Roundness and cylindricity factors influencing measurement and its effect on interference fits
Loading...
Date
item.page.authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Round or cylindrical components are very common in engineering products. Their
newlinedeparture from the ideal round or cylindrical geometry is quantified in terms of roundness
newlineand / or cylindricity. The designer has to specify these values to ensure satisfactory
newlineperformance of the product and the aim of the metrologist has to measure the components
newlineto verify that they are within specifications. Measurement of roundness/ cylindricity is
newlinecommonly carried out using a roundness testing machine. The work reported in the thesis
newlinecan be divided into two parts, a study of how the uncertainty of the measurement of
newlineroundness is affected by the various measurement parameters and a study on the effect of
newlineroundness and cylindricity error on the extraction load of interference fits.
newlineThe first part reports on an experimental study to determine the effect of various
newlinefactors on the resulting uncertainty of roundness measurement. The factors considered
newlinewere the number of sampling points, method of initial setup, rotary table speed,
newlineworkpiece diameter and form error of the specimen. As a typical workpiece commonly
newlinemeasured in such roundness testing machines, the poppet valve of internal combustion
newlineengines was chosen. The roundness of the valve face is a critical parameter to minimize
newlineleakage and ensure satisfactory operation of the engine. An experimental design was
newlineplanned which simultaneously varied the factors and levels. Since valves with high value
newlineof roundness error were not available, available valves were modified using wire electro
newlinedischarge machining to obtain the desired roundness error. Repeated measurements were
newlinecarried out of each valve to determine the measurement uncertainty at each condition.
newlineThe stability of the measurement process at each level was verified using control charts.
newlineUsing statistical methods, an empirical model was developed, which could be used to
newlinepredict the standard uncertainty of measurement at various measurement conditions.
newline