Measuring Surface Topography

Because the basic types of surface geometry are caused by different factors and tend to have different relationships to the performance of the component, it is conventional to consider them separately during analysis. In general, if control of aspects of component performance related to surface topography is required (e.g., wear characteristics, friction, reflectivity, resistance to stress failure, or lubrication properties), roughness is analyzed. If control of some aspect of machine tool performance or component performance (e.g., noise or vibration generation) is required, waviness is analyzed.

Surface Texture Recorder. The most common type of contact method for measuring surface topography is the surface texture recorder, the principle of which is shown in Fig. 2. The stylus of the instrument is moved across the surface via a guiding mechanism to produce the "traced profile," which is defined by the interaction of the stylus with the component. The transducer produces a signal that is the difference between the traced profile and a "reference profile" or "datum profile" provided by the guideway. The transducer signal is then converted into a digital signal via an analog-to-digital converter. At this point the transducer contains only the vertical or Z-component of the profile. The horizontal or X-component generated by the traversing mechanism is combined with the Z-component to produce the "total profile." The total profile is then filtered to remove unnecessary information, which produces a "primary profile." This profile can then be subjected to filtering techniques that can separate the roughness, waviness, and form features of the surface.

Fig. 2 Basic principle of a surface texture recorder

Noncontact techniques are becoming increasingly popular in the measurement of surface topography, especially for surfaces that may be subject to damage using contact techniques. The results obtained are very similar to those of stylus methods and can use the same parameter definitions. Some noncontact methods, such as diffraction measurement, can measure surfaces quickly and easily and can potentially be used on machine tools at the point of manufacture.

Some noncontact methods do have limitations in measuring certain surfaces. For instance, in surfaces with high slope, an insufficient intensity of light is reflected back to the detector and the focus lens begins to follow inaccurately. Another example is that on contaminated surfaces, the contamination is measured because there is no force to remove it, and this distorts the results. Oxide layers on surfaces such as aluminum can also present problems, because the focusing lens will oscillate between the top and bottom of the oxide layer, giving the false impression of a very rough surface.

The focus-follow method (Fig. 3) involves the use of a moving lens to try to keep a spot of light focused on the surface. The lens movement correlates to the profile of the surface and its vertical movement is controlled by an electric motor. The analog electrical signal generated to drive the motor is digitized and processed in the same manner as a contact stylus. A variation of this method involves the use of a separate transducer to monitor the position of the lens; the electric signal from this transducer is used in the same manner as a contact stylus.

Fig. 3 Focus-follow method for noncontact measurement of roughness

Multiple-beam interferometers, such as the Fizeau interferometer, also have useful application in microtopography. They are used with a microscope to provide high resolution in three dimensions. The interference microscope divides the light from a single-point source into two or more waves. In multiple-beam interference microscopes, this is done by placing a partially transmitting and partially reflecting reference mirror near the surface of the specimen (Fig. 4).

Fig. 4 General principle of a multiple-beam interferometer

The multiple beams illustrated in Fig. 4 are superimposed after traveling different lengths. This produces interference patterns, which are magnified by the microscope. The interference fringes having a perfectly flat surface appear as straight, parallel lines of equal width and spacing. Height variations cause the fringes to appear curved or jagged, depending on the unit used. With multiple-beam interferometers, height differences as small as X/200 can be measured, where X is the wavelength of the light source.

Lasers can provide a monochromatic light source, which is required in interference microscopes. Typical systems can provide displays of isometric plots, contour plots, and qualitative parameters, such as surface roughness, camber, crown, radius of curvature, cylindrical sag, and spherical sag.

The Profile. Before surface topography assessment can be understood clearly, an important factor needs to be explained concerning the measured profile generated from the surface. To obtain a reasonably clear display of the height and spacing of the surface profile, the vertical stylus deviation typically needs to be magnified by at least 5,000*, whereas the horizontal measurement length is magnified by about 100*. So it is important to note that the resultant display is not a magnified cross section of the surface. Figure 5 provides an illustration of this principle.

Fig. 5 Diagram illustrating how the profile shape varies as Vh, horizontal magnification, is reduced relative to Vv, vertical magnification. (i) Surface profile magnified 5000x equally in all directions. (ii) Profile with Vv:Vh ratio of 5:1. (iii) Profile graph recorded with a Vv:Vh ratio of 50:1. As the horizontal magnification of the profile is increased, the length X-X is expanded to X'-X' and the peaks A, B, C, and D appear flatter. Increasing the horizontal magnification still further until it equals the vertical magnification expands the length Y-Y to Y'-Y'. Peaks E and G and valleys F and H now appear much flatter, but the actual difference in heights of the corresponding peaks and valleys in (i), (ii), and (iii) are exactly the same.

Fig. 5 Diagram illustrating how the profile shape varies as Vh, horizontal magnification, is reduced relative to Vv, vertical magnification. (i) Surface profile magnified 5000x equally in all directions. (ii) Profile with Vv:Vh ratio of 5:1. (iii) Profile graph recorded with a Vv:Vh ratio of 50:1. As the horizontal magnification of the profile is increased, the length X-X is expanded to X'-X' and the peaks A, B, C, and D appear flatter. Increasing the horizontal magnification still further until it equals the vertical magnification expands the length Y-Y to Y'-Y'. Peaks E and G and valleys F and H now appear much flatter, but the actual difference in heights of the corresponding peaks and valleys in (i), (ii), and (iii) are exactly the same.

0 0

Post a comment