Uncertainty analysis in turbomachine and cascade testing

Journal article

The analysis of experimental uncertainties is generally accepted today as an important part of the reporting of test results, and although the debate continues concerning the best methods of estimating and presenting uncertainties, the common practices are widely recognized and understood. In this paper we examine critically the accepted methods for analyzing uncertainties. Our concern, however, is not simply with the estimation and reporting of experimental uncertainties, but with their active use in planning and setting up experiments and tests to optimize the instrumentation and test procedures for best quality (minimum uncertainty) data. Using the techniques developed, it is possible for any given experimental arrangement to optimize measurement techniques for minimum uncertainty. This has been demonstrated in critical turbine cascade data, where improvements up to an order of magnitude have been achieved. In the measurement of blade profile loss it has been found necessary to measure some pressures, such as the inlet total pressure, to an order of magnitude better accuracy than others. The analysis is developed and results are presented for turbomachine and blade cascade testing, but the principles are generally applicable.

Authors: Baines, N; Mee, D; Oldfield, M

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Gulf Publishing Company
• Journal: International Journal of Engineering Fluid Mechanics
• Volume: 4
• Issue: 4
• Pages: 375-401
• Publication date: 1991-01-01
• Language: English
• Subjects: Engineering & allied sciences