First-generation microturbines are based on the use of existing materials and proven technology, and with low levels of compressor pressure ratio and modest turbine inlet temperatures, have thermal efficiencies approaching 30% for turbogenerators rated up to 100 kW. For such small machines the goal of advancing beyond this level of performance is unlikely to include more complex thermodynamic cycles, but rather will be realised with higher turbine inlet temperatures. Advancing engine performance in this manner has a significant impact on recuperator technology and cost. In the compact heat exchanger field very efficient heat transfer surface geometries have been developed over the last few decades but further improvements perhaps using CFD methods will likely be only incremental. Automated fabrication processes for the manufacture of microturbine recuperators are in place, and on-going developments to facilitate efficient higher temperature operation are primarily focused in the materials area. Based on the assumptions made in this paper it is postulated that in the 100 kW size the maximum thermal efficiency attainable for an all-metallic engine is 35%. To achieve this the recuperator cannot be designed in an isolated manner, and must be addressed in an integrated approach as part of the overall power conversion system. In this regard, temperature limitations as they impact the recuperator and turbine are put into perspective. In this paper there is strong focus on recuperator material selection and cost, including a proposed bi-metallic approach to establish a cost-effective counterflow primary surface recuperator for higher temperature service. If indeed there is a long-term goal to achieve an efficiency of 40% for small microturbines, it can only be projected based on the utilisation of ceramic hot end components. Alas, the high temperature component that has had the minimum development in recent years to realise this goal is the ceramic recuperator, and efforts to remedy this situation need to be undertaken in the near future. (C) 2003 Elsevier Science Ltd. All rights reserved.
