Chapter 02: Surface Roughness Analysis and …

1 2 Surface RoughnessAnalysis andMeasurement Techniques The Nature of Analysis of Surface Roughness Average Roughness Parameters Statistical Analyses Fractal Characterization Practical Considerations in measurement of Roughness Parameters measurement of Surface Roughness Mechanical Stylus Method Optical Methods Scanning Probe Microscopy (SPM) Methods Fluid Methods Electrical Method Electron Microscopy Methods Analysis of Measured Height Distribution Comparison of measurement Methods Closure The Nature of surfaces A solid Surface , or more exactly a solid gas or solid liquid interface, has a complex structure and complexproperties depending on the nature of the solids, the method of Surface preparation, and the interactionbetween the Surface and the environment. Properties of solid surfaces are crucial to Surface interactionbecause Surface properties affect real area of contact, friction, wear, and lubrication.

2 In addition totribological functions, Surface properties are important in other applications, such as optical, electricaland thermal performance, painting, and surfaces , irrespective of their method of formation, contain irregularities or deviations from theprescribed geometrical form (Whitehouse, 1994; Bhushan, 1996, 1999a,b; Thomas, 1999). The surfacescontain irregularities of various orders ranging from shape deviations to irregularities of the order ofinteratomic distances. No machining method, however precise, can produce a molecularly flat surfaceon conventional materials. Even the smoothest surfaces , such as those obtained by cleavage of somecrystals, contain irregularities, the heights of which exceed the interatomic distances. For technologicalapplications, both macro- and micro/nanotopography of the surfaces ( Surface texture) are important(Bhushan, 1999a,b).

3 In addition to Surface deviations, the solid Surface itself consists of several zones having physico-chemical properties peculiar to the bulk material itself ( Figure ) (Gatos, 1968; Haltner, 1969; Buckley,1981). As a result of the forming process in metals and alloys, there is a zone of work-hardened or Bharat Bhushan The Ohio State University deformed material on top of which is a region of microcrystalline or amorphous structure that is calledthe Beilby layer. Deformed layers would also be present in ceramics and polymers. These layers areextremely important because their properties, from a Surface chemistry point of view, can be entirelydifferent from the annealed bulk material. Likewise, their mechanical behavior is influenced by theamount and depth of deformation of the Surface of the surfaces are chemically reactive. With the exception of noble metals, all metals and alloysand many nonmetals form Surface oxide layers in air, and in other environments they are likely to formother layers (for example, nitrides, sulfides, and chlorides).

4 Besides the chemical corrosion film, thereare also adsorbed films that are produced either by physisorption or chemisorption of oxygen, watervapor, and hydrocarbons, from the environment. Occasionally, there will be a greasy or oily film derivedfrom the environment. These films are found on metallic and nonmetallic presence of Surface films affects friction and wear. The effect of adsorbed films, even a fractionof a monolayer, is significant on the Surface interaction. Sometimes, the films wear out in the initialrunning period and subsequently have no effect. The effect of greasy or soapy film, if present, is moremarked; it reduces the severity of Surface interaction often by one or more orders of Chapter covers the details on the Analysis and measurement of Surface Roughness . Analysis of Surface Roughness Surface texture is the repetitive or random deviation from the nominal Surface that forms the three-dimensional topography of the Surface .

5 Surface texture includes (1) Roughness (nano- and microrough-ness), (2) waviness (macroroughness), (3) lay, and (4) flaws. Figure is a pictorial display of surfacetexture with unidirectional lay (Anonymous, 1985).Nano- and microroughness are formed by fluctuations in the Surface of short wavelengths, character-ized by hills (asperities) (local maxima) and valleys (local minima) of varying amplitudes and are large compared to molecular dimensions. Asperities are referred to as peaks in a profile (twodimensions) and summits in a Surface map (three dimensions). Nano- and microroughness include thosefeatures intrinsic to the production process. These are considered to include traverse feed marks andother irregularities within the limits of the Roughness sampling length. Waviness is the Surface irregularity FIGURE Solid Surface details: Surface texture (vertical axis magnified) and typical Surface layers.

6 Of longer wavelengths and is referred to as macroroughness. Waviness may result from such factors asmachine or workpiece deflections, vibration, chatter, heat treatment, or warping strains. Wavinessincludes all irregularities whose spacing is greater than the Roughness sampling length and less than thewaviness sampling length. Lay is the principal direction of the predominant Surface pattern, ordinarilydetermined by the production method. Flaws are unintentional, unexpected, and unwanted interruptionsin the texture. In addition, the Surface may contain gross deviations from nominal shape of very longwavelength, which is known as errors of form. They are not normally considered part of the Surface FIGURE Pictorial display of Surface texture. (From Anonymous (1985), Surface Texture ( Surface Roughness ,Waviness and Lay), ANSI/ASME , ASME, New York.)

7 With permission.) texture. A question often asked is whether various geometrical features should be assessed together orseparately. What features are included together depends on the applications. It is generally not possibleto measure all features at the same very general typology of a solid Surface is seen in Figure Surface textures that are deterministicmay be studied by relatively simple analytical and empirical methods; their detailed characterization isstraightforward. However, the textures of most engineering surfaces are random, either isotropic oranisotropic, and either Gaussian or non-Gaussian. Whether the Surface height distribution is isotropicor anisotropic and Gaussian or non-Gaussian depends upon the nature of the processing method. Surfacesthat are formed by cumulative processes (such as peening, electropolishing, and lapping), in which thefinal shape of each region is the cumulative result of a large number of random discrete local events andirrespective of the distribution governing each individual event, will produce a cumulative effect that isgoverned by the Gaussian form.

8 It is a direct consequence of the central limit theorem of statistical processes (such as turning and shaping) and extreme-value processes (such as grinding andmilling) generally lead to anisotropic and non-Gaussian surfaces . The Gaussian (normal) distributionhas become one of the mainstays of Surface this section, we first define average Roughness parameters, followed by statistical analyses and fractalcharacterization of Surface Roughness that are important in contact problems. Emphasis is placed onrandom, isotropic surfaces that follow Gaussian distribution. Average Roughness Parameters Amplitude Parameters Surface Roughness most commonly refers to the variations in the height of the Surface relative to areference plane. It is measured either along a single line profile or along a set of parallel line profiles( Surface maps). It is usually characterized by one of the two statistical height descriptors advocated bythe American National Standards Institute (ANSI) and the International Standardization Organization(ISO) (Anonymous, 1975, 1985).

9 These are (1) R a , CLA (center-line average), or AA (arithmetic average)and (2) the standard deviation or variance ( ), R q or root mean square (RMS). Two other statistical FIGURE General typology of surfaces . height descriptors are skewness (Sk) and kurtosis (K); these are rarely used. Another measure of surfaceroughness is an extreme-value height descriptor (Anonymous, 1975, 1985) R t (or R y , R max , or maximumpeak-to-valley height or simply P V distance). Four other extreme-value height descriptors in limiteduse, are: R p (maximum peak height, maximum peak-to-mean height or simply P M distance), R v (maximum valley depth or mean-to-lowest valley height), R z (average peak-to-valley height), and R pm (average peak-to-mean height).We consider a profile, z(x), in which profile heights are measured from a reference line Figure Wedefine a center line or mean line such that the area between the profile and the mean line above the lineis equal to that below the mean line.

10 R a , CLA, or AA is the arithmetic mean of the absolute values ofvertical deviation from the mean line through the profile. The standard deviation is the square rootof the arithmetic mean of the square of the vertical deviation from the mean mathematical form, we write( )and( )where L is the sampling length of the profile (profile length).The variance is given as( )( )where, is the standard deviation and R q is the square root of the arithmetic mean of the square of thevertical deviation from a reference line, or( ) FIGURE Schematic of a Surface profile z(x).RCLAAAL zmdxaL=== 10mLzdxL= 10 2201= () LzmdxL= Rmq22 RRMSL zdxqL22201==() For the special case where m is equal to zero,( )In many cases, the R a and are interchangeable, and for Gaussian surfaces ,( )The value of R a is an official standard in most industrialized countries. Table gives internationallyadopted R a values together with the alternative Roughness grade number.