The invention of shape memory alloys spurred a period of intense interest in the area of heat engines in the late 70's and early 80's. It was believed that these engines could use heat from low temperature sources such as solar heated water, geothermal hot water and rejected heat from conventional engines as a significant source of power. The interest has since dwindled, largely because small prototype devices developed in the laboratory could not be scaled up to produce significant power. It is believed that the scaled-up designs failed because they were dependent on friction as the driving mechanism, which led to large energy losses and slip. This thesis proposes a new chain and sprocket driving mechanism that is independent of friction and should therefore allow for large-scale power generation. This thesis begins by presenting properties and applications of shape memory alloys. The proposed design is then described in detail, followed by a review of the evolution that led to the final design. A brief chapter on thermodynamic modeling and a summary chapter suggesting improvements on the current design follow. / Master of Science