12.1 SURFACE GEOMETRY
Before
discussing the types of wear mechanisms in detail, it will be useful to define
the characteristics of an engineering surface that are relevant to these processes.
(Material strength and hardness will also be factors in wear.) Most solid
surfaces that are subject to wear in machinery will be either machined, ground,
or EDM’d, though some will be as-cast or as-forged. In any case, the surface
will have some degree of roughness that is concomitant with its finishing
process. Its degree of roughness or smoothness will have an effect on both the
type and degree of wear that it will experience.
Even
an apparently smooth surface will have microscopic irregularities. These can be
measured by any of several methods. A profilometer passes a lightly loaded,
hard (e.g., diamond) stylus over the surface at controlled (low) velocity and
records its undulations. The stylus has a very small (about 0.5
m) radius tip that acts, in effect, as a
lowpass filter, since contours smaller than its radius are not sensed.
Nevertheless, it gives a reasonably accurate profile of the surface with a
resolution of 0.125
m or better. Figure 12-1 shows the profiles and
SEM* photographs (100
x
) of both (
a
) ground and (
b
)
machined surfaces of hardened steel cams. The profiles were measured with a
Hommel T- 20 profilometer that digitizes 8 000 data points over the sample
length (here 2.5 mm). The microscopic “mountain peaks” on the surfaces are
called
asperities
.
From
these profiles a number of statistical measures may be calculated. ISO defines
at least 19 such parameters. Some of them are shown in Figure 12-2 along with
their mathematical definitions. Perhaps the most commonly used parameters are
Ra
,
which is the average of the absolute values of the measured points, and
Rq
,
which is their rms average. These are very similar in both value and meaning.
Unfortunately, many engineers specify only one of these two parameters, neither
of which tells enough about the surface. For example, the two surfaces shown in
Figures 12-3
a
and
b
have the same
Ra
and
Rq
values, but are clearly different in nature.
One has predominantly positive, and the other predominantly negative, features.
These two surfaces will react quite differently to sliding or rolling against
another surface.
In
order to differentiate these surfaces that have identical
Ra
or
Rq
values, other parameters should be calculated.
Skewness
Sk
is a measure of the shape or symmetry of the
amplitude distribution of the surface rougness contour. A negative value of
Sk
indicates
that the surface has a predominance of valleys (Figure 12-3
a
)
and a positive
Sk
defines a predominance of peaks (Figure 12-3
b
).
Many other parameters can be computed (see Figure 12-2). For example,
Rt
defines
the largest peak-to-valley dimension in the sample length,
Rp
the
largest peak height above the mean line, and
Rpm
the average
of the five largest peak heights. All the roughness measurements are calculated
from an electronically filtered measurement that zeros out any slow-changing
waves in the surface. An average line is computed from which all peak/valley
measurements are then made. In addition to these roughness measurements
(denoted by
R
), the waviness
Wt
of
the surface can also be computed. The
Wt
computation filters out the
high-frequency contours and preserves the long-period undulations of the raw
surface measurement. If you want to completely characterize the surface-finish
condition, note that using only
Ra
or
Rq
is
not sufficient. A cam surface finish specification should, at a minimum,
include limits for
Ra, Rt,
Rpm, Sk
, and
Wt
.
* Scanning Electron Microscope
Copyright 2004, Industrial
Press, Inc., New York, NY