Civil Engineering Reference
In-Depth Information
FIGURE 1.2
A vision for manufacturing fingerprinting.
surface analysis and synthesis; (2) representing structure in the stochastic nature of manufacturing
processes; and (3) clearly separating the deterministic aspects in the presence of stochastic and time-
varying effects. A simple implementation of this vision is shown in
Fig. 1.2
:
surface measurements are
decomposed and fingerprinted for the purpose of isolating the faulty submechanisms in the manufac-
turing system. The remainder of this chapter expounds on this vision through the development of
practical tools to understand surface characteristics at a new level. These tools are illustrated with
mathematical representations, numerical examples, and industrial applications.
1.2
The Never-Ending Search for the Perfect
Surface Analysis Method
Understanding manufactured surfaces has been an interesting research problem for a long time. Surface
characterization is meant to decompose the surface geometry into its basic components in order to
understand it better [74]. The progressive understanding gained from the numerous different techniques
to characterize surfaces has been valuable [74]. However, while the use of some of these techniques has
been standardized successfully, the need to develop better and more accurate methods has not ceased.
In fact, the growing need for automated manufacturing and design necessitates a revolutionary technique,
which will provide a unified means of characterizing all significant components simultaneously. In this
section, we present an overview of the traditional techniques developed for surface analysis purposes.
We then present some alternative ways of looking at surfaces and our criteria in helping to advance the
field of surface analysis.
Traditional Surface Characterization Methods
Manufactured surfaces are typically viewed as containing various types of deviations from the intended
nominal surface profile [74]. The widely recognized types of deviations are identified as: (1) irregularities
known as surface “roughness” that often result from the manufacturing process, such as tool marks left
as a result of turning or marks resulting from a grinding wheel; (2) irregularities known as surface
“waviness” of a longer wavelength caused by improper manufacturing, such as vibration between a
workpiece and a cutting tool; and (3) low-frequency waveforms or linear trends referred to as errors of
“form,” such as errors in rotating members of a machine or thermal distortion errors.
The most traditional means of characterizing these various types of surface errors is by using average
profile parameters [74]. Two primary parameters are roughness and waviness height measures. The height
parameters are typically specified on part models in an attempt to control the manufacturing process.
However, efforts to optimize the functionality of a part leads to a need to supplement the height
information. One such supplement parameter is the spacing parameter, defined as the average distance
between the positive crossings of the profile with the mean line [44, 74]. Other parameters define the
maximum range of the profile, as well as the skewness (asymmetry) and the randomness of the profiles.
