A Preliminary Turbine Design for an Organic Rankine Cycle
An increasing trend in exploiting low enthalpy content energy sources, has led to a renewed interest in small-scale turbines for Organic Rankine Cycle applications. The design concept for such turbines can be quite different from either standard gas or steam turbine designs. The limited enthalpic content of many energy sources enforces the use of organic working media, with unusual properties for the turbine. A versatile cycle design and optimization requires the parameterization of the prime turbine design. In order to address the major challenges involved in this process, the present study discusses the preliminary design of an electricity-producing turbine, in the range of 100 kWel, for a low enthalpy organic Rankine cycle. There are many potential applications of this power generating turbine including geothermal and solar thermal fields or waste heat of PEM type fuel cells. An integrated model of equations has been developed, accordingly. The model aims to assess the performance of an organic cycle for various working fluids, including NH3, R600a and R-134a. The most appropriate working fluid has been chosen, taking into consideration its influence on both cycle efficiency and the specific volume ratio. The influence of this choice is of particular importance at turbine extreme operating conditions, which are strongly related to the turbine size. In order to assess the influence of various design parameters, a turbine design tool has been developed and applied to preliminarily define the blading geometry. Finally, a couple of competitive turbine designs have been developed. In one approach, the turbine speed is restricted to subsonic domain, while in the other approach the turbine speed is transonic, resulting to choked flow at the turbine throat. The two approaches have been evaluated in terms of turbine compactness and machine modularity. Results show that keeping the crucial parameters of the geometrical formation of the blade constant, turbine size could become significantly smaller decreasing up to 90% compared its original value.