Development of an oil-free swing vane compressor with liquid refrigerant injection

A conventional compressor in a vapour-compression refrigeration system often requires the extensive use of active oil lubrication to ensure reliable operations. The use of active oil lubrication, however, has resulted in the low portability and poor compactness of current compressors because the lubrication system often requires components such as the oil filter, oil pump, oil cooler and oil sump. In addition, these compressors must always be operated in the upright orientation to ensure that the oil sump is always located at the lowest position to allow the lubricant to flow back to the oil sump by gravity. This is required to maintain a functional lubrication system that provides a continuous re-circulation of lubrication flow. As a result, the conventional vapour-compression refrigeration system has limited cooling and heating applications since the orientation of the compressor is fixed. The current refrigeration system that relies on oil-lubricated compressor is not feasible for applications such as wearable cryogenic cooling bags for the ease of vaccine transportation or compact cryogenic coolers for space applications. With the successful development of an oil-free compressor, the vapour-compression refrigeration system would be able to operate freely under any orientation and become an alternative for small-scale portable cooling and heating applications. This work aims to develop a simple positive-displacement vane type compressor for oil-free operation. Potential key failure modes and the performance characteristics of an oil-free compressor were determined theoretically and experimentally. Without oil lubrication, an oil-free compressor would easily get overheated and seizure may occur. To resolve this issue, liquid refrigerant injection cooling was used in the new compressor design to improve its reliability. In this study, an oil-free swing vane compressor prototype was designed and its performance was evaluated through experimental testing with air and R134a as the working fluid. Theoretical models were formulated to predict its operational characteristics. The models developed include the mathematical descriptions of the geometry and kinematics of the swing vane mechanism, the thermodynamics analysis of the working cycle, the dynamic analysis of the moving parts and valve vibrations, the leakage flow analysis and the heat transfer study. In the theoretical analysis of the liquid refrigerant injection cooling, a two-phase orifice flow model was also used to predict the injected mass flow of the refrigerant and the injected liquid refrigerant was assumed to evaporate instantaneously after absorbing heat from the compression chamber. Experimental studies of the oil-free prototype were conducted at operating speeds ranging from 1080 to 1800 rev min^-1 with air as the working fluid. Compression ratios between 2.11 and 4.08 with mechanical efficiency ranging from 19.3% to 30.3% and volumetric efficiency ranging from 30.9% to 67.5% were achieved. The overall isentropic efficiency achieved from these experiments varies from 0.52 to 0.71. Predicted results from the theoretical models were compared with measured data and the maximum discrepancy of each measured output was found to be less than 15%. Experimental tests of the prototype in a vapour-compression cycle with R134a were conducted at an operating speed of 1620 rev min^-1 and the prototype achieved a mechanical efficiency of 35.6% and a volumetric efficiency of 55.5%. The coefficients of performance (COPs) measured and predicted were 1.25 and 1.28, respectively. The low COPs values were largely due to substantial internal leakages in the prototype in the absence of oil lubrication. Liquid refrigerant was also injected into the compression chamber of the prototype for durations of 0.005s and 0.01s to cool the compressor mechanism and the performance of the prototype was analysed. It was found that the injected liquid refrigerant was able to cool the compressor significantly without increasing the compression power and decreasing the compressor efficiency. No significant improvement to the COP for operations under effects of liquid injection cooling was observed. However, the surface temperature of the compressor was cooled by 6.5 to 8.1 deg C while only a small additional 2.4% increment in the average power input was measured. With a liquid refrigerant injection duration of 0.01s, the isentropic efficiency was predicted to improve from 0.70 to 0.90 due to lesser compression work required. Overall, the conceptual prototype of an oil-free swing vane compressor was successfully tested experimentally. As compared to other existing oil-free refrigeration compressors, the COPs achieved in the experiments were low. Higher COPs can be achieved with better fabrication accuracy to reduce internal leakages and with better self-lubricating materials with higher wear resistance and lower friction coefficient.Doctor of Philosoph