A multi-variable optimization program has been developed to investigate the performance of Organic Rankine Cycles (ORCs) for low temperature heat recovery applications using both turbine and twin-screw expanders when account is taken of performance variation due to changes in ambient conditions. The cycle simulation contains thermodynamic models of both types of expander. In the case of the twin-screw machine, the methods used to match the operation of the expander to the requirements of the cycle are described. The performance of turbine expanders in a superheated ORC has been modelled using correlations derived from operational data for single stage reaction turbines to predict the turbine efficiency at “off-design” conditions. Several turbine configurations have been considered including variable nozzle area and variable speed. The capability of the cycle model has been demonstrated for the case of heat recovery from a steady source of pressurized hot water at 120 °C. The system parameters are optimised for a typical operating condition, which determines the required size of heat exchangers and the expander characteristics. Performance at off-design conditions can then be optimized within these constraints. This allows a rigorous investigation of the effect of air temperature variation on the system performance, and the seasonal variation in net power output for the turbine and twin-screw ORC systems. A case study is presented for a low temperature heat recovery application with system electrical power output of around 100 kWe at design conditions. The results indicate that similar overall performance can be achieved for ORC systems using either type of expander.