CFD modelling and analysis of variable and constant wall thickness scroll expanders and their performance.

Higher efficiencies and more compact designs in spite of larger expansion ratios are associated with variable wall thickness scroll expander geometries. However, the literature for these innovative scroll designs is mainly limited to theoretical studies since the research and development of scroll expanders is still in the early stages. Considering the potential benefits of less overall leakage areas, shorter residence time of the gas and less time for leakages and heat transfer, the scroll machine with variable wall thickness could be a promising candidate to further improve the efficiency and power output in organic Rankine cycle systems. It has also the opportunity of opening up new application fields such as hybrid vehicle powertrain systems in which power cycles with high pressure ratio are needed. The main aims of this PhD research were to investigate variable and constant wall thickness scroll expanders for small scale organic Rankine cycle systems using three-dimensional and transient CFD simulations. The aerodynamic performances are compared to examine the features of the novel variable wall thickness scroll expander design. This is followed by the investigation of spark-ignition scroll engines and their performance by means of a heat release rate analysis and CFD based combustion modelling tools. The evaluation of the CFD simulations of variable wall thickness scroll expanders reveals that the geometrical effects of varying wall thicknesses did not affect the characteristic scroll machine operation. The validation, verification and the findings had proven consistency with the theory of scroll expanders. The optimum performance was achieved at a pressure ratio of 3.5 regardless of the rotational speed. The decrease of radial clearance from 200μm to 75μm had a positive effect on isentropic efficiency and specific power output. The isentropic efficiency at the optimum performance point was significantly improved by 22% from 31.9% to 53.9%. It is also found that the lower number of working chambers resulted in a shorter gas residence time, associated with less time for flank leakages, in comparison to the constant wall thickness scroll expander. Thus, the fluid friction was reduced, converting less kinetic energy into enthalpy. The large-scale swirls were completely dissipated in the expansion chambers of the variable wall thickness scroll expander at the crank angle of 600°, in contrast to 672°in the expansion chambers of the constant wall thickness design. In addition, the shorter scroll profile length of the variable wall thickness scroll expander generated lower average radial and axial gas forces. Moreover, higher pressure gradients between individual working chambers contributed to a higher peak of the tangential gas moment but at the expense of higher transient radial and axial gas force and tangential gas moment variations. More significant pressure drops occurred along the local radial clearance reducing the isentropic efficiencies in spite of the shift towards higher pressure ratio. The heat release rate analysis reveals that a more thorough expansion was achieved by employing the scroll engine in the Miller/Atkinson cycle instead of the conventional Otto cycle. The highest power output of 44.5kW was achieved for a compression ratio of 10.1:1 and an expansion ratio of 17.8:1 (V=4.62dm^3) at a rotational speed of n=3000rpm. The thermal efficiency followed the same trend reaching a peak value of 43.1% but for a lower compression ratio of 8.2:1. The evaluation of the CFD based combustion model results shows that the third combustion cycle was technically not feasible because the entire domain was filled with burned gas due to the lack of flame quenching. No steady-state solution was achieved and all the results are therefore hypothetical