On Thermal Performance of Seawater Cooling Towers

Seawater cooling towers have been used since the 1970’s in power generation and other industries, so as to reduce the consumption of freshwater. The salts in seawater are known to create a number of operational problems including salt deposition, packing blockage, corrosion, and certain environmental impacts from salt drift and blowdown return. In addition, the salinity of seawater affects the thermophysical properties which govern the thermal performance of cooling towers, including vapor pressure, density, specific heat, viscosity, thermal conductivity and surface tension. In this paper, the thermal performance of seawater cooling towers is investigated using a detailed model of a counterflow wet cooling tower. The model takes into consideration the coupled heat and mass transfer processes and does not make any of the conventional Merkel approximations. In addition, the model incorporates the most up-to-date seawater properties in the literature. The model governing equations are solved numerically and its validity is checked by data in the literature. Based on the results of the model, a correction factor is obtained which characterizes the degradation of the cooling tower effectiveness when seawater is used.

A Planar Labyrinth Micromixer

Rapid mixing of two fluids in microchannels has posed an important challenge to the development of many integrated lab-on-a-chip systems. In this paper, we present a planar labyrinth micromixer (PLM) to achieve rapid and passive mixing by taking advantage of a synergistic combination of the Dean vortices in curved channels, a series of perturbation to the fluids from the sharp turns, and an expansion and contraction of the flow field via a circular chamber. The PLM is constructed in a single soft lithography step and the labyrinth has a footprint of 7.32 mm × 7.32 mm. Experiments using fluorescein isothiocyanate solutions and deionized water demonstrate that the design achieves fast and uniform mixing within 9.8 s to 32 ms for Reynolds numbers between 2.5 and 30. Compared to the mixing in the prevalent serpentine design, our design results in 38% and 79% improvements on the mixing efficiency at Re = 5 and Re = 30 respectively. An inverse relationship between mixing length and mass transfer Pe´clet number (Pe) is observed, which is superior to the logarithmic dependence of mixing length on Pe in chaotic mixers. Having a simple planar structure, the PLM can be easily integrated into lab-on-a-chip devices where passive mixing is needed.

Effect of Surface Conditions on the Adhesion of Crystallization Fouling

In this study, effects of surface conditions in terms of surface roughness and oxide layer, on adhesion of crystallization fouling on heat transfer surfaces were investigated. The experimental results showed that the surface roughness has no obvious effect on the adhesion of crystallization fouling. The polished sample did not present better anti-fouling properties compared to other rough samples. While the formation of Fe2O3 layer on the surface is proved to be able to accelerate the adhesion of calcite fouling with hexagonal structure, because there are similar crystalline structure and lattice parameter between the Fe2O3 and calcite fouling. Therefore, in order to improve the anti-fouling property of heat transfer surfaces, inhibiting the formation of oxide layer is more important than efforts to improve surface roughness.

Performance of Plate Heat Exchangers Used as Refrigerant Liquid-Overfeed Evaporators

The heat transfer and pressure drop characteristics of plate heat exchangers were measured, when used as refrigerant liquid over-feed evaporators. The three units all had 24 plates but with different chevron-angle combinations of 28°/28°, 28°/60°, and 60°/60°. R134a flowing upwards was used as the refrigerant, in a counter-current arrangement with water flowing on the other side. Heat transfer and pressure drop measurements were made over a range of mass flux, heat flux and corresponding outlet vapour fractions. The effect of system pressure on the evaporator performance was not evaluated due to the small range of evaporating temperature. Experimental data were reduced to obtain the refrigerant-side heat transfer coefficient and frictional pressure drop. The results for heat transfer showed a strong dependence on heat flux and weak dependence on mass flux and vapour fraction. Furthermore, the chevron angle had a small influence on heat transfer but a large influence on frictional pressure drops. Along with observations that were obtained previously on large ammonia and R12 plate evaporators, it is concluded that the dominating heat transfer mechanism in this type of evaporator is nucleate-boiling rather than forced convection. For the two-phase friction factor, various established methods were evaluated; the homogeneous treatment gives good agreement.

Numerical Investigation of Heat Transfer for Laminar and Turbulent Flow in a Plate Heat Exchanger

The problem of turbulent flow and heat transfer in a two-channel plate heat exchanger was numerically investigated, considering its complex geometry as well as inlet and outlet ports effects. Results obtained for the flow and thermal field have clearly shown their asymmetrical behavior, which has important influence on the local heat transfer. Friction factor are found to be in good agreement with theoretical correlation.

Falling Film Evaporation of Water on Horizontal Configured Tube Bundles

This paper reports an experimental study on falling film evaporation of water on 6-row horizontal configured tube bundles in a vacuum. Three types of configured tubes, Turbo-CAB-19fpi and −26fpi, Korodense, including smooth tubes for reference, were tested in a range of film Reynolds number from about 10 to 110. Results show that as the falling film Reynolds number increases, falling film evaporation goes from tubes partial dryout regime to fully wet regime; the mean heat transfer coefficients reach peak values in the transition point. Turbo-CAB tubes have the best heat transfer enhancement of falling film evaporation in both regimes, but Korodense tubes’ overall performances are better when tubes are fully wet. The inlet temperature of heating water has hardly any effects on the heat transfer, but the evaporation pressure has controversial effects. A correlation with errors within 10% was also developed to predict the heat transfer enhancement capacity.

Vacuum Operation of a Thermosyphon Reboiler

The research facility at the University of Manchester in the Morton Laboratory is a full scale replica of an industrial sized natural circulation thermosyphon reboiler, which comprises 50 tubes of 3 m length and 25.4 mm OD. The facility is operated under vacuum. Water is used as the process fluid and condensing steam is the heating source. Experimental datasets were obtained for the reboiler and have been presented in the form of profile plots of feed rate, fluid recirculation, recirculation ratio and vapour quality. The data elucidate the effect of pressure [0.1 to 1.0 bar] and heat duties [78 to 930 kW] on the performance of the reboiler. Three distinct modes of operation have been observed. Mode one is defined as a flow-induced instability or geysering (low heat duty) and exists below a definite transitional point that is independent of process pressure. Mode two is a region of stable operation that occurs above the threshold of the flow-induced instability, while mode three, which is defined as the heat-induced instability (density-wave instability), is pressure dependent obtained at high duties and is characterised by violent oscillations. These instability thresholds represent the lower and upper limits of operation of the reboiler. The region of stable operation is enveloped between the two limits and is very dependent on process pressure as it progressively becomes smaller as the vacuum becomes lower. These studies led to unique experimental observations, which revealed the existence of intermittent reversed flow in the entire loop. The use of throttling in the heat-induced unstable region to return to stable operation tends to be over a narrow range, outside of which the sole way to regain stability is to lower the heat load or increase the process pressure. In the region of flow-induced instability, throttling the fluid at the inlet is useless and actually makes the situation worse. These instabilities are alleviated by increasing the heat load.

Simulation of Char Dust Combustion Inside a Pyroscrubber Downstream of a Petroleum Coke Calcining Kiln

A pyroscrubber is a furnace used in the petroleum coke calcining industry to recover energy from the carbonaceous contents, including char dust and hydrocarbon volatiles of the exhaust gas from the calcination kiln. The combusted hot gases are used to generate steam and produce electricity, so it is important to optimize the pyroscrubber performance to produce high-grade combusted gases to generate steam but with minimal emissions. A previous study employed the locally-homogeneous flow (LHF) model to study rhe means to improve combustion efficiency and reduce emissions. In the LHF model, the inter-phase exchange rates of mass, momentum and energy are assumed to be infinitely fast, so the dispersed phase (char dust) can be simplified as the gas phase, and the complex two-phase flow is then treated as a single-phase flow. In this study, LHF model is replaced with a solid particle combustion model by incorporating both finite-rate heterogeneous and homogeneous combustion processes. Results reveal that the particle combustion model generates much higher local flame temperature (2200K) than in LHF model (1800K). All char particles are burned before or in the high-bay area. Total energy output of the case with particle combustion model is 92% of the LHF model. Furthermore, motivated by the potential energy saving from removing the air blower power supply, this study further investigates the possible benefit of running the pyroscrubber with the ventilation doors open. Three cases with different combinations of air injections and door opening have been studied. Results show that the gas flow is stably stratified with a large amount of the entrained cold air moving at the bottom of the chamber and the hot combusted gas moving through the top. With bottom doors completely open, sufficient air can be drawn into the pyroscrubber without the need of blowing air in, but the combustion gases will be overcooled making this practice unfavorable from the energy saving point of view.

1D Heat Transfer Model for the Chain Section of a Lime Kiln

This work was aimed at gaining a better understanding of heat transfer within lime kilns, by both experiments and detailed modeling of heat transfer phenomena in the chain section. Experiments were conducted using a laboratory mockup of a rotating kiln to obtain convective heat transfer coefficient data for cooling of a steel rod in dry or wet lime mud. For moisture contents of 0% and 30%, the mud heat transfer coefficient was determined to be 170 and 320 W/m2°C, respectively. A 1-D, unsteady heat conduction model was used to predict the temperature variations of all the chain rings in the chain system and calculate the amount of heat transferred by each chain ring to the lime mud. A thermal model was then developed to predict the steady axial temperature profiles of lime mud, gas and kiln wall throughout a rotating lime kiln equipped with a typical chain system.

Use of Optimum Seeking Method in Test of Heat Transfer Performance of Finned Tubes

Curve fitting method is often used in the test of heat transfer performance of heat exchangers. The convective heat transfer coefficient may be obtained with the method. However, there are defects with the method. A New Curve fitting method is introduced in this paper. Optimum seeking method is adopted in the New Curve fitting method. The new method is a practical method for test of convective heat transfer coefficient. There are few constraint with the method. Moreover, the results are more precise and the test is more convenient. The method is especially suitable for test of convective heat transfer coefficient of finned tubes outer surface. Two instances of use of the method are given.