Hypersorption was recently developed by the Union Oil Company of California. The hypersorption process consists of a moving bed of an adsorbent passing counter currently to the gaseous flow. The gases are separated by selective adsorption. Of basic importance for successful operation is the heating of the adsorbent to obtain desorption. Considerable difficulty in achieving this transfer of heat is encountered because of the non-conducting nature of the adsorbent. The purpose of this investigation was to determine the heat transfer characteristics for the design of an improved heat exchanger for a moving bed system composed of air and activated carbon. This investigation was made studying the transfer of heat to five sizes of activated carbon, from 0.078 inches to fines, at flow rates of 2.2 to 11.0 pounds per hour. Steam at pressures of 15 to 75 pounds per square inch, gage, was used to heat the carbon moving through a three-quarter inch black iron pipe 36 inches long with an effective heating area of 0.65 square foot. The results of this investigation showed that the desorption of moisture in the activated carbon caused an unexpected break in the thermal conductivity curves between 200 and 220 °F, resulting in variable thermal conductivity-specific heat ratios. Rod-like flow was evidenced through e physical test, but poor correlation was observed between the date and the rod-like equation• The over·all heat transfer coefficient varied from one to three Btu per hour—square foot-°F per foot, increasing as the carbon flow rate and the Graetz number, KL/Wcp, increased. / Master of Science

This thesis contributes an in-depth comparative study of u-tube and coaxial borehole heat exchangers. While it is widely accepted that the lower resistance of the coaxial heat exchanger should result in a performance advantage, the findings of several studies comparing the heat exchanger configurations did not definitively establish the mechanisms causing differences in performance. This study employs numerical modelling to consider heat exchangers over a broad range of time scales and under carefully controlled geometry and flow conditions, resulting in the identification of the key parameters influencing borehole heat exchanger performance. The first part of this study consists of a comparison of u-tube and coaxial heat exchangers under continuous loading. A detailed conjugate heat transfer numerical model was developed in OpenFOAM, designed to capture both short and long time scales of heat exchange, necessary to understand the nuanced differences between designs. A novel transient resistance analysis was employed to understand the dominant factors influencing performance. This study established that marginal differences exist between u-tube and coaxial borehole heat exchangers (BHEs) when operated continuously long term but that greater differences occur early in operation. The second phase of this investigation provided a framework for analysing borehole heat exchanger performance during intermittent operation, while also comparing u-tube and coaxial designs. During this study, it was found that reducing operating time, improving the the rate of the ground's recovery to its original temperature, and lowering the duty cycle improved BHE performance. Transit time was identified as a influential time scale, below which heating at the outlet was limited. Further, the benefits of operating below the transit time were mitigated by design-specific interaction between inlet and outlet flows. Finally, this study found that non-dimensionalizing operating time by transit time causes the differences between u-tube and coaxial performance to vanish, leading to the conclusion that differences in BHE performance are caused by variations in flow rather than thermal mass. / Thesis / Doctor of Philosophy (PhD) / This thesis provides an in-depth comparative study of two different designs of borehole heat exchanger, the u-tube and coaxial, which are used in geothermal applications to transfer heat to and from the ground. While many researchers anticipated that the coaxial design would perform better, several studies comparing the heat exchangers were not able to provide a clear answer about which heat exchanger performed best. This study addressed this gap by using detailed numerical simulations which showed that there was a marginal difference in performance between the two heat exchangers when operated for periods longer than a few hours, but that larger differences occurred early in operation (under 15 minutes). The results also showed that operating intermittently resulted in improvements in performance of the heat exchanger, particularly when operated for periods less than the time it takes fluid to travel the length of the piping.

The object of this thesis was to determine water film coefficients and condensing steam film coefficients for a single tube heat exchanger. A shell and tube apparatus was constructed and these coefficients were determined by Wilsons graphical method. Test runs were made at various pressures and water velocities. It was determined that for flow through a horizontal tube the water film coefficient closely approximates 416 V_w^0.8 Btu/hr-ft²-F, where V_w equals water velocity in feet per second, and the Reynolds number lies between 17,000 and 100,000. It was also found that an average condensing steam film coefficient for filmwise condensation was 2000 Btu/hr-ft²-F. It was discovered that this coefficient is much higher if the condensing surface is highly polished. / Master of Science

Metal foams have many attractive properties such as light weight, low relative density, energy absorption capability etc. One of the main advantages of metal foam is that the foam inherits several properties of the parent metal, at the same time, at a fraction of the weight. Metal foams are basically of two types; closed pore and open pore. In the open pore configuration the highly porous structure with large surface to volume ratio is attractive in thermal applications such as heat exchangers, small scale refrigeration, diesel exhaust cooling and heat sink for electronics. Large surface area to volume ratio of the heat transfer area is an important parameter in design of heat exchangers. Application of open cell metal foam as a heat exchanger involves production of the metal foam, cutting/drilling the metal foam to required dimensions and attaching it to a substrate or duct. Foams are cut by various methods such as by using circular saw, band saw, abrasive sawing wire or electrical discharge machining. Cutting or drilling operations plastically deform the struts and affect the surface roughness of the struts and hence, the contact area between the foam and the substrate. The foam and the substrate are then joined to get the final product. Various techniques are adopted to join the foam and substrate that includes, press fit, welding, soldering, brazing and use of epoxy adhesives or thermal glue. These methods either deform the foam plastically or involve a bonding material which involves an additional step in manufacturing and is generally necessary to reduce the thermal resistance at the interface. Every secondary step involved in machining the foam and joining it to substrate/duct add to the energy, time and cost of the component. Significant amount of materials wastage occurs during the production and machining steps of the metal foam. Bonding material used for attaching foam to the substrate makes the recycling of the heat exchangers difficult. In the present research work the above issues were rectified by introducing a novel method of fabricating the heat exchanger in a single step. This can be done by producing open cell foam, bonded to the substrate in a single step to get the ready to use heat exchanger. The uniqueness of the method/ process is that it provides an advantage of manufacturing heat exchangers consisting of open cell aluminium foam both inside and outside the aluminium duct/substrate. Here open cell metal foam is metallurgic ally bonded to the aluminium duct without producing any distortion in the aluminium duct. The present method avoids the secondary cutting and joining operations, hence reducing material and energy wastage. This heat exchanger does not need a bonding material at the foam duct interface which makes the product completely recyclable without even having to separate the aluminium foam and, many-at-times, the copper substrate. Further, in the present process no hazardous material is involved in the fabrication process of the heat exchanger and all the materials used for the foam production can be recycled. Another unique advantage of this process is that the foam can also be cast inside and outside the tube in a single step. This helps increase the heat transfer area per unit volume inside the tube increasing the effectiveness significantly. First, an attempt was made to cast aluminium foam over a Cu substrate. Spheres made of Plaster of Paris (PoP) were used as space holders to create pores in the foam. First, a dough of PoP was prepared by mixing sufficient amount of water with the powder of PoP. Small pieces of PoP were taken from the dough and were rolled by hands to prepare spherical balls. Next, a casting setup was made where a die made of stainless steel was placed in a crucible whose bottom was filled with sand. A tube/duct made of copper was placed at the centre of the die and PoP balls were dropped around the duct. This setup was then placed in a furnace and was preheated to remove all the moisture from the PoP. Molten aluminium at around 700 °C was poured into the preheated die. After solidification, the die was opened and cast was allowed to cool in ambient air. PoP balls were removed by using a sharp needle and by dipping the casting in acetic acid. After removal of PoP from the cast, interconnected holes/cavities formed in the place of space holders/PoP balls, forming pores in the foam. There are some limitations of this method such as removal of PoP was tedious and needed chemicals that need to be discarded, PoP cannot be recycled and creates waste, small amount of moisture present in PoP balls can cause an explosion. The bonding between aluminium foam and Cu substrate obtained was not good, giving rise to thermal contact resistance. Due to the above limitations further implementation of this process using PoP was not explored further. There was a need of space holder material which can withstand the temperature of molten Al and also can be removed easily from the cast without any use of chemicals. Obtaining metallic bonding between foam and Cu substrate was difficult due to the corrosion layer formation at the interface of Al and Cu substrate due to preheating. If preheating was not carried out full penetration of the molten aluminium did not take place in the space available in between the spheres. Therefore, it was decided to cast Al foam over Al substrate. The main challenge and difficulty was to cast open cell Al foam inside and outside the tube/duct made of the same material (Al) without distorting the tube/duct as well as achieving consistent metallic bonding between the two. This has been successfully done by gravity casting method a single step manufactured and ready to use open cell Al foam heat exchanger were fabricated. A casting setup was prepared, which consisted of a commercially pure aluminium tube placed in the middle of a stainless steel split die. The gap between the tube and die was filled with the salt spheres. An uncommon and new approach was adopted to produce NaCl salt spheres. NaCl salt balls (spherical and ovoid) of different diameters were processed by casting route. The casting step of NaCl is necessary as the moisture present in NaCl can be completely removed during the melting of NaCl. NaCl was chosen as it had a melting point higher than aluminium. The casting setup was placed in a furnace and was preheated to various temperatures up to 550 °C. Commercially pure aluminium was melted separately in a crucible and was poured into the steel die at 700oC. The liquid metal flows through the die and fills the cavities between the salt balls. The die was opened immediately after solidification of molten Al and cast was allowed to cool in ambient air. The salt (NaCl), which was still solid, was dissolved in water to get the foam structure. With proper control of the preheat temperature and temperature of liquid aluminium no distortion of the aluminium duct was observed throughout the length of the heat exchanger. Consistent and complete fusion/ metallic bonding was observed at the interface of Al foam and Al substrate/duct. Several heat exchangers with different porosity and pore geometry with the aluminium foam cast outside the tube and both inside and outside of the tube were fabricated. The beauty of the designed method is that it is simple and cost effective and eliminates the major issue of thermal contact resistance since the foam and the duct are made of the same material and are bonded in the liquid state leaving no interface between the foam and the duct. Further, foam can also be cast inside the duct in the same step while casting the foam outside the tube, giving an integral heat exchanger which has higher heat transfer surface area to volume ratio inside and outside the duct. This is expected to further improve the efficiency and effectiveness of the heat exchanger An added advantage of this method is that the heat exchanger can be recycled easily in a single step re-melting route. Further, the heat exchanger does not use any hazardous material during manufacture that needs attention during recycling. After the production and fabrication of the heat exchangers, the thermal performance or effectiveness of the heat exchangers was assessed, to evaluate its usefulness and suitability for heat transfer application. An experimental test setup was fabricated in the laboratory to perform the heat transfer tests. The experimental test setup consists of the following major components;1) A test chamber whose function was to insulate the heat exchangers from the surroundings and to avoid any heat loss to the surroundings, 2) An air blower used to supply cold fluid (air) to the test chamber, 3) A constant temperature bath was used to supply the hot fluid, which was water in this case, in the duct of the heat exchanger, 4) A rotameter was used to measure the volumetric flow rate of the cold fluid and 5) A pressure gauge having the pressure measurement range between 1 mbar to 160 mbar to measure the pressure drop across the test chamber. K-type chromel – alumel thermocouples having temperature measurement range between -270 °C to 1,260 °C were used to measure the temperature of hot and cold fluids during the experiments. By aid of the data logger system and computer, temperature readings were recorded during the tests and were used further for the heat transfer calculations. For testing the aluminium foam heat exchangers was placed in the insulated test chamber. Hot water was supplied inside the duct of heat exchanger whereas air at room temperature was supplied around the foams at varying flow rates during the tests. During the tests, temperature readings were taken at steady state condition. NTU-Effectiveness method was used to evaluate the thermal performance of heat exchangers. Overall results obtained by this experimental study are as follows • As the inlet temperature difference between hot and the cold fluids increases the heat transfer rate and the effectiveness of the heat exchangers also increases. • At a constant flow rate of hot fluid, heat exchangers exhibits significantly better thermal performance at lower flow rate of cold fluid compared to higher flow rate. As the flow rate of cold fluid increases, the velocity of the fluid increases and consequently, reduces the optimum interaction time between hot and the cold fluids required for the efficient heat transfer. • At a constant and low flow rate of cold fluid the effectiveness of the heat exchanger increases as the porosity of the foam increases. But when the flow rate of cold fluid was increased further after a certain limit, the effectiveness value of the heat exchanger decreases. • Heat exchanger consisting of foam of higher porosity exhibits higher effective. • Heat exchanger having foam inside and outside of the duct/tube exhibits significantly higher effectiveness compared to Al duct, Cu duct and other heat exchanger tested. • At a higher flow rate of the cold fluid, the heat exchangers consisting of foams of higher porosity, experience more drop in effectiveness compared to the heat exchanger having foams of low porosity. • Pressure drop across the length of the foam/fin increases as the volumetric flow rate of the cold fluid (m3/s) increases. • Surface area per unit volume and effectiveness values for bare Al tube is very low compared to Al foam heat exchangers resulting in the bare Al tube exhibiting much lower effectiveness compared to heat exchanger made of Al foam. • For a certain flow rate of fluids, the effectiveness of the heat exchanger increases up to a certain thickness of the Al foam. • Regardless of the thickness of the foam, the effectiveness of the heat exchangers is low at higher flow rate of cold fluid compared to lower flow rate. • These foam based heat exchanger had a much higher effectiveness when compared to that of other heat exchangers, data of which were got from literature. The present experimental study concludes that fuse bonding open cell aluminium foam over an Al duct or Al substrate can improve the thermal performance of the heat exchanger significantly. The thesis includes five chapters. Chapter 1 gives a detailed introduction about the metal foam, heat exchangers, thermal contact resistance and its effect on the heat transfer rate has been explained. This chapter also includes the overall aim and motivation for the research work. Chapter 2 covers the literature available on production methods of metal foam and its limitations has been listed out. And conventional methods of manufacturing open cell metal foam heat exchangers and its disadvantages have been explained in detailed. Chapter 3 covers in detail the novel method of production and fabrication of open cell metal foam heat exchangers. Chapter 4 includes an experimental study, where thermal performance of heat exchangers has been assessed through heat transfer experiments. Chapter 5 is the conclusions and future works.

Call number: LD2668 .T4 1979 H43 / Master of Science

Thesis (MScIng)--University of Stellenbosch, 2004. / ENGLISH ABSTRACT: As the demand for less expensive energy is increasing world-wide, energy conservation is becoming a more-and-more important economic consideration. In light of this, means to recover energy from waste fluid streams is also becoming more-and-more important. An efficient and cost effective means of conserving energy is to recover heat from a low temperature waste fluid stream and use this heat to preheat another process stream. Heat pipe heat exchangers (HPHEs) are devices capable of cost effectively salvaging wasted energy in this way. HPHEs are liquid-coupled indirect transfer type heat exchangers except that the HPHE employs heat pipes or thermosyphons as the major heat transfer mechanism from the high temperature to the low-temperature fluid. The primary advantage of using a HPHE is that it does not require an external pump to circulate the coupling fluid. The hot and cold streams can also be completely isolated preventing cross-contamination of the fluids. In addition, the HPHE has no moving parts. In this thesis, the development of a range of air-to-air HPHEs is investigated. Such an investigation involved the theoretical modelling of HPHEs such that a demonstration unit could be designed, installed in a practical industrial application and then evaluated by considering various financial aspects such as initial costs, running costs and energy savings. To develop the HPHE theoretical model, inside heat transfer coefficients for the evaporator and condenser sections of thermosyphons were investigated with R134a and Butane as two separate working fluids. The experiments on the thermosyphons were undertaken at vertical and at an inclination angle of 45° to the horizontal. Different diameters were considered and evaporator to condenser length ratios kept constant. The results showed that R134a provided for larger heat transfer rates than the Butane operated thermosyphons for similar temperature differences despite the fact that the latent heat of vaporization for Butane is higher than that of R134a. As an example, a R134a charged thermosyphon yielded heat transfer rates in the region of 1160 W whilst the same thermosyphon charged with Butane yielded heat transfer rates in the region of 730 W at 23 °C . Results also showed that higher heat transfer rates were possible when the thermosyphons operated at 45°. Typically, for a thermosyphon with a diameter of 31.9 mm and an evaporator to condenser length ratio of 0.24, an increase in the heat transfer rate of 24 % could be achieved. Theoretical inside heat transfer coefficients were also formulated which were found to correlate reasonably well with most proposed correlations. However, an understanding of the detailed two-phase flow and heat transfer behaviour of the working fluid inside thermosyphons is difficult to model. Correlations proposing this behaviour were formulated and include the use of R134a and Butane as the working fluids. The correlations were formulated from thermosyphons of diameters of 14.99 mm, 17.272 mm, 22.225 mm and 31.9 mm. The evaporator to condenser length ratio for the 31.9 mm diameter thermosyphon was 0.24 whilst the other thermosyphons had ratios of 1. The heat fluxes ranged from 1800-43500 W/m2. The following theoretical inside heat transfer coefficients were proposed for vertical and inclined operations (READ CORRECT FORMULA IN FULL TEXT ABSTRACT) φ = 90° ei h = 3.4516x105Ja−0.855Ku1.344 φ = 45° ei h = 1.4796x105Ja−0.993Ku1.3 φ = 90° l l l ci l l v h x k g 1/ 3 2.05 2 4.61561 109Re 0.364 ν ρ ρ ρ − ⎡ ⎡ ⎛ ⎞⎤ ⎤ = ⎢ ⎢ ⎜ ⎟⎥ ⎥ ⎢ ⎢ ⎜ − ⎟⎥ ⎥ ⎣ ⎣ ⎝ ⎠⎦ ⎦ φ = 45° l l l ci l l v h x k g 1/ 3 1.916 2 3.7233 10 5Re 0.136 ν ρ ρ ρ − ⎡ ⎡ ⎛ ⎞⎤ ⎤ = ⎢ ⎢ ⎜ ⎟⎥ ⎥ ⎢ ⎢ ⎜ − ⎟⎥ ⎥ ⎣ ⎣ ⎝ ⎠⎦ ⎦ The theoretically modelled demonstration HPHE was installed into an existing air drier system. Heat recoveries of approximately 8.8 kW could be recovered for the hot waste stream with a hot air mass flow rate of 0.55 kg/s at an inlet temperature of 51.64 °C and outlet temperature of 35.9 °C in an environment of 20 °C. Based on this recovery, energy savings of 32.18 % could be achieved and a payback period for the HPHE was calculated in the region of 3.3 years. It is recommended that not withstanding the accuracies of roughly 25 % achieved by the theoretically predicted correlations to that of the experimental work, performance parameters such as the liquid fill charge ratios, the evaporator to condenser length ratios and the orientation angles should be further investigated. / AFRIKAANSE OPSOMMING: As gevolg van die groeiende aanvraag na goedkoper energie, word die behoud van energie ‘n al hoe belangriker ekonomiese oorweging. Dus word die maniere om energie te herwin van afval-vloeierstrome al hoe meer intensief ondersoek. Een effektiewe manier om energie te herwin, is om die lae-temperatuur-afval-vloeierstroom (wat sou verlore gaan) se hitte te gebruik om ‘n ander vloeierstroom mee te verhit. Hier dien dit dan as voorverhitting van die ander, kouer, vloeierstroom. Hittepyp hitteruilers (HPHR’s) is laekoste toestelle wat gebruik kan word vir hierdie doel. ‘n HPHR is ‘n vloeistof-gekoppelde indirekte-oordrag hitteruiler, behalwe vir die feit dat dié hitteruiler gebruik maak van hittepype (of hittebuise) wat die grootste deel van sy hitteoordragsmeganisme uitmaak. Die primêre voordele van ‘n HPHR is dat dit geen bewegende dele het nie, die koue- en warmstrome totaal geïsoleer bly van mekaar en geen eksterne pomp benodig word om die werkvloeier mee te sirkuleer nie. In hierdie tesis word ‘n ondersoek gedoen oor die ontwikkeling van ‘n bestek van lug-totlug HPHR’s. Hierdie ondersoek het die teoretiese modellering van so ‘n HPHR geverg, sodat ‘n demonstrasie eenheid ontwerp kon word. Hierdie demonstrasie eenheid is geïnstalleer in ‘n praktiese industriële toepassing waar dit geïvalueer is deur na aspekte soos finansiële voordele en energie-besparings te kyk. Om die teoretiese HPHR model te kon ontwikkel, moes daar gekyk word na die binnehitteoordragskoëffisiënte van die verdamper- en kondensordeursneë, asook R134a en Butaan as onderskeie werksvloeiers. Die eksperimente met die hittebuise is gedoen in die vertikale en 45° (gemeet vanaf die horisontaal) posisies. Verskillende diameters is ook ondersoek, maar met die verdamper- en kondensor-lengteverhouding wat konstant gehou is. Die resultate wys dat R134a as werksvloeier in die hittebuise voorsiening maak vir groter hitteoordragstempo’s in vergelyking met Butaan as werksvloeier by min of meer dieselfde temperatuur verskil – dít ten spyte van die feit dat Butaan ‘n hoër latente-hittetydens- verdampings eienskap het. As voorbeeld gee ‘n R134a-gelaaide hittebuis ‘n hitteoordragstempo van omtrent 1160 W terwyl dieselfde hittebuis wat met Butaan gelaai is, slegs ongeveer 730 W lewer by 23 °C. Die resultate wys ook duidelik dat hoër hitteoordragstempo’s verkry word indien die hittebuis bedryf word teen ‘n hoek van 45°. ‘n Tipiese toename in hitteoordragstempo is ongeveer 24 % vir ‘n hittebuis met ‘n diameter van 31.9 mm en ‘n verdamper- tot kondensor-lengteverhouding van 0.24. Teoretiese binne-hitteoordragskoëffisiënte is ook geformuleer. Dié waardes stem redelik goed ooreen met die meeste voorgestelde korrelasies. Nieteenstaande die feit dat gedetailleerde twee-fase-vloei en die hitteoordragsgedrag van die werksvloeier binne hittebuise nog nie goed deur die wetenskaplike wêreld verstaan word nie. Korrelasies wat hierdie gedrag voorstel is geformuleer en sluit weereens die gebruik van R134a en Butaan as werksvloeiers in. Die korrelasies is geformuleer vanaf hittebuise met diameters van onderskeidelik 14.99 mm, 17.272 mm, 22.225 mm en 31.9 mm. Die verdamper- tot kondensor-lengteverhoudings vir die 31.9 mm deursnit hittebuis was 0.24 terwyl die ander hittebuise ‘n verhouding van 1 gehad het. Die hitte-vloede het gewissel van 1800-45300 W/m2. Die volgende teoretiese geformuleerde binne-hitteoordragskoëffisiënte word voorgestel vir beide vertikale sowel as nie-vertikale toepassing (LEES KORREKTE FORMULE IN VOLTEKS OPSOMMING) φ = 90° ei h = 3.4516x105Ja−0.855Ku1.344 φ = 45° ei h = 1.4796x105Ja−0.993Ku1.3 φ = 90° l l l ci l l v h x k g 1/ 3 2.05 2 4.61561 109Re 0.364 ν ρ ρ ρ − ⎡ ⎡ ⎛ ⎞⎤ ⎤ = ⎢ ⎢ ⎜ ⎟⎥ ⎥ ⎢ ⎢ ⎜ − ⎟⎥ ⎥ ⎣ ⎣ ⎝ ⎠⎦ ⎦ φ = 45° l l l ci l l v h x k g 1/ 3 1.916 2 3.7233 10 5Re 0.136 ν ρ ρ ρ − ⎡ ⎡ ⎛ ⎞⎤ ⎤ = ⎢ ⎢ ⎜ ⎟⎥ ⎥ ⎢ ⎢ ⎜ − ⎟⎥ ⎥ ⎣ ⎣ ⎝ ⎠⎦ ⎦ Die wiskundig-gemodelleerde demostrasie HPHR is geïnstalleer binne ‘n bestaande lugdroër-sisteem. Drywing van om en by 8.8 kW kon herwin word vanaf die warm-afvalvloeierstroom met ‘n massa vloei van 0.55 kg/s teen ‘n inlaattemperatuur van 51.64 °C en ‘n uitlaattemperatuur van 35.9 °C binne ‘n omgewing van 20 °C. Na aanleiding van hierdie herwinning, kan energiebesparings van tot 32.18 % verkry word. Die HPHR se installasiekoste kan binne ‘n berekende tydperk van ongeveer 3.3 jaar gedelg word deur hierdie besparing. Verdamper- tot kondensator-lengteverhouding, vloeistofvulverhouding en die oriëntasiehoek vereis verdere ondersoek, aangesien daar slegs ‘n akkuraatheid van 25 % verkry is tussen teoretiese voorspellings en praktiese metings.

Thesis (MScEng) -- Stellenbosch University, 2004. / ENGLISH ABSTRACT: Around the world the implementation of heat recovery systems is playing an increasingly important role in the engineering inqustry. The recovered energy is utilised in the plants and saves companies millions in expenses per year. Not only is this seen on the grand scale of industry, but also in everyday life, where for instance turbochargers are used to boost the performance of automobiles by utilising the wasted energy expelled along with exhaust gasses. The aim of this project is to investigate a small scale waste heat recovery system, and to determine the optimum method by which to convert the recovered energy into electrical energy, which can be used as a secondary energy source. The research contained in this thesis, centres on the main components and theory needed for the construction of a small scale waste heat recovery system. Also included, is a theoretical analysis concerning the design and construction of the system, utilising researched theory and a simulation program of the recovery system. The simulation is control volume-based and generates property data on the fluid and exhaust gas throughout the heat exchanger. The final design included a finite element stress analysis of certain parts of the system to ensure safe testing at high pressures and temperatures. The final design resulted in a high pressure, cross flow, stainless steel fintube heat exchanger that, by using a continuous combustion unit as energy source and water as the working fluid, reached efficiencies of up to 74% in direct steam generation testing. The tube-side of the heat exchanger was designed to withstand pressures of up to 2MPa (20bar), which is imperative for the implementation of the next phase, where a turbocharger will be connected to the heat exchanger. The completion of this part of the project has paved the way for further development and implementation of the heat recovery system. / AFRIKAANSE OPSOMMING: Die herwinning van energie begin 'n toenemend belangrike rol in die ingenieurs industrie speel. Die herwonne energie word in fabrieke ben ut en spaar maatskappye milj oene aan uitgawes per jaar. Hierdie beginsel word nie net in die grootskaalse nywerhede toegepas nie, maar ook in die allerdaagse lewe, soos byvoorbeeld in voertuie waar turbo-aanjaers gebruik word om die energie-uitset van enjins te verhoog deur bloot gebruik te maak van die verlore energie wat saam met die uitlaatgasse in die atmosfeer gepomp word. Die doel van hierdie projek is om 'n kleinskaalse energieherwinningstelsel te ondersoek en die mees effektiewe metode te vind om die herwinde energie na elektriese energie om te skakel wat as 'n sekondere energiebron gebruik kan word. Die navorsing bevat in die tesis, kyk na al die hoofkomponente en teoretiese kennis wat nodig is vir die konstruksie van 'n kleinskaalse hitteherwinningstelsel. Ook ingesluit is 'n teoretiese analise ten opsigte van die ontwerp en konstruksie van die sisteem. Dit behels die gebruik van nagevorsde teorie saam met 'n simulasie program van die herwinnings stelsel. Die simulasie program is op kontrole volumes gebasseet en genereer uitlaatgas- en water eienskappe soos dit deur die hitteruiler vloei. Die finale ontwerp bevat 'n eindige element spannmgs analise van sekere kritiese komponente in die stelsel om die veilige gebruik van die sisteem by hoe drukke en temperature te verseker. Die finale ontwerp was 'n hoedruk, kruisvloei, vlekvrye staal finbuis hitteruiler. Deur 'n konstante verbrandingseenheid as energiebron te gebruik saam met water as werksvloeier, het die hitteruiler effektiwiteite van tot 74% in direkte stoomgenerasie-toetse bereik. Die hitteruiler is ontwerp om hoe drukke van tot 2MPa (20bar) te hanteer wat baie belangrik is vir die implementasie van die volgende fase van die projek waar 'n turbo-aanjaer aan die stelsel gekoppel sal. Die suksesvolle voltooiing van hierdie fase van die projek het die weg gebaan vir die verdere ontwikkeling en implimentasie van die energieherwinningsstelsel.

Thesis (MScEng)--University of Stellenbosch, 2000. / ENGLISH ABSTRACT: Measurements on air-cooled reflux steam condensers or dephlegmators at different power plants have shown that sections of these units do not transfer heat effectively over a range of operating conditions. The ineffective sections may be due to flooding in the finned tubes although entrainment of condensate in certain steam inlet manifolds is usually the main reason for the poor performance. In this dissertation factors that limit effective dephlegmator operation are discussed and the influence of two inlet manifold designs on dephlegmator operation is investigated. Laboratory experiments are conducted to show under which conditions liquid entrainment occurs and to visualize the flow distribution within the respective manifolds. An alternative, essentially horizontal arrangement of the dephlegmator is proposed. In order to evaluate the performance of such a system, the heat transfer and pressure drop on the steam-side is determined experimentally in an air-cooled finned tube. No flooding was observed during tests conducted at zero and negative tube angles to the horizontal. / AFRIKAANSE OPSOMMING: Toetse op lugverkoelde stoom terugvloeikondensors, of deflegmators, by verskeie kragstasies het getoon dat sekere dele van hierdie eenhede onder verskeie werkstoestande nie warmte effektief oordra nie. Hierdie oneffektiewe dele kan deur vloeding van die vinbuise veroorsaak word alhoewel die meesleur van kondensaat in sekere stoom inlaatspruitstukke gewoonlik die hoof oorsaak is. In hierdie dissertasie word faktore wat effektiewe deflegmator werksverrigting beinvloed bespreek en die invloed van twee inlaatspruitstukontwerpe op deflegmator werksverrigting ondersoek. Eksperimente is in 'n laboratorium uitgevoer om aan te toon onder watter werkstoestande vloeistof samesleping voorkom en om vloeiverdeling binne die onderskeie inlaatspruitstukke te visualiseer. 'n Altematiewe, wesenlike horisontale deflegmator opstelling word voorgestel. Die werksverrigting van hierdie voorstelling is ondersoek deur die warmteoordrag en stoorn-kant drukval eksperimenteel te bepaal in 'n lugverkoelde vinbuis. Geen vloeding is opgemerk vir toetsgevalle waar klein negatiewe of zero hoeke tot die horisontaal ondersoek is nie.

Thesis (PhD (Mechanical and Mechatronic Engineering))--University of Stellenbosch, 1992. / The effect of cross-winds on the performance of natural draft dry-cooling towers is studied by means of isothermal model tests, a numerical simulation and full scale measurements. The action of the wind on such towers is found to be complex and is influenced by a number of different parameters including the wind speed, the shape of the approaching wind profile, the inlet diameter to the inlet height ratio of the tower, the tower height, the shape of the tower shell, the pressure loss coefficient of the heat exchangers and the amount of heat rejected by the tower. For a horizontal arrangement of the heat exchangers the wind effect on the tower is shown to be strongly dependent on both the shape and pressure loss coefficient of the tower supports. In practical cooling towers the heat exchangers are either arranged horizontally in the inlet cross-section of the tower or vertically around the circumference of the tower and the wind effect is found to be dependent on the particular layout. The wind effect on a tower is furthermore found to increase if the heat exchangers are arranged in the form of A-frames. Additional reductions in the heat rejection rate of the tower are caused by a non-uniform air temperature distribution inside the tower and flow distortions through the heat exchanger. Significant reductions in the wind effect on a cooling tower can be achieved by installing windbreak walls below the heat exchangers if the latter are arranged horizontally in the tower inlet.

Thesis (PhD (Mechanical and Mechatronic Engineering)--University of Stellenbosch, 1990. / 260 leaves printed single pages, preliminary pages i-xxi and numbered pages Chapter 1/1.1-1.3, Chapter 2/2.1-2.17, Chapter 3/3.1.1-3.10, Chapter 4/4.1-4.18, Chapter 5/5.1-5.3, References pages R.1-R.7, Appendix A pages A.1-A.34, Appendix B pages B.1-B.34, Appendix C pages C.1-C.26, Appendix D pages D.1-D.16, Appendix E pages E.1-E.30, Appendix F pages F.1-F.39. Includes bibliography, list of tables, figures and symbols. / Digitized at 600 dpi grayscale to pdf format (OCR), using a Bizhub 250 Konica Minolta Scanner. / ENGLISH ABSTRACT: The ultimate goal of this project is to ensure a better understanding of the governing mechanisms present when flow distorting components are installed in close proximity of an axial flow fan. The effect of different parameters on the operation of axial flow fans is investigated. These parameters are divided into flow enhancing and flow reduction effects. The performance of an axial flow fan can be enhanced by changing the tip clearance, by adding a solid disc to the hub of the fan or by varying the number of fan blades. Flow reductions are caused by components such as inlet grids, walkways and their supporting structures, heat exchangers and windwalls. The effects of flow enhancing components are measured and compared to the results of other authors. The sensitivity of these effects to parameters such as the type of fan rotor and the specific system in which the rotor is installed is highlighted. The system effect (the interaction between the fan rotor and flow resistances in close proximity of each other) of individual components, as well as the combination of different components, is predicted both theoretically and experimentally. These predictions are compared to measured data relevant to the components in an installation where the system effects are present. The results are correlated to the kinetic energy flux coefficient of the flow at different locations within the installation. Experimental data obtained from a full scale unit (inlet shroud diameter of 9,216 m) are used to compare to scaled data from the model (inlet shroud diameter of 1,542 m). The hub to tip ratio of the axial flow fans investigated is 0,15. The most important conclusions are that the performance of the type of axial flow fan under investigation can be improved by reducing its tip clearance and by installing a solid disc to the downstream side of the rotor. An increase in the number of blades of the fan leads to only marginal improvements in the fan performance. The overall performance of the system can also be improved by removing some of the flow resisting components, or by changing their relative positions. All these conclusions are based on the assumption that the power input to the fan rotor remains constant. / AFRIKAANSE OPSOMMING: Die uiteindelike doel van hierdie projek is om te verseker dat die beherende meganismes wat teenwoordig is wanneer vloeiversteurende komponente in die nabyheid van 'n aksiaalwaaier geinstalleer word, beter verstaan word. Die effek van verskillende parameters op die werkverrigting van aksiaalwaaiers word ondersoek. Hierdie parameters word verdeel in vloeiverbeterings- en vloeiverminderingseffekte. Die werkverrigting van 'n aksiaalwaaier kan verbeter word deur die lempuntspeling te verstel, deur 'n soliede skyf aan die naaf van die waaierrotor te installeer, of deur die aantal lemme te verander. Die vloeiverminderings word veroorsaak deur inlaatsiwwe, loopvlakke en hul ondersteuningsstrukture, warmteruilers en windwande. Die effekte van vloeiverbeteringskomponente word gemeet en vergelyk met die resultate van ander outeurs. Die sensitiwiteit van hierdie effekte op parameters soos die tipe rotor en die spesifieke stelsel waarin die rotor geinstalleer is, word uitgelig. Die stelseleffek (die interaksie tussen die rotor van die waaier en vloei weerstande wat naby mekaar geinstalleer is) van individuele, sowel as 'n kombinasie van verskillende komponente, word teoreties en eksperimenteel voorspel. Hierdie voorspellings word dan vergelyk met eksperimentele data wat van toepassing is op die komponente in 'n installasie waar stelseleffekte voorkom. Die resultate word gekoppel aan die kinetiese energievloedkoeffisient van die vloei by verskillende posisies binne die installasie. Eksperimentele data, verkry vanaf 'n volskaaleenheid (inlaatmondstukdiameter van 9,216 m), word met die geskaleerde data van die model (inlaatmondstukdiameter van 1,542 m) vcrgelyk. Die naaf- tot hulsverhouding van die aksiaalwaaiers wat ondersoek word is 0,15. Die belangrikste gevolgtrekkings is dat die werkverrigting van die tipe aksiaalwaaier wat ondersoek word verbeter kan word deur die lempuntspeling te verminder en deur 'n soliede skyf te installeer by die stroomaf kant van die rotor. 'n Toename in die aantal lemme van die waaier lei slegs tot marginale verbeterings in die werkverrigting van die waaier. Die totale werkverrigting van die stesel kan ook verbeter word deur sommige vloeiweerstandskomponente te verwyder, of deur hulle relatiewe posisies te verander. Al hierdie gevolgtrekkings is gebasseer op die aanname dat die drywingsinset na die waaierrotor konstant bly.