University of Southern Queensland (USQ), Toowoomba, QLD, 4350, Australia
Gas dynamic models are used to simulate the entrainment ratio and the compression ratio of ejectors but it is generally necessary to apply ejector-specific calibration factors to the models in order to achieve adequate agreement with experimental data. In an effort to avoid the application of arbitrary calibration factors that can mask multiple deficiencies in the model, data from a range of different ejectors working with air, R141b, and steam are compared to results from a gas dynamic model that does not included any loss factors. This approach facilitates a consistent assessment of ejectors relative to the inviscid gas dynamic model and enables identification of justifiable calibration factors that are applicable across a range of ejectors. Ejectors for which the entrainment ratio has been maximized by adjusting the primary nozzle exit position relative to the ejector throat consistently achieve entrainment ratios that exceed the simulated gas dynamic value. However, the compression ratio of such ejectors falls below the simulated gas dynamic value. The present efforts to correlate ejector operation across a wide range of conditions is showing some promise: a correlation for the entrainment ratio excess is presented using a variable that reflects the capacity for mixing between the primary and secondary streams, and the deficit in the compression ratio is correlated as a linear function of the ejector expansion ratio.