocouple wire, inserted through a hollow fiber rod, is split on the end and each half inserted into grooves of a fiber rod, (2) the grooved fiber rod is secured inside the hollow rod with epoxy resin (Epocast RlO - F resin and 10 - F 1951 hardener) and the thermocouple Junction is made at the end of the grooved rod, (3) the grooved fiber rod is coated with epoxy resin, (U) pressure fitting 8 is secured to the lug lid and to secure the rod to pressure fitting B, pressure fitting A is screwed into fitting, expanding a washer. A piece of rubber tubing is fitted over the end of the hollow rod and sealed with epoxy resin. This piece of rubber tubing keeps the flexing of the thermocouple wire at the end of the rod to a minimum. The fiber rod positioned the thermocouple at the cold point within the Jars. In this study the assumption is that the cold point is 1/10 the height of fill measuring from the bottom of the Jars. This is the cold point location for liquids in glass Jars found by Pflug and Nicholas (1961). From the Jars, the thermocouple wires were connected to input of a Brown 12-point temperature recording potentio- meter which printed the temperature every 1.33 minutes. To make sure that the errors due to a faulty data collect- ing system, were kept to a minimum, the temperatures being Ieaifrom the thermocouple system were frequently checked k:— TO RECORDER THER MOCOUPL E JUNCTION Figure 1. Measuring the temperature at the cold point Of 16 oz. Jars. 12 .’ 4/ /i TOWIEECORDER ECKLUND I [PRESSURE FITTING A THERMOCOUPLE PACKiNG GLANDV\ PRESSURE FITTING B $ $‘-%-.~..: : .- - - ..... Figure 2. Setup for measuring temperatures in Jars. 13 against terperatures being read from a mercury in glass thermometer. The thermometer and thermocouple readings being compared were taken at the same location and time within the water'bath. Heating: 2-3 3 :: E3333w33;1. The Jars were heated in hot water in a rectangular, uninsulated steel tank, 2“" x 48" x 18" (see Fig. ). A water depth of at least 12 inches, which was 5 inches above the tallest Jar, was maintained. The tank was heated by steam flowing through a pipe coil at the bottom of the tank. The water bath temperature was maintained by a Taylor Model 87 R U M7 temperature controller modulating an air Operated (Fisher-Governor type 667 A) control valve. A hand Operated diaphram type pressure reducing valve was used to adjust the upstream steam pressure to allow the steam con- trol valve to work in the Optimum control range. Three different bath temperatures were used: 1650?, 1800?, and 1950F. The steam line, after leaving the tank, was normally left open. However it had to be throttled to raise the temperature of the bath to 195°F. At 1950? the temperature cycled as much as t 20? from the mean, which was greater than found when maintaining a temperature of 1800 or 1,;09 due to the throttling of the line. Each test consisted of heating two Jars. The two Jars for all tests were located at the same positions in the water bath. The water bath was not agitated; the average temperature between the two positions was 30?. 11+} .685 Empmhm Span. Hops: 9.3 we 395.93 .m onnmdm m m>..<> a n u MuauuAm “ . a _ 22mg _ u 3 wmxw IGC) 15C) I40 = PCUXLISZW—PC'U- 130 f 8 l ' ' .zo 16:30.24 MtN. _: HO 0.: ‘1 \J >— \J K9 0 __J 0.: Z O TEMPERATURE °F 8C) 052 4 e 8H0 12 $4 10 \8 2c 2224» L<————a———+4 PRKNT CYCLE UNITS Figure 5. Determining f and J factors. 21 Surface to volume ratios, Table 10, were calculated from the Jar volumes, water fill data, and Jar surface areas evaluated. The number of marbles per Jar, Table 4, were estimated by dividing the average marble weight into the marble fill weight. Analysis of variance tests, t and F tests, were used when determining the effect of marble size on heating rates. The overall effect of Jar size, heating medium, and the presence or absence of marbles in the Jars are summarized in Tables 6 and 7. In Tables 6 and 7, average values for the heating rate of the three sizes of marbles are given. The effects of marble size on the heating rate at the different temperatures for the different size containers in the two different heat transfer media are summarized in Table 8. The surface conductance of the two heating media as measured by the copper and aluminum transducers is given in Table 9. The heating data for the three sizes of marbles were analyzed statistically and the results of the statistical analysis are tabulated in Table 10. The average J-value for all tests was 1036. Discussion of Results The results were analyzed and will be discussed in tents of heating rate or f-value. The J-values were measured but the meaningfulness of the J-value in convection heating is not fully known. Since the significance of the f-value is understood and is in general independent of J, it will be used in the analysis. . ffect cf Heating Hedium and Heating Medium Temperature. An analysis of the heating medium data in Table 6 shows that A»: (at; 23 in all instances as the heating medium temperature increased, the f-value decreased. Tests conducted by Pflug and Nicholas (1961) showed this same relationship between f-value and heating medium temperature. Pflug and Nicholas considered the possibility that the larger temperature differentials accompanying the higher processing temperatures produced stronger convection currents which were responsible for the difference in the rate of heating. The data develOped in this study verify this observation. The f value ratios, f in steam-air/f in water in Table 11 and Figures 6, 7, and 8 were prepared to aid in the com- paring the relative heating rates of water and steam-air mixtures. In general, differences are small and only trends can be pointed out. The effect of heating medium temperature on the f value for 16 oz Jars is shown in figure 6; 26 oz Jars, figure 7, and for 48 and 64 oz Jars figure 8. In these figures the relative change in f value with heating medium type and temperature is evident. The f value of Jars heated in steam- air decreases more with the increase in temperature then the f value for Jars heated in water baths. Probably, there is significance in the fact that the 16 oz Jar of marbles, the smallest Jar with the lowest heat capacity and the 48 and 64 oz. Jars of water, the largest Jars with the greatest heat capacity behave differently then the rest of the group. The results of surface conductance measurements (data in Table 9) reflect the data in Figures 6, 7 and 8 that at 1950F there should be essentially no difference in heating 24 NH NH HH NH mH on OH on NH .02 ooH. 0mm. mmm. one. «am. qua. qu. omN. HHn. HNq. H¢.mH mm.mH MO.NH hm.mH mm.m 0N.HH No.0 Hm.0H mm.NH mH.¢H wo.m mw.w No.0H mw.OH N Amm mwma so conancoov w NH m NH NH mH an m NH m on 0 NH .02 mmm. wHa. osm. Hom. mmm. mom. HNM. qnm. 04¢. Nms. mHm. 0mm. me. wmq. *Qm moomH No.0H mm.wH mm.mH mm.ON mH.NH NO.NH Hq.HH mm.HH mm.MH om.¢H sw.m Hq.m ON.HH om.HH m 0 NH @ w NH HH NH mN 0 NH 002 qmm. MHm. Nmm. mos. mum. mom. mHm. Nan. mmo. 0mm. mow. van. 3. mwN. *Qm mommH mmasamaoosmp sanos wsfipwom qo.mH >m.FH NH.nN mm.nN Hm.nH mm.MH sm.na mm.nH om.oa om.ma mm.HH mm.oa NN.MH NH.NH m hHm m span new a span ads a 3969 has a snag has a me9 add a Span has a swan swoum amps: ammpm amps: swopm amps: somam amps: swmpm amps: smopm amps: asoam amass BSHUms maprmm .maoHuHocoo Hmam>mm one now moszpuh yo magma sH muHSmmL mama msupwmn map mo hawBESm mHnaws a poem: mHnaas a amps; amps: amps: mHnams a amps: mHnaws a amps: mHnaws a gown: oHpams a amps: ponds amps: oHans @ Loud; mHQLmE a pops: amps: amps: HHHm .w mqmqe oNO .NO .NO .NO .NO .N0 0N0 .NO .NO oNO .NO .NO .NO .NO ma ma wq Nm NM mN 0N 0N oN 0H 0H 0H ©H mNHm ash 25 wH NH .oz scapwdpon oawocmpm cmpospaoo mamma Mo amnssz n HHH. wm.qH m qu. Hom. Nw.mH wH mqm. HNq. MN.NH m Nmm. qN. cm.oH NH own. *0m 9 .oz #Qm momma moomfl mmasamamaaop ssfioms msdpwom mm.wH oq.©H mm.mH Fm.ON M a #H a m OOZ MNN. Hom. co». «mm. *Qm mome .02 mm.ma sm.wa swam N©.mN Q has a Bmmpm sump amps: aHm a ammum Span amps: ssHoos mnHuwmm Anmscchoov oHnaws a Loud: moHpawe a Loud: amps: amps: HHHm 0 qu<9 .No am .No #w .NO #0 .NO Q0 wNHm awn /O a L H\. r4CD O LINN bAo 5A0 \O (\1 OO 0 \O\O omma. omma. momH. mmwo. mmmH. onmo. sons. mama. » rt msH mbaso wsaumo: on» mo mayo» :H muHSmoH mush oaso. 0:3.H mmoa. mmm.a omeo. 0:3.H oomo. mmm.a mmsa. mns.a mmzH. mm:.H mama. oo:.a mmoa. mmm.a mama. om:.a mamo. mam.a mama. Ham.a Gama. mmw.a ammo. oaa.a mama. mmm.a am a s a oma mmaSumHm ANN ammo so vossHocoov sans. mass. omHH. mmHH. MHMH. mmmo. meow. mJOH. monH. 3:0H. mHmo. mmmH. NOOH. Ohmo. Sm * ooa.a mmm.a omN.H NdN.H 00\ \OC'“. Mg?- 0 HH mmm.a NmN.H osm.H Hmm.H H:N.H mmm.H omm.H omm.H H mome damp Edavms mcfiuwom mQHpmmS tsp Has a swoon zpmfl hops? has a Emmpw gasp amps: an a_ssopm sump amass Add @ Edmpm Span amass has a Esmpm sump amass aHs was mdopm Spmn Honda aHs mcs Eamon Sump amass :dHcvs flfifiummm “nut. s Hmpmg QHQHQE .H w... ) «w as... mmxrmman .Hmpmwbw H®D®3 amass mHoams rH .9 «w. m.» magma: amass mHoass a ma mflc m& r amps: amass amps: oHQams and? mHuLCF amass amass HHHm .H Honomm mo mamafidm .s mamaa .No as .No as .No as .No ms .No mm .No mm .No pm .No mm .No 0N .No am .No OH 0NO ml.” .No 0H .No 0H mmmam amp some. ammo. mmON. mmmo. *Qm momma omm.H oqmo. owN.H mmN.H wbmo. bom.H onm.H mono. omN.H mmN.H mMHH. Nom.H a *Qm n .mOOwH ‘ madamamosme ssHomz mchwmm mmmH. mnHH. awno. mmbo. *Qm mommH oom.H ¢¢N.H ONm.H QMH.H H has a smmpm gown amps: aHm a swoon gown amass schoe mGHuwmm goes: a oanms amps: a mHnams amps: amps: HHHm ADmSEHacoov N mqmge onpwH>cQ camosmpm u * oNO #6 .No as .No so .NO #W oNHm ash 28 TABLE 8. Rearrangement of the f-value data to make possible comparison of the f-values for the three sizes of marbles. Water bath Marble size medium Small Medium Large f No.* r No.* r No.* min. min. min. 16 oz. 165°F 10.45 11 10.53 10 10.67 4 180°F 9.31 12 9.39 12 9.64 6 195°? 8.72 12 8.80 12 9.24 6 26 oz. 165°F 13.37 12 12.89 11 13.49 11 180°F 11.74 12 11.33 12 11.57 10 195°F 10.74 12 10.94 12 11.17 6 32 oz. 165°F 14.05 6 13.94 5 180°F 12.24 6 11.79 6 195°F 11.31 6 11.22 6 48 oz. 165°F 18.00 6 18.02 6 17.88 5 180°? 16.34 6 16.25 6 16.41 5 195°F 15.74 6 15.54 5 15.47 6 64 oz. 165°F 18.22 4 18.49 4 18.09 6 180°F 16.45 6 16.40 6 16.54 6 195°F 15.87 6 15.74 6 15.85 6 *N0. = Number of tests conducted Marble Size Steam plus Small Medium Large Air Medium f No. f No. f min. min. min. 26 oz. 165°F 13.48 4 13.43 4 13.20 4 180°F 11.35 6 11.37 6 11.50 6 l95°F 9.92 5 10.17 4 9.87 6 32 oz. 165°F 14.05 6 12.97 6 180°F 12.22 6 12.08 6 195°F 10.19 5 9.82 6 29 TABLE 9. Average surface conductance of the transducers in the two heating mediums at the three temperatures. (These data are the averages of surface conductances determined using the c0pper and the aluminum cylinders in the heating mediums at the respective temperatures; the curves were broken, the first f value (f1) and the second f value (f2) were treated separately. he Surface Conductance in the Two Heating Mediums 165 deg. F. 180 deg. F. 195 deg. F. Water Bath h for f1 147.24 168.33 189.17 h for f2 127.50 150.33 167.00 Steam Plus Air h for f1 56.25 71.80 143.25 h for f2 65.75 93.00 192.50 3v IASLi 10. desults of the statistical analysis of the f-value data for Jars containing marbles. Significance of Karble Fill Data When Heating the Jars in a Water Bath Jar size Heating temperature F-value Level of range significance 1. 16 oz. 95-165O F 1.289 Hone 2. 16 oz. 95-1800 2 3.092 rooe 3. 16 oz. 93-1950 8 19.992 396 4. 26 oz. 95-1650 9 7.214 99z 5. 26 oz. 95-1800 8 6.247 973 6. 26 oz. 95-1950 2 14.591 992 7. 48 oz. 95-1650 2 .118 None 8. 48 oz. 95-1800 2 .336 None 9. 48 oz. 95-1950 8 .693 None 10. 64 oz. 95-1650 2 2.056 Hone ll. 64 oz. . 95-180O F .llr None 12. 64 oz. 95-195O F .254 None P-value 13. 32 oz. 95-1650 2 .982 None 14. 32 oz. 95-180O F 3.221 None 15. 32 oz. 95-1950 F .934 Eons Significance of m rble fill data when heating Jars in steam and air Jar size Heating temperature P value Level of range sirrificance ./ , o - 4 1. 20 oz. 95-165 9 .358 .3ne 2. 26 oz. 95-180O F .369 None 3. 26 oz. 95-1950 F .940 None T value 4. 2 oz. 95-1650 2 8.200 993 5. 32 oz. 95-1300 F 1.129 None 6. 32 02. 95-195 F 1.001 None [11 11 TABL Calculated ratios, f-value (steam plus air) / f value (water) and ratios of surface area / volume for the five Jar sizes Water fill f value (steam plus air)/ f value (water) Jar size Surface to 1650 E 1800 9 1950 F volume ratio 16 oz. .1027 1.091 .969 .923 26 oz. .0967 1.052 .918 .915 32 oz. .0956 ----- ---- ~--- 48 oz. .0761 .969 .933 .870 64 oz. .0764 .958 .965 .863 Water plus marble fill 16 oz. .1027 1.102 1.049 1.025 26 oz. .0967 1.009 .988 .914 32 oz. .0956 .966 1.011 .887 48 oz. .0761 1.060 .984 .860 64 oz. .0764 1.036 1.010 .908 f (minutes) A 1 1 32 150 141 13+ I 12.. m; 11.1 101!- . (D Hater fill-water bath 91' Q '.-Jater fill-steam 91' air (3 water 4 marble fill-water bath C3 Hater & marble fill-steam 4 air 8 4 t 4— ‘ f *4 150 160 170 180 190 200 210 Temperature F Figure 6. The relation of f-value and heating medium temperature for 16 oz Jars heated in a non- agitated water bath and in steam plus air mixtures. f (minutes) 17 H 13 '\ 15 13 12 11 13 T 33 4 1 Th 4» (agater fill-water bath (7Nater fill-steam 4 air (BWater & marble fill-water bath CJNater % marble fill-steam & air 5 t HL Y U r lfiH 150 160 170 180 190 200 210 figure 7. O . Temperature r The relation of f-value and heating medium temperature for 26 oz Jars heated in a non- agitated water bath and in steam plus air mixtures. f (minutes) A) N 2O 19 18 17 l6 15 14 13 12 11 34 1 4 n 4)- 4” “'8 OZ. 61" OZ. Jars Jars ' .\ O 0 Water fill \\ " <7 V Water fill \ a A Water 3c marble fill U C] Water 9 marble fill " water bath ______. Steam 1 air 1 : 1 s 4 '1 1 150 160 170 180 190 200 210 Figure 8. On Temperature r The relation of f value and heating medium temperature for 48 and 64 oz Jars heated in a non-agitated water bath and in steam plus air mixtures. ' 9L“. 35 rate between water bath and steam—air mixture with some difference expected at 16508 and 1800?. In general the results confirm this; the trend in Table 11 is for the steam-air to become more effective as the heating medium temperature goes from 165°_to 1800 to 19508. The f-values ratios for the two heating mediums appear roughly to group themselves with Jar surface to volume ratios; at lower surface to volume ratios, 0.076 compared to 0.096 or 0.103 the f-value heating medium ratio for water in Jars appear to be smaller, whereas for water plus marbles the difference is less pronounced or there is no difference. For both water and water plus marbles there appears to be a decrease in the f-value ratio as temperature increases which suggests that the relative effect of the surface film of water vs. steam plus air, changes with heating medium temper- ature. Steam plus air becomes relatively more effective, f is smaller, as we go from 165 to 1950?. Comparing the h values at 1650 and 195°? we find that the h ratios are 56/147 and 1 143/189 and the h ratios are 65/127 and 192/167 reapectively. 2 This h ratio comparison would seem to explain the change in f ratio. This result suggests that the rate of heating of water in jars is more dependent on heat transfer coefficient than the rate of heating of water plus marbles; this is true even though the f-value of water plus marbles is smaller than the f value of water. (The relative heat capacity of the Jar of water is sufficiently larger than the heat capacity of the Jar of water plus marbles to make this possible) It follows that in Jars of water plus marbles, flow resistance 36 is probably the limiting factor as far as rate of heating is concerned. The results of these eXperiments appear to fit into the overall pattern of steam-air heating. Pflug and Nicholas (1961) using a non flow steam-air heating system found that steam-air mixtures at very low velocity were not as efficient as a water bath in cases when the external film coefficient had a controlling influence. Pflug and Blaisdell (1961) established that the effectiveness of steam-air mixtures varies directly with velocity, that at the low velocities used by Pflug and Nicholas (1961) steam-air mixtures can be very bad but at higher velocities the differences between steam-air mixtures and water are small. The eXperiments in this thesis project were carried out under controlled steam-air velocity conditions selected to approximate commercial flow conditions. Obviously under the steam-air flow conditions evaluated the steam-air was in general less efficient than water at 165 and more effective than water at 195°. Water vs. water plus marbles. The effect of water vs. water plus marbles is shown graphically in Figures 9 and 10 where regardless of heating medium or fill ratio the f-values are smaller for the water plus marbles than for the water. Rephrasing in terms of heating rates: the jars containing water plus marbles heat more rapidly than jars of water. In jars of water plus marbles the heat capacity of the system is smaller than for water alone due to the relative difference of density x Specific heat of glass, 150 lb/ft3 f (minutes) 37 224. 21 4r 20 + 19 1f 18,1 17 1, 16 i“ 15 + 14«+ 13 1- 12 T- .Jater gt deter Harble fill fill o 11-» o o 165 F \ a . 1808? \ D |. 195 3 \ I1 - 1 1 1 1 V Y I I 0.06 ’ 0.07 0.08 0.09 9.10 0.11 1 surface to volume ratios Figure 9. Phe relation of f value and surface to volume ratio for jars heated in steam plus air mixtures. f (minutes) 2h . — 23 .. 22 21 o 20 19 w 18 l 17 1 16 1 15 11+ 13 1 h I 12 Eater fill marble fill ‘ O \ 165°? ! m I 195°F \ \ 111 o 10 8 l g; A r ‘1 I 0.06 0.07 0.08 0.09 0.10 0.11 4r d Surface to volume ratio Figure 10. The relation of f value and surface to volume ratio for jars heated in a non- agitated water bath. x 0.18 BTU/1b OF = 27, compared with water, 62.4 1b/ft3 x 1.0 BTU/lb 0; = 62.4. The solid glass marble heats by con- duction, therefore not only is the final heat capacity of the system reduced 56.7% for that part of the volume replaced by glass, but the glass portion of the system will absorb heat at a lower rate than the water portion (the temperature of the glass will lag the temperature of the water). Since the surface area of the jar remains constant we are theoretically increasing the surface to volume ratio which produces faster heating (smaller f-values). Obviously we are not reducing the f-value linearly as we theoretically increase the surface to liquid volume ratio. In the jar of water plus marbles the water will be flow- ing through a series of small channels (spaces between the marbles) therefore the resistance to flow will be higher than in jars of water. The velocity of the convection fluid flow will be a function of the flow resistance or friction drag; consequently heating should be faster in a water filled jar than in a jar with water plus marbles. In the convection heating system the convection flow driving force, temperature difference, which is a function of the heat transfer rate to the jar is going to be about the same for jars of water plus marbles as for jars of water since water contact surface area will be only slightly reduced by the point contact of the marbles with the jar. The result is probably that there is sufficiently more convective flow pressure in jars with water plus marbles to overcome the increased friction. If the size of marble is reduced to a no point where the friction becomes quite large the result would be slower heating. It can be concluded that since spherical particles do not block the flow when heating the liquid mass and since they make only point contact, the addition of particles in the’ range of 3/8 to 3/h inch diameter to a liquid”system should not appreciably affect the rate of heating. If the particles are large and have flat sides that can prevent liquid wall contact in a significant surface area the heating rate will be reduced. Effect of jar size. In figures 9 and 10 the f-value data from Table 6 are shown as a function of the surface to volume ratios. The rate of heating increased consistently (f-value decreased) as.the surface to volume ratio increased. Nicholas and Pflug (1961) showed that correlation of heating rates with surface to volume ratios are more meaningful than correla- tion of heating rates with Jar capacity. The rather good agreement of differentisized containers with similar surface to volume ratios in Figure 7 (for example, the 48 and 64 oz. Jars have similar surface to volume ratios, 0.0761 and 0.0764, and have similar f-values when the type of fill and heating medium are the same) suggest that the heating rates of water or water plus marbles in Jars with other surface to volume ratios can be predicted if in the same overall range of conditions. .Effect of marble size. The effect of marble size is shown in Table 8. A statistical analysis was made to determine if the differences in Table 8 were significant; the results of the statistical analysis are shown in Table 10. It was found 41 the f-value deviation of replicate runs was greater than the difference in f-value due to marble size variation for 16 of the 21 comparisons. A 0.75 marble should heat at a rate 25% as fast as a 0.375 marble therefore jars of water plus the 0.75 marbles should heat faster because the rate of heat removal is smaller plus the fact that the flow path in the 0.375 marbles should have a higher resistance which would slow the rate of heating. Since in these experiments there appear to be no major differ- ences in the rate of heating of the jars with either large or small marbles, it must be concluded that neither of these effects are significant in this range of conditions. Decreas- ing the size of marbles to 0.25 or .1875 inch may change the results dramatically. It can be concluded that the effect of the size of particle over the range tested in this experiment do not pro- duce significant effects as a function of particle size. These data cannot be extrapolated since there is certainly a critical particle size that has a significant effect on heating rate. SUKKARY A study was made of the heating characteristics of water and water plus marbles in five sizes of glass containers in water and Steam air mixtures at 165, 180 and 19503. Glass jars in common use in the food science industry having 16, 26, 32, h? and 6h fluid oz capacity were studied filled with water, water plus 3/8 in. diameter, water plus 1/2 in. diameter and water plus 3/b in. diameter glass marbles to determine the effect of the particulate objects on the heating rate. The jars were heated in a water bath and in steam - air mixtures at 165, 180 and 1950?. The temperature - time heating characteristics were determined by thermocouples located at the slowest heating zone in the container. The temperature-time data were plotted and the resulting curves 5 analyzed for I and j values. The effect of jar size, water vs. marbles, marble size, water bath vs steam-air, ani heating medium temperature were determined as a function of the heat- ing rate. The results of the study showed a good correlation between the surface to volume ratio of the container and the heating rate. The rate of heating of the jar increased (f value decreased) as the surface to volume ratio increased. There was no detectable difference in the rate of heating I'i ‘/ H) O H d .30 i0 (9 .1) H (D size marbles. Jars with the water plus marbles heated faster than jars of water. :1) H H (.1. )n H l') heated faster with increasing heating medium temperature. Greater changes in heating rate were observed in steam-air mixtures then in a water bath. The difference between water bath and steam-air mixtures was small due to the relatively high flow rate of the steam-air mixtures; how- ever, the water bath was more effective at 1650 P with the steam air more effective at 1950 F. CONCLUSIONS The following conclusions can be drawn with reSpect to heating 16 to 64 oz. jars in a water or steam air medium from 95-1650F, 95-1800F, or 95-1950F. l. 5. As the heating medium temperature increases the f value decreases. In these tests water was in general more efficient at 165°F with steam plus air being more efficient at 195°F. The f values correlate well with surface to volume ratios rather than jar size, the f value decreases with an increasing surface to volume ratio. Jars with water plus marbles heat faster than jars of water. No difference in the rate of heating was detected due to differences in the three sizes, 3/8", 1/2", or 3/4" of marbles. 44 b1 h APPENDIK Nomenclature is the time required to produce a given sterlization effect at 250 dearees F. T - T, is the log factor computed from d T1 - To is heating medium temperature in degrees F. is the initial temperature in degrees F. is the temperature at th = O; the intercept value of the straight line asymptote in degrees F. indicates heating time, the time the container is subject to a given heating medium temperature. is the time required, in minutes, for the asymptote of the heating or cooling curve to cross one 10g cycle, i.e., the time required for a 90A change in temperature on the linear portion of the curve. Subscripts are use” to denote successive values if more than one linear portion is used to describe a heating or cooling curve. is the surface heat transfer coefficient in BTU/hr. (ft.) 0 . 1 L 1“?" *fi wry-"('1 4 u. «I. ~ 4 7...]:‘14'3 4; ‘14-.) Ball, C. C. and F. C. u. Olson. (1957; Sterlization in “ Food Lechnology. McGraw-hill Inc., New York. 633 pp. Blaisdell, J. L. (1963} Natural convection heatinz of liquid: in unagitated food containers. Thesis for the degree of Ph.D., Mich. State Univ., East Lansing. Coulsozz, J. N. and J. 19. Richardson. (1956} 2333193; .. 3rd edition. Pergamon Press. New York, London 531 Paris. 331 pp. hersom, A. C. and E. D. Hulland. (1963) Canned quig. 5th edition. J. A. Churohhill Ltd. London. 279 pp. Joslyn, K. A. and J. L. Reid. (1963; Food Processing operations. Volume II. The A. V. I. Publishin: Company Inc. W¢stport, Connecticut. 580 pp. Nicholas, 3. C. and I. J. Pflug. (1961) Heat processing characteristics of fresh cucumber pickle products. II. Q art. Bull. Mich. Agr. Exp. Sta., Mich. §tate Univ. 4U:§22. Pflug, I. J., J. L. Blaisdell. (1961; The effect of velocity of steam air mixtures on the heating rates of glass 0:3tainers. Quart. Bull. Eich. Agr. exp. ata., Mich. Ftate Univ. UU:235. l 0 LA“ deferences (cantinued) 8. Pflug, l. J., J. L. Blaisdell and R. C. Nicholas. (1 65; Rate of heating and location of the slowest heatinm zone in sweet fresh cucumber pickles. Food Technol. 19:121. Pflug I. J. and R. C. Nicholas. (1961} Heating rates in glass containers as affected by heating medium and product. Quart. Bull. Mich. Aar. Exp. Sta., Mich. State Univ. ““3153. M193HIGAN STATE UNIVERSITY LIBRAR IE 1131113313 11313111311111131111 1