Modeling of an Oil Free Carbon Dioxide Compressor Using Sanderson-Rocker Arm Motion (S-Ram) Mechanism
The multi-piston axial reciprocating compressor using the Sanderson-Rocker Arm Motion (S-RAM) mechanism is expected to have less leakage from the cylinder to the shell space due to the vertical motion of the piston assembly in the cylinder. The main frictional power loss is located at the ball-socket joints connecting the piston with the wobble plate of the S-RAM mechanism. The stroke of the compressor can be changed by altering the inclined angle between the connecting shaft and machine driving shaft, to change the delivered refrigerant mass flow rate to match the capacity requirement in industry. A comprehensive simulation model for a prototype reciprocating compressor using the S-RAM mechanism has been developed. The natural refrigerant carbon dioxide (CO2) is used as the working fluid. The comprehensive model is comprised of a kinematics model, compression process model, dynamics model and an overall energy balance model. In the kinematics model, the movement of the piston is given including its displacement, velocity and acceleration. It is found that the moving path of the center of the ball in the ball-socket joint is moving around a corresponding cylinder centerline with a ‘figure 8’ motion instead of moving along the cylinder centerline. In the compression process model, the system of governing equations is solved, which incorporates a valve sub-model and a leakage sub-model. The variable step size RKF45 method and Broyden’s method are employed to solve the non-linear system of equations to find the in-cylinder refrigerant state (instantaneous temperature and pressure) at each rotational angle of the machine driving shaft. The variations of suction and discharge valve movements with respect to driving shaft rotational angle are also given. The values of the cylinder wall temperature, the actual suction and discharge temperatures in the connecting pipes are required to initiate the solving of the compression process model. These temperatures are solved simultaneously by incorporating the overall energy balance model with the compression process model. A lumped temperature assumption is employed in the overall energy balance model to assume there is no temperature gradient in each compressor component at steady-state. The dynamics model, which focuses on the frictional power loss, runs based on the in-cylinder refrigerant pressure determined in the solution of the compression process model. The presented modeling methodology is validated by modeling a conventional reciprocating piston compressor instead of the S-RAM compressor since there are no test data available for the prototype S-RAM compressor. The simulation model was updated to predict the performance of a conventional reciprocating compressor. Required modifications are made to account for the difference between these two compressors. The simulation results are validated using test data obtained from a hot-gas bypass test stand and the data from the published compressor map. Using this validated methodology, the simulation results for the S-RAM compressor are given. Moreover, the effects of different parameters on the compressor performance are investigated in the parametric studies (such as the effect of the valve spring stiffness and the effect of stroke-to-bore ratio) so that proper parameters are chosen to optimize the compressor.
Groll, Purdue University.
Off-Campus Purdue Users:
To access this dissertation, please log in to our