Estimating Precision in Heat Rate Testing

Abstract

Obtaining an estimate of the precision realized in heat rate testing is discussed in this paper. Heat rate precision depends largely on the scatter of the most important measurements, so we examine their scatter. All solutions proposed are based on the assumption that the data arise from a normal population, although answers are also useful for near-normal distributions. We propose a three-part procedure designed to ensure the gathering of data of a given precision with a given confidence. The procedure should fit easily into the test engineer’s routine.