Experimental Correlation of a Thermal / Fluid Dynamic / Electrical Performance Model of a Multiple-Tube, Vapor-Anode AMTEC Cell

Abstract

An AMTEC (Alkali-Metal Thermal-to-Electric Conversion) cell performance analysis model described by Hendricks et al. (1998) has been correlated with experimental data by comparing its predictions for Beta” alumina solid electrolyte (BASE) and evaporator temperatures, voltage, power output, and conversion efficiency with experimental measurements on two versions of a PX-6 AMTEC cell. The critical features of this AMTEC cell model, the PX-6 experimental set up and testing, and the model correlation PX-6 test data are discussed in this work. Model prediction/test data comparisons are presented across a wide range of current-voltage conditions and hot side temperatures for two significantly different cell lengths. Cell model predictions demonstrate good agreement with experimental PX-6 test data in simultaneously predicting BASE tube and evaporator temperatures, the onset of sodium (Na) condensation in the BASE tubes, current-voltage characteristics, and power output in high current ranges (i.e., > 1.5 A). The model also has demonstrated good capability to predict cell conversion efficiency at high currents when Na is not condensing in the BASE tubes. The good model prediction/test data comparisons have demonstrated the progress in developing this cell performance model and increased confidence in its technical foundations, algorithm implementation, and capability to predict AMTEC cell performance. The AMTEC cell model’s capability to simultaneously predict many critical cell performance parameters across a wide range of hot side temperatures, at high current conditions, and different cell lengths demonstrates the progress that has been made in its development. It has demonstrated good predictive capability, utility, and flexibility as a performance design and analysis tool for sophisticated AMTEC cell design. Testing limitations prevented testing at low current levels (i.e., < 1.5 A), so future experimental validation studies should focus on correlating model predictions at low currents.