Search

Bu çalışmada ağaç kurutmanın tarihi gelişimi, sanayideki yeri ve ağacın fiziksel ve kimyasal özellikleri detaylı olarak incelenerek ağaç kurutmasının ağaç endüstrisinde uygulanış şekilleri araştırıldı ve bunlardan en uygun kurutma yöntemi olan sıcak hava ve su buharı ile kurutma metodu üzerinde detaylı inceleme yapıldı. Daha sonra ısı ve kütle transferi analojisi ve gözenekli, higroskopik yapıya sahip olan ağacın içinden sıvı akışı incelendi. En son olarak da model olarak ele alman 11 mm kalınlığındaki bir ağaç parçacığının üzerinde cereyan eden ısı ve kütle transferi olayım tek boyutlu olarak matematiksel modelleyip bu model üzerinde sıcaklık ve nem değişimlerini belirlenen sınır şartlan ve kabüller çerçevesinde 24 saatlik bir kurutma süresi içinde değişimi, oluşturulan model eşitliklerinin hazırlanan bir bilgisayar Fortran programı yardımı ile çözülmesi ile grafik olarak elde edildi. Elde edilen sonuçlar daha önce yapılmış olan deneysel verilerle karşılaştırılarak uygunlukları gözlendi. In this study, heat and mass transfer between porous material and its environment during the drying of porous materials are investigated. For this purpose, as an application field of drying, especially application modes of wood drying in industry are examined. Matematical modelling of heat and mass transfer in drying processes is done. This constructed model is applied to wood materials drying. Obtained equations are solved by using finite differance method in computer. Results are discussed and compared with experimental data in the literature by considering temperature and moisture distributions.

06.03.2018 tarihli ve 30352 sayılı Resmi Gazetede yayımlanan “Yükseköğretim Kanunu İle Bazı Kanun Ve Kanun Hükmünde Kararnamelerde Değişiklik Yapılması Hakkında Kanun” ile 18.06.2018 tarihli “Lisansüstü Tezlerin Elektronik Ortamda Toplanması, Düzenlenmesi ve Erişime Açılmasına İlişkin Yönerge” gereğince tam metin erişime açılmıştır. Son yıllarda, mikro kanallarda akış ve ısı geçişi karakteristikleri üzerine çok sayıdadeneysel ve teorik çalışmalar yapılmıştır. Bu çalışmalar, bazen klasik bağıntılarlauyuşması bazen de onlarla ters düşmesi nedeniyle sonuçların tek anlamlılıktan nekadar uzak olduğunu açıkça göstermektedir. Bu tezde, esas olarak bu sapmalarınnedeni analiz edilecek ve tartışılacaktır. Bu deneysel çalışmaların kronolojik analiziyapıldığında, mikro kanallardan elde edilen deneysel veri ile geleneksel teoriarasındaki farklılığın azaldığını görmek mümkündür. Bu gerçek, mikro kanalınpürüzlülüğü ve geometrisinin daha iyi kontrolünün sonucu olan mikro üretimtekniklerindeki büyük gelişme ve yapılan deneysel testlerin doğruluğundaki artışlakısmen açıklanabilir. Bu nedenle, daha önceden yapılmış çalışmaların sonuçlarıbazen hatalarla dolu ve yanıltıcı olabilmektedir.Bu tezde, mikro kanallardaki akışta ısı taşınımı ve basınç kaybı üzerindeki ölçek vemikro etkileri rolünün, klasik teoriyle kanıtlanabilen, literatürde yayınlanan pek çokdeneysel sonuçla açıklanması hedeflenmektedir. Mikro ölçülerde ısı geçişi alanındakianalitik yöntemler ve karmaşık ölçüm tekniklerine dair son 5 yılda büyük ilerlemekaydedilerek, mikro kanal akışının anlaşılmasına büyük katkıda bulunulmuş vegeleneksel teorik modellerin hangi durumlarda mikro kanal akışındaki ısı geçişidavranışını doğru tahmin etmek için kullanılabileceğini belirlemede yardımcıolmuştur. Bu tezde sunulan, sürmekte olan birçok araştırma projesine dairtartışmalarla birlikte, ısı geçişi ve mikro kanal akışını içeren eserlerin bir özetidir.Teorik ve deneysel çalışmalar, hesap analizleri, daha yenilikçi modellemetekniklerinin bazılarının tanıtımı ve bunlara ait tüm önemli bulgular sunulmuş,karşılaştırılmış ve tartışılmıştır.Anahtar kelimeler: Mikro Kanallar, Mini Kanallar, Faz Dönüşümünde Isı Geçişi,Basınç Kaybı, Tek Faz, İki Faz, Isı Taşınımı, Ölçek etkisi, Mikro Etkiler. In the last years a large amount of experimental analyses have been addressed to thestudy of the fluid-dynamical and heat transfer characteristics of single-phase flows inmicro channels. These reviews show clearly how the results are far from univocal,sometimes agreeing with the classical correlations, at other times contradicting them.In this paper the main aspects which can be responsible for these deviations will beanalysed and discussed. From a chronological analysis of these experimental works itis possible to note that the discrepancy between the experimental obtained for microchannels and the predictions of the conventional theory is decreasing. This fact canbe partially explained by considering the dramatic improvements in micro fabricationtechniques with the consequent more appropriate control of the roughness andgeometry of the micro channel and with the increase in the accuracy of theexperimental tests made. For this reason the results of the older studies aresometimes fraught with errors and can be misleading.In this work the role played by the main scaling effects and micro-effects on thepressure drop and on the convective heat transfer for single-phase flows in microchannels will be analysed with the aim to explain that many experimental resultspublished in the open literature can be justified by using the conventional theory. Thetremendous progress over the past five years in sophisticated measurementtechniques and analytical methodologies in the area of micro scale heat transfer havegreatly added to the understanding of the phenomena unique to micro channel flowand has helped to clarify those situations where conventional theoretical models canbe used to accurately describe and predict the flow and heat transfer behavior inmicro channel flows. Presented here is a review of the published literature along withdiscussions of several ongoing research projects involving micro channel flow andheat transfer. Theoretical and experimental studies, computational analyses, adescription of some of the more innovative instrumentation techniques, andimportant findings from all of these are presented, compared and discussed.Key Words: Microchannels, Minichannels, Phase Change Heat Transfer, PressureDrop, Single-Phase, Two-Phase, Convection, Scaling Effects, Micro Effects.

Bu tezin amacı mühendislik uygulamalarında oldukça geniş bir kullanım alanı olan poroz ortamın araştırılmasıdır. Soğuk duvar yüzeyi ve diğer yüzeyler arasında doğal taşınımla ısı transferini, porozitenin etkisini incelemek üzere küçük bir buzdolabından oluşan deney düzeneği hazırlanmıştır. Boş soğutucu içinde ve sınır tabaka içindeki sıcaklık dağılımı gözlenmiştir. Poroz ortam doğal taşınımla olan ısı transferini artırmıştır. Sonuç olarak, bloklar özellikle soğuk duvar yakınındaki ısı transferini artırmıştır. Bu çalışma risk taşıyan hassas besin gruplarının depolanmasına ve istiflemesine bir model teşkil edebilir., The aim of this thesis is investigation of porous medium that using a usage field at the engineering aplications. An experiment was carried out using a refrigerator model that heat is transferred by natural convection between a cold wall an the other walls,which are exposed to heat losses. The air temperature profile in the boundary layers and in the central zone of the empty refrigerator model was investigated. Porous medium was increased heat transfer by natural convection in the refrigerator. As a result of, blocks enhance heat transfer particularly near the cold wall. This paper can be model to keep and stack of sensitive foods.

Yüksek hızlı hava jetleri, tekstil ve kağıt sanayisinde, metallerin ısıl işlemleri ile elektriksel ve elektronik ekipmanların soğutulmasında genellikle yaygın bir şekilde kullanılır. Bu çalışmada yuvarlak bir lüleden dönerek çıkan hava jeti, düz bir levha üzerine dik açıda çarptırılarak, levha üzerindeki ısı transfer karakteristikleri çeşitli parametrelere bağlı olarak incelenmiştir. Deneyler normal (boş), çok kanallı (0=0°) ve dönen (0=22.5°, 41° ve 50°) jetler için çeşitli Reynolds sayılarında ve lüle-levha mesafelerinde yapılmış ve lüle çapı 15 mm alınmıştır. Deneylerde levhanın yüzey sıcaklıkları, sıvı kristal metodu ile ölçülmüştür. Sonuçta, normal, çok kanallı ve dönen jetlerin Nusselt sayılarının durgunluk noktası uzaklığıyla değişimleri elde edilmiştir. Reynolds sayısı arttıkça Nusselt sayısının arttığı, maksimum Nusselt sayısının normal, çok kanallı ve 0=22.5° açılı dönen jetlerde durma noktasında, 0=41° ve 50° açılı dönen jetlerde ise durma noktasından yaklaşık X/D=2.5 değerinde meydana geldiği gözlenmiştir. Re=20 000 için maksimum Nusselt sayısı yaklaşık olarak H/D=6 lüle-levha mesafesinde elde edilmiştir. High speed air jets are commonly used in the textile and paper industry, thermal treatment of metals and cooling of electrical and electronical equipments. In the present work, air jet issuing by swirling from a cylindrical nozzle was perpendicularly impacted to a narrow plate and the heat transfer charecteristics on the plate were investigated depending on various parameters. Experiments were performed for normal (empty), multi-channeled (0=0°) and swirling jets (0=22.5°, 41° ve 50°) at different Reynolds numbers and nozzle-to-plate distances, and the diamater of the nozzle was taken as 15 mm. In the experiments, the surface temperatures of the plate were measured using liquid crystal method. Consequently, the changes of Nusselt numbers in radial direction from stagnation point were obtained for the normal, the multi-channeled and the swirling jets. It was observed that Nusselt number increased with increasing Reynolds number, and that the maximum Nusselt number occured at the stagnation point for the normal, the multi-channel, and the swirling jets with 0=22.5° while it occured at approximately X/D=2.5 for swirling jets with 0=41° and 50°. For Re=20 000, the maximum Nusselt number was obtained for a nozzle-to-plate distance of approximately H/D=6. 69

ÖZET Plakalı ısı değiştiricilerindeki sinüzoidal dalgalı levhalar arasındaki ısı taşınım katsayısının ve basınç düşüşünün incelendiği bu çalışmada, çeşitli parametrelere göre önceden yapılmış deney sonuçları veri olarak kullanılmış ve sayısal yöntemler yardımıyla incelenmiştir. Elde edilen değerler tablo ve grafiklerle ortaya konulmuştur. Veriler değerlendirilerek levhalar üzerindeki yerel ısı geçişinin, ele alınan geometri üzerindeki değişimi grafikler halinde verilmiştir. Sonuçların irdelendiği bölümde, daha önce yapılan çalışmalarda bulunan değerler elde edilen sonuçlarla karşılaştırılmış ve uyum içerisinde oldukları görülmüştür. Ayrıca, bu çalışma sırasında dikkate alınan ve ihmal edilen diğer noktalar üzerinde durulmuş ve bunların etkisinin ne yönde sisteme ne yönde etki edeceğine ilişkin değerlendirmeler yapılmış ve bundan sonra yapılacak çalışmalar yönünde fikir verilmiştir. SUMMARY This study primarily centers the effect of various physical parameters on the heat transfer rate and heat transfer coefficient of the fluids which flow between horizontal wavy walls. Heat exchangers are used to heat recovery in industry and to obtain different temperatures which is needed in process. Plate heat exchangers is used commonly because of various usage possibilities and their compact structures. A plate exchanger consist of pack of rectangular pressed plates, suspended vertically, and clamped together in a frame by tie bars or screws. The plates are corrugated to increase strength and to improve heat transfer, and are fitted with peripheral sealing gaskets. Four corner ports, which communicate with appropriate connections mounted on the frame, are arranged so that the two liquids flow through alternate passages between adjent plates, usually in countercurrent flow, with heat transfer taking place across the plate surfaces. Plate heat exchangers find their main applications in liquid-liquid heat transfer duties. They are most common in the dairy, beverage, general food processing, and pharmaceutical industries where their ease of cleaning and the thermal control required for sterilization/pasteurization makes themideal. They are also used in synthetic rubber industry, paper mills and petrochemical plants, and in process heaters, coolers and closed-circuit cooling systems. Over 60 different plate patterns have been developed worldwide. Since the geometry of present study has not been studied previously in literature, firsly two adjacent flat walls channels were analyzed and then two adjacent sunizoidal wavy walls channels were analyzed. In the further researches can be also employed to examine graphically dimensionless numbers with mass transfer and velocity distribution. Besides this.various channel geometries can be analyzed. In this study, numerical method was chosen due to limited time and possibility of easy apply. The problem has been modeled and solved using finite difference scheme. Present results are compared to previous experimental studies and industrial usage. In the first step, the finite difference scheme model is arranged 400 grid in the x direction and 24 grid in the y direction. Boundary conditions in the entrance and exit areas can be velocity and pressure for the flow analyses and temperature, heat flux for the heat transfer analyses. The boundary conditions for the geometries in figure 8. 1 can be express as below: > Entrance velocity is constant and horizontal and vertical velocity component is zero. > The velocity on the channel wall is zero. > Gravitational effects are neglected. xii> The physical properties of fluid in the channel such as density, viscosity, specific heat are assumed constant. > Fluid is assumed as isotropic. > The pressure gradian in the exit is zero. > The temperatures in the entrance and exit of the channel are zero. > The entrance and exit effects and secondary flows are neglected. if ``v.iA ?Wrt. (a) (b) Figure 1 The grid domain of the problem. (c) These analyses are compared to the each other and the results of the previous research, [4], and the results of the present study are showed using tables and graphics. The heat transfer rate between flowing fluids in two paralel horizontal sinuzoidal wavy wall is examined and the results are analyzed The results shows that heat transfer coefficient primarily depends upon fluid flux and channel geometry. It is determined minumum and maximum values at certain points on sunizoidal wave using graphical results. Fig. 2 Numax-Re corelation at case (a) Especially, as shown Fig. 2 when the fluid flux is high, this stuation can be seen clearly. The results of the problem for various Reynolds numbers are as below: It can be seen that the higher Reynolds number cause the higher Nusselt number and the higher heat transfer coefficient. On the other hand, when Reynolds number increase, the pressure drop between entrance and exit also increase. It could be seen clearly at Fig. 3. In the high Reynolds number values, there is a big difference between minumum and maximum local heat transfer coefficients on the wavy channel. (see Fig. 4) XUl.7```. ı;` i Kanal geometrisi H_3.8rrs p/h2 hs H_38rre p/h2 ta H_19ms p/h2 te 0.78 Kanal Boyu Fig 3. Pressure distribution along a sinusoidal wavy channel Kanal Geometrisi Pres_3.8ms p/h2 hava Pres_19ms p/h2 hava Pres_38ms p/h2 hava Kanal boyu Fig 4. Local heat transfer coefficient in a sinusoidal wavy channel xiv 88

ÖZET Hava şartlandırma sistemlerinde ve endüstriyel proseslerde her zaman için sistemden çekilmesi ve uzaklaştırılması gerekli bir ısı bulunmaktadır. Bir başka deyişle bir soğutma işlemine ihtiyaç vardır. Geçmişte bu işlem doğal bir su kaynağından veya şebekeden çekilen soğuk su ile gerçekleştiriliyordu. Bugün ise şebeke suyu ile bunu gerçekleştirmek pahalıya mal olmaktadır. Aynı şekilde soğutma suyunun doğal kaynaklardan çekilmeside nispeten zor olmakta ve ayrıca ekolojik dengeye zarar vermektedir. İşte bu noktada su soğutma kuleleri sistemdeki atik ısıyı uzaklaştırmak için ideal bir çözüm olmakta ve yukarıda belirtilen sorunların üstesinden gelmektedir.Bu çalışmada soğutma kulelerinden genel olarak bahsedilmekle birlikte,odak noktasını ters akımlı soğutma kuleleri oluşturacaktır. Ters akımlı soğutma kulelerinde ısı ve kütle geçişim irdelerken,teoriyi oluşturmak ve konuyu anlamayı kolaylaştırmak için3ölüm 2 ve 3 'de temel bağıntı ve kavramlar açıklanmıştır.Bu bölümlerde hava ve subuharın ideal gaz olarak kabul edeceğimizden karışımlar ve ideal gazlar anlatılmış,soğutma kulesindeki ısı ve kütle geçişini açıklayabilmek için birarada ısı ve kütle geçişi konusuna değinilmiştir. Daha sonra sırasıyla kütle geçiş katsayısı, Lewis bağıntısı, psikrometrik diyagram, bağıl ve özgül nem, adyabatik yaş termometre sıcaklıkları açıklanrmşur.Bölüm 4'de ASHRAE Handbook esas alarak soğutma kulelerinin çalışma prensipleri, tipleri ve tasarım esasları hakkında bilgi verilmiş ve kulede sis oluşumu açıklanmıştır. Son bölümde ısı ve kütle geçiş bağıntılarından faydalanarak, ters akımlı su soğutma kuleleri için ısı ve kütle geçiş teorisi çıkarılmış ve Merkel eşitliği elde edilmiştir.Buradan kule hacim hesabı için NTU(Number of Transfer Unit) değerini veren bir integral bağıntısı bulunmuştur. Daha sonra giriş ve çıkış şartlan verilen, örnek bir su soğutma kulesi için hacim ve kule yükseldiği hesabı yapılmıştır Bu hesap yapılırken NTU integrali Simpson 1/3 metodu kullanılarak sayısal olarak çözülmüştür.Bu amaçla FORTRAN 77 ile hazırlanmış basit bir bilgisayar programı bölüm sonunda verilmiştir.Aynı örnek problem daha sonra adım adım integrasyon metodu kullanılarak çözülmüş ve sayısal çözümle karşılaştırılmıştır. Daha sonra örnek kule için, kule hacminin yaş termometre sıcaklığı ve su/hava debileri oranı ile değişimi, grafik olarak sunulmuş ve açıklanmıştır. viii SUMMARY Heat and Mass Transfer Analysis of Counterflow Cooling Towers Most air-conditioning systems and industrial processes generate heat that must be removed and dissipated. Water is commonly used as a heat transfer medium to remove heat from refrigerant condensers or industrial process heat exchangers. In the past, this was accomplished by drawing a continuous stream of water from a natural body of water ot a utility water supply, heating it as it passed through the process, and then discharging the water directly to sewer or returning it to the body of water. Water purchased from utilities for this purpose has now become prohibitively expensive because of increased water supply and disposal costs. Similarly, cooling water drawn from natural sources is relatively unavailable because the increased temperature of the discharge water disturbs the ecology of the water source. Air-cooled heat exchangers may be used to cool the water by rejecting heat directly to the atmosphere, but the first cost and fan energy consumption of these devices are high. Cooling towers overcome most of these problems and, as such, are commonly used to dissipate heat from water-cooled refrigeration, air-conditioning, and industrial process systems. A cooling tower cools water by a combination of heat and mass transfer. The water to be cooled is distributed in the tower by spray nozzles, splash bars, or filming- type fill, which exposes a very large water surface area to atmospheric air. Atmospheric air is circulated by (1) fans, (2) convective currents, (3) natural wind currents, or (4) induction effect from sprays. A portion of the water absorbs heat to change from a liquid to a vapor at constant pressure. This heat of vaporization at atmospheric pressure is transferred from the water remaining in the liquid state into the airstream. The temperature difference between the water entering and leaving the cooling tower is the range. For a system operating in a steady state, the range is the same as the water temperature rise through the load heat exchanger, provided the flow rate through the cooling tower and heat exchanger are the same. Accordingly, the range is determined by the heat load and water flow rate, not by the size or capability of the cooling tower. IXThe difference between the leaving water temperature and the entering air wet-bulb temperature is the approach to the wet bulb or simply the approach of the cooling tower. The approach is a function of cooling tower capability, and a larger cooling tower produces a closer approach ( colder leaving water ) for a given heat load, flow rate, and entering air condition. Thus, the amount of heat transferred to the atmosphere by the cooling tower is always equal to the heat load imposed on the tower, while the temperature level at which the heat is transferred is determined by the thermal capability of the cooling tower and the entering wet-bulb temperature. The thermal performance of a cooling tower depends principally on the entering air wet-bulb temperature. The entering air dry-bulb temperature and relatively humidity, taken independently, have an insignificant effecet on fliermal performance of mechanical-draft cooling towers, but they do affect the rate of water evaporation within the cooling tower. The thermal capability of any cooling tower may be defined by tile fblowing parameters : 1. Entering and leaving water temperatures 2. Entering air wet-bulb or entering air wet-bulb and dry-bulb temperatures 3. Water flow rate The entering air dry-bulb temperature affects the amount of water evaporated from the water cooled in any evaporative-type cooling tower. It also affects m flow through hyperbolic towers and directly establishes thermal capability within any indirect-contact cooling tower component operating in a dry mode. The thermal capability of cooling towers for air conditioning is identified in nominal capacity, based on heat dissipation of 1.25 kw per condenser kilowatt and a water circulation rate of 54 mL/s per kilowatt cooled from 35°C to 29.4 ° C wet-bulb temperature. For specific applications, however, nominal capacity ratings are not used, and the thermal performance capability is usually stated in terms of flow rate at specified operating conditions ( entering and leaving water teperatures and entering air wet-bulb and/or dry-bulb temperatures ). According to the location of th fan corresponding to the fill and to the flow arrangements of air and water, currently widely used mechanical draft cooling towers for HVAC&R can be classified into the following categories:. Counterflow induced draft. Crossflow induced draft. Counterflow forced draft Counterflow Induced Draft Cooling Towers In a counterflow induced draft cooling tower, the fan is located downstream from the fill at the air exit. Atmospheric air is drawn by the fan through the intake louver or, more simply, an opening covered by wire mesh. Cooling water from thecondenser or recirculating water from the coil, or a mixture of the two is evenly sprayed or distributed over the fill and falls down into the water basin. Air is extracted across the fill and comes in direct contact with the water film. Because of the evaporation of a small portion o the cooling water, usualy about 1 percent of the water flow, the temperature of the water gradually decreases as it falls down through the fill countercurrent to the exracted air. Evaporated water vapor is absorbed by the airstream. Large water droplets entrained in the airstream are collected by the drift eliminators. Finally, the airstream and drift are discharged at the top exit. The evaporatively cooled water falls into the water basin and flows to the condenser. In a counterflow induced draft cooling tower, the driest air contacts the coldest water. Such a counterflow arrangement shows a better tower performance than a crossflow arrangement. In addition, air is drawn through the fill more evenly by the induced draft fan and is discharged at a higher velocity from the top fan outlet. Both higher exhaust air velocity and even velocity distribution reduce the possibility of exhaust air recirculation. Crossflow Induced Draft Cooling Towers In a crossflow induced draft cooling tower, the fan is also located downstream from the fill at the top exit. The fill is installed at the same level as the air intake. Air enters the tower from the side louvers and moves horizantally through the fill and the drift eliminator. Air is then turn upward and finally discharged at the top exit. Water sprays from the nozzles, falls across the fill, and forms a crossflow arrangement with the airstream. The crossflow induced draft cooling tower has a greater air intake area. Because of the crossflow arrangement, th tower can be considerably lower than the counterflow tower. However, the risk of recirculation of tower exhaust air increases. Counterflow Forced Draft Cooling Towers In a counterflow forced draft cooling tower, the fan is positioned at the bottom air intake, that is, on the upstream side of the fill. Cooling water sprays over the fill from the top and falls down to the water basin. Air is forced across the fill and comes in direct contact with the water. Because of the evaporation of the water, its temperature gradually decreases as it flows down along the fill in a counter-flow arrangement with air. In the airstream, large water droplets are intercepted near the air exit by the eliminator. Finally, the airstream containing drift is discharged at th top opening. Because the fan is located near the ground level, the vibration of the counterflow forced draft tower is small compared with that of the induced draft tower. Also, if the centrifugal fan blow toward the water surface, there is a better evaporative cooling effect over the water basin. However, the disadvantages of this type of cooling tower are the uneven distribution of air flowing through the fill, which is caused by the forced draft fan. In addition, the high intake velocity may xirecapture a portion of the warm and humid exhaust air. Counterflow forced draft cooling towers are often used in small and medium installations. Materials of Construction Materials found in cooling tower costruction are usually selected to resist the corrosive water and atmospheric conditions. Wood. Wood has been used extensively for all static components except hardware. Redwood and fir predominate, usually with postfabrication pressure treatment of waterborne preservative chemicals, typically chromated-copper-arsenate (CCA) or acid-copper-chromate (ACC). These microbiocide chemicals prevent the attact of wood-destructive organisms, such as termites or fungi. Metals. Steel with galvanized zinc is used for small-and medium-size installations. Hot-dip galvanizing ?fter fabrication is used for larger weldments. Hot- dip galvanizing and cadmium and zinc plating are used for hardware. Brasses and bronzes are selected for special hardware, fittings, and tubing material. Stainless steels (principally 302, 304, and 316) are often used for sheet metal, drive shafts, and hardware in exceptionally corrosive atmospheres. Cast iron is a common choice for base castings, fan hubs, motor or gear reduction housings, and piping-valve components. Metals coated with polyurethane and polyvinyl-chloride are used selectively for special components. Epoxy-coal tar compounds and epoxy-powdered coatings are also used fbr key components or entire cooling towers. Plastics. Fiberglass-reinforced polyester materials are used for components such as piping, fan cylinders, fan blades, casing, louvers, and structural connecting components. Polypropylene and ABS are specified for injection-molded components, such as fill bars and flow orifices. PVC is increasingly used as fill, eliminator, and louver materials. Reinforced plastic mortar is used in larger piping systems, coupled by neoprene O-ring-gasketed ball and socket joints. Concrete, masonry, and tile. Concrete is typically specified for cold-water basins of fielderected cooling towers and is used in piping, casing, and structural systems of the largest towers, primarily in the tower industry. Special titles and masonry are used when aesthetic considerations are important. The basic theory of cooling tower operation was first proposed in 1923 by Walker et al. However, the first practical use of the differential equations was developed by Merkel in 1925. He combined the equations for heat and water vapor transfer and used enthalpy as the driving force to allow for both sensible and latent heat transfer. Heat is removed from the water by a transfer of sensible heat due to a difference in temperature levels and by the latent heat equivalent of the mass transfer resulting from the evaporation of a portion of the circulating water. Merkel combined these into a single process based on enthalpy difference as the driving force. The theory used by Merkel requires two main assumptions, namely, that the water loss by evaporation is neglected and that the Lewis number for air/water vapor systems is unity. The theory states that all of the heat transfer taking place at any xiiposition in the tower is proportional to the difference between the enthalpy of air saturated at the temperature of the water at that point in the tower. Quantative treatment of cooling tower performance by dealing with heat and mass transfer separately is very laborious. Therefore the simplifying approximation of MerkeFs enthalpy theory has been almost universally adopted for the calculation of tower performance. MerkeFs differential equation for the cooling tower was redeveloped by Nottage and converted to a graphical method of solution by Lichtenstein. Another graphical prosedüre, for determining the air process line in a cooling tower, was suggested by Mickley. Simpson and Sherwood carried out experimental studies on several small-scale cooling towers, and examined the dependence of the mass transfer coefficient on the various air and water properties. Carey and Williamson extended MerkeFs theory to be applicable to gas cooling and humidification, and proposed the Stevens diagram (devised by W.L.Stevens ) for the solution of the cooling tower integral necessary for determining the required volume of a tower. Baker and Shryock reviewed MerkeFs work and examined the effects of some of the approximations. Further theoritical work on cooling towers has been carried out by Berman, Hsu et al., Threlkeld, Yadigaroglu and Pastor, and Whiller. Extensive sets of curves for cooling tower design, based on MerkeFs theory, have been prepared by the American Society of Heating, Refrigerating and Air Conditioning Engineers. In this study, heat and mass transfer theory of counterflow cooling towers are presented. The performance of counterflow cooling towers is analyzed by applying by so-called Merkel model. This model is based on equations developed in a paper published in German by Merkel (1925). The equations express an energy balance and describe simultaneous mass and heat transfer coupled through the Lewis relation. However, in the interest of tractibility the equations were simplified by omitting a term and as a result do not account for the mass of water lost by evaporation. Given the inlet water and air conditions the Merkel equations predict the enthalpy (hence wet-bulb temperature ) of the outlet air, but not its humidity. The equations also predict the required number of transfer units (NTU) to accomplish the process. XIII 59

ÖZET Dünya enerji gereksiniminin önemli bir kısmını karşılamakta kullanılan sıvı hidrokarbon yakıtlar, genellikle, püskürtme yoluyla çok küçük damlacıklara parçalanarak demet halinde yakılmaktadır. Bu nedenle, sıvı yakıt damlacıklarının, ya da genel olarak sıvı damlacıkların, buharlaşması olgusu bilimsel araştırma konulan araşma girmiş ve bu alanda bir hayli araştırma yapılmıştır. Konu üzerinde yapılan çalışmaların geçmişine bakıldığında, atomize olmuş yakıtların yanması üzerine yapılan araştırmaların çoğu tek bir damlacık için yapılmıştır. Damlacıklar arası etkileşimi irdeleyen araştırmaların sayısı ne yazık ki çok azdır ve bu çalışmalar da hep durgun ortamlarla sınırlı kalmış ve yalnızca dİffüzyon gözönüne alınmıştır. Oysa, uygulamada kullanılmakta olan yanma odalarında, damlacık çapma dayalı Reynolds sayısı 200 gibi yüksek değerlere dek artabilmektedir. Damlacık düzeneklerinde hız dağılımları ve yerel ısı geçişine ilişkin deneysel bilgilere raslamak mümkün olmamıştır. Sunulan bu doktora çalışmasında, silindirik hücre modeli kullanılarak, sıvı damlacık dizileri çevresinde, momentum, ısı ve kütle geçişi bilgisayar yardımıyla sayısal olarak irdelenmiştir. Çözüm yöntemi olarak sonlu farklar yöntemi kullanılmıştır. Sonlu farkların eğrisel sınırlara olan uyumsuzluğu nedeniyle, uygun bir çözüm alam oluşturmak için Sayısal Grid Üretimi yöntemi kullanılarak koordinat dönüşümüne gidilmiştir. Üretilen grid üzerinde, Navier-Stokes bağıntıları girdap- akımfonksiyonu formülasyonuyla çözülmüş, ayrıca kütle geçişinin momentum ve ısı geçişi üzerindeki etkileri de gözönüne alınmıştır. Damlacıklar arasındaki mesafenin değişik değerleri için çözüm yinelenmiş ve böylece damlacıklar arasındaki etkileşimin damlacıkların birbirine yakınlığı ile ilişkisi gösterilmiştir. Kurulan modelin, tek damlacık için verdiği sonuçlar da kapsama alınmıştır. Tek ve üç damlacık hali için, kütle geçişinin olduğu ve olmadığı hallerde, hız dağılımları, akım çizgileri, eş-sıcaklık eğrileri, girdap ve kütle kesri dağılımları elde edilmiş ve bunlardan yararlanarak damlacıkların yüzeyinde yerel Nusselt ve Sherwood dağılımları çıkartılmış, bunların integrasyonu ile de genel Nusselt ve Sherwood sayılarının Reynolds sayılarına bağlı değişimleri incelenmiştir. Sonuçlar, kütle geçişinin ve komşu damlacıkların geçiş süreçleri üzerinde etkili olduğunu göstermiştir. xı MOMENTUM, HEAT AND MASS TRANSFER IN LIQUID FUEL DROPLET ARRAYS SUMMARY Combustion of liquid fuel droplets is a very complex process. Many workers have studied evaporation of a single isolated liquid droplet into a stagnant gas. A real liquid spray behavior is affected by many features. Some of these are relative motion between gas and droplet leading to convective heating and internal circulation, transient heating of the liquid, droplet-droplet interactions, droplet - turbulent eddy interactions, and multicomponent liquid behavior. Relatively few works have been carried out on the interactions between droplets and all of these studies were limited to quiscent environments. Since the Reynolds number based on droplet diameter can be as high as 200 in real combusters, the diffusion-only analysis used in these studies should be inadequate to understand the interactions between droplets. Therefore a cylindirical cell model was developed firstly by Tal and Sirignano. This model provided significant improvement in compliance with experimental results as compared to the spherical cell model, another model first applied to the range of intermediate Reynolds numbers by LeClair and Hamielec. Considering a certain number of droplets in a periodic configuration (array) it can be possible to produce very useful results on the effect of interactions between droplets on the various transport processes such as momentum, heat and mass transfer. A schematic representation of the assemblage geometry which is also used in the present Ph.D. study can be seen in Fig.2-1. Due to symmetry and nearly periodic character, there is no momentum, heat and mass transfer at the streamwise equidistant planes between the droplets. By this assumption the problem is reduced to a multitude of spheres in tandem in a square streamtube (Fig.2-2) which is subsequently replaced with a cylindirical duct of equal cross-sectional area. Details of this model can be found in the paper by Chen and Tong, but the focus of our work is the effect of mass transport on other transport phenomenas. The most effective way to overcome the problems which are encountered in the use of finite-difference method is the use of boundary-fitted coordinate (RFC) system. The technique is based on a method of automatic numerical generation of a general curvilinear coordinate system having a coordinate line coincident with each boundary of a general multiconnected domain under study. With coordinate system generated to maintain coordinate lines coincident with the boundaries, finite- difference codes can be written which are applicable to general configurations without the need of special procedures at the boundaries. It is also possible to distribute the Xllcurvilinear grid lines in the physical plane with concentration of lines in regions of high gradients while maintaining the square grid in the transformed (computational) plane. Once the curvilinear coordinates have been generated, any set of partial differential equations of a physical problem may be solved on this new coordinate system by transforming the equations and the boundary conditions and solving the resulting set of equations. In order to eliminate the difficulties encountered in the solution of transfer equations and to improve the numerical accuracy, the numerical grid generation technique is used in the present study. Elliptic partial differential equations are used to generate grids for several droplet array configurations. All derivatives are approximated by second order central difference. The resulting difference equations are solved by successive point overrelaxation. Since the problem contains some gradients on droplet surfaces, the values of coefficients in the control functions are selected so that grid lines is attracted toward the droplet surfaces. It can not be possible to concentrate the grid lines on droplet surfaces by taking the control functions as zero. The Navier-Stokes equations in vorticity-streamfiinction form, and energy and mass transfer equations are solved numerically. The governing dimensionless equations for the axisymmetric flow field, heat and mass transfer in cylindrical coordinates are as follows: d2 ¥ +- d2 W dx2 dy2 1 dy y dy (1) dco dco v U--+V- G)= dx ay y Re 2 ( d2 CO d2 co 1 dco +- CO dx2 dy2 y dy y (2) BT dT 2 u---+v--=- dx dy RePr ( 32 d2T d2T ]_dT dx2 dy2 y dy +o (3a) y/=-y2 (O=0 7=1 mj =0 (6) ^=0 => /j/=0 ©=0 - =0 ^-=0 (7) dy dy 2 dy dy X=X => 4^=0 42-0 ^=0 ^=0 (9) dx ax ax ox surface=> ^=0 a>=-^-^- T=0 »z,=l (10) ox ay -A,^=(i-«,')n (ii) dn The local heat and mass transfer to the droplet can be expressed in terms of the local Nusselt and Sherwood numbers which are given by *r ~dT or « dm,,`N Nue=2- She=-2- (12) an dn Subsequently, the overall average heat transfer to the whole droplet can be expressed as 1 * 1 K Nu =- J Nug sin 6 dd Sh =-jShe sin 6 dd (13) 2 0 2 0 As for the drag coefficients, the derivations for friction drag and pressure drag can be found in Jenson [43]. They are given by 8 r C`= - / asm2 Odd (14). ReJ0 XIVCdp=jPe smlBde (15) where 4 /( da / p>=p'+rJX^m r (16) In order that the computations can be performed on a rectangular grid in the computational plane, it is convenient to transform the governing equations from the physical domain ( x,y ) to the computational domain ( £,T]). Using the chain rule of partial differentiation the partial derivatives become *, = §,*e+l,*, (17) *` = §`*« + »?,?,, (18) 0» =§**«+ n« 0` + ?/ 0« + nx2 m+Hx nx ^ (19) ^ = ^ 0+ TV ^ + C *« + V 4>m+2Zy % *fr (20) The matrices ( §x, £,, *î,, fy,§«,§`,, i]` » ^ ) appearing in these equations are obtained from the following expressions: -f=x{yv-x1lyf (21) L=^r (22) ^=`^ (23) n,»-^ (24) ny=-j (25) XVk-1 b~ J -y` ( *« ?/ + 2*^ £x T], +xw T]x2 ) (26) ->'l,(xj5^2+2xft^Tjy+xinT]/) (27) *?« = -*« ( y