Theory and design for mechanical measurements

Abstract

Listen

Translate article

The assignment of numerical values to physical quantities underlies all quantitative statements in engineering and the physical sciences. This assignment is achieved by the process of measurement. The physical quantity being measured and the precision required in the numerical value determines the instrumentation to be used. The design of a measurement system therefore involves the analysis of the attribute to be measured, the means available for its detection and the verification that the measurement system performs as intended and can achieve the desired accuracy and precision. In this book Figliola and Beasley first discuss in general terms the basics of measurement, analogue and digital data acquisition systems and signal processing, the statistics of measurement and the analysis of error and uncertainty. In successive chapters they concentrate on the instruments and their physical basis in the areas of electricity, temperature, fluid flow, elastic strain and mechanics (displacement, motion, force and power). The coverage is directed towards measurements in various branches of engineering, with numerous worked examples and problems for students (approximately 30 to 40) at the end of each chapter. Since it is an American engineering text, the book uses both SI and English units. Unfortunately, the text is flawed by numerous errors. Some of the more egregious are that in chapter 1, `dimension' is used in place of `unit', the definitions given for the ampere and the ohm are in terms of `international' units that were abandoned in 1948 and derived units are expressed, for example, as `' in place of the standard forms (SI, ISO, ANSI) of `' or `'. There are furthermore numerous minor numerical errors and inconsistencies. One of the more serious flaws is the failure to distinguish between bias errors and uncertainty due to bias in the discussion of chapter 5. There is also a misuse of the student in the evaluation of uncertainty (although this error is not exclusively Figliola and Beasley's, since it occurs in ANSI documents on fluid flow measurement). Given the estimate for variance, with degrees of freedom, an uncertainty interval at confidence level p is properly , while the uncertainty for a combined quantity is where is evaluated from the Welch - Satterthwaite expression In spite of its shortcomings, the book collects a great deal of material in one place and, in the hands of a careful instructor who is aware of its flaws, could be useful as a supplementary text on measurement. E Richard Cohen