Characteristic Shapes and Wavelength Decomposition of Surfaces in Machining

In the past, characterization of machined surfaces by a few parameters, even at the risk of a less comprehensive description, was essential for the common manual shop floor use. In this era of explosive growth of electronic instrumentation and computers, restriction to few numbers is no longer required. Therefore this paper presents a comprehensive characterization of surfaces including detailed quantitative estimate of their wavelength content, random or uncorrelated component and a graphical representation of characteristic shapes. It includes the autocorrelation and autospectrum in a natural way, but its main advantage is that each wavelength component can be related and analyzed with respect to cutting conditions and the physical mechanisms or processes generating the surfaces. Examples of surface profiles generated in Electro Discharge Machining, turning and grinding are used to illustrate the effects of machining parameters on the characteristic shapes. The characterization can be readily derived by the Data Dependent Systems methodology and is particularly relevant and useful in computer aided manufacturing and design.