Study on the Use of Springs in a Dual Free Piston Engine Alternator

The free piston engine combined with a linear electric alternator has the potential to be a highly efficient converter from fossil fuel energy to electrical power. With only a single major moving part (the translating rod), mechanical friction is reduced compared to conventional crankshaft technology. Instead of crankshaft linkages, the motion of the translator is driven by the force balance between the engine cylinder, alternator, damping losses, and springs. Focusing primarily on mechanical springs, this paper explores the use of springs to increase engine speed and reduce cyclic variability. A numeric model has been constructed in MATLAB^®/Simulink to represent the various subsystems, including the engine, alternator, and springs. Within the simulation is a controller that forces the engine to operate at a constant compression ratio by affecting the alternator load. The complex interdependence of the free piston engine alternator is analyzed with respect to parametric changes to the spring stiffness. For a fixed compression ratio, it is shown that an increase in spring stiffness from 50 to 350 kN/m (which practically must be associated with an increase in total moving mass) raises system frequency (18%) and power (12%), but can also lead to a relatively small loss of system efficiency (2%). This is due to the decrease of charging efficiency (EGR increased by 12%) for fixed intake/exhaust conditions and higher frictional losses (4%). The gain in system frequency and power output is diminished according to the increased moving mass associated with stiffer springs. This study also investigates the ability of springs to dampen cyclic variation in response to combustion variation. Normally distributed noise is added to combustion efficiency and duration. Coefficients of variation of compression ratio and peak pressure are used to represent cycle to cycle variation response and compared for varied spring stiffnesses. It is shown that the stiff springs can be used to dampen the effects of combustion stochastics and the resulting variation brought on by cylinder pressure variation. This results in lower controller demand and higher operational sustainability.