Leakage modelling of revolving vane expander

• Main Author: Hans, Hendrik.
• Other Authors: Ooi Kim Tiow
• Format: Final Year Project
• Language: English
• Published: 2010
• Subjects: DRNTU::Engineering::Mechanical engineering::Fluid mechanics
• Online Access: http://hdl.handle.net/10356/39775
• Institution: Nanyang Technological University

Conventionally, before arriving to evaporator and after the condenser, the refrigerant gas will pass through a throttling valve which will cause a decrease in the pressure of the refrigerant gas. Now, we are trying to include the use of expander so that, instead of the refrigerant gas being expanded isenthalpically (without extracting any work), it will be expanded more towards isentropically. Hence, the energy that we obtained from the expansion process could be supplied to the compressor, decreasing the demand for electricity of the compressor. Through preliminary analysis, the COP could be increased by up to 60%, this translates to savings in energy of up to 37%. Therefore, we decided that detailed analysis would be very much of interest in order to evaluate the performances of the expander. And one of the main problems with rotary type expanders is leakage. Therefore, the leakage modeling becomes the main issue to be resolved. For most of the time, researchers on leakage always assumed that radial leakage is more important than endface leakage and more prominent, therefore they made assumptions that endface leakage is negligible as compared to radial leakage. So to this date, no work on endface leakage modeling is developed. The farthest researchers have gone, is represented by the work of Yanagisawa and Shimizu[10]. But the modeling done in [10] is leakage of lubricant from the shaft to the endface that mixed with the refrigerants hence we can say that no work up to this date has been done on the endface leakage from high pressure chamber to low pressure chamber. As for the radial leakage modeling, there are quite a number of models developed and one of the most well-established models is done by Yanagisawa and Shimizu in [8]. But in that model, there’s an inherent flaw both on his experimental setup and on his theoretical modeling. Hence, we perform in modeling in a different manner with results that converge well within his range of results. We require 8 models to completely describe all the leakages under different circumstances. This is due to the fact that the main leakages on the expander are radial leakage and endface leakage. The endface leakage is later divided into rotor endface leakage and leakage through vane endface. For each of these leakages, we establish choked flow model, unchoked compressible flow model and unchoked incompressible flow model, except for radial leakage.