This paper presents a method for evaluating the steady state performance of loss-in-weight powder feeding equipment and then demonstrates the use of the method in evaluating the performance of a K-Tron KT35 loss-in-weight feeder for three pharmaceutical grade powders. The method describes in detail the experimental setup, the procedure used in collecting feedstream data, and the filtering and subsequent analysis that is needed to determine differences in the steady state feeding performance. The data stream is shown to be a complex data set that can be simplified, after applying a suggested filtering and analysis algorithm, to two quantities (average feedrate and relative standard deviation) that can be used for comparison of different feeding treatments (powder, feeder tooling, and setpoint). Through analysis of variance (ANOVA), the significance of various tooling on optimizing feeder performance can be evaluated. (C) 2012 Elsevier B.V. All rights reserved.