More than a singular concept, surface finish is really a function of four distinct factors. The most important factor is roughness, or the closely-spaced surface irregularities that result from manufacturing and/or cutting (as by tools or abrasive materials, see Figure 84). These irregularities are typically measured in microinches (millionths of an inch) or micrometers (millionths of a meter). To make a topographic analogy, roughness is akin to a plowed field where the churned dirt forms countless small pockets in the topsoil.

The second factor is lay, or the direction of the primary roughness pattern (again, see Figure 84). In other words, the way in which the surface irregularities are oriented. In terms of our analogy, lay would denote the particular pattern left in the dirt after it has been churned.

The third factor is waviness, or surface irregularities with considerably longer wavelengths than those referenced as roughness (see Figure 85). Waviness irregularities can be caused by, among other things, machinery vibrations or material warping. If roughness is analogous to a plowed field, waviness can be thought of as a slowly rolling hill.

Flaws are the fourth factor that should be considered. Flaws are surface imperfections that occur only infrequently, i.e. not in a pattern (see Figure 86). Flaws may be caused by inconsistencies within the metal itself, or through impact or abrasion after processing, as with scratches, cracks, etc. Depending on the severity, a single flaw may be enough to compromise the functionality of the surface. A flaw is like an isolated sinkhole or fissure in an otherwise unmarred plain.

The superimposition of these four factors onto one another determines the characteristics of a given surface (see Figure 87). Roughness, lay, waviness, and flaws must all be measured to get a complete picture of the surface. The question then becomes: how best to reflect these measurements?

For many years, surface finish has been noted in terms of RMS, or Root Mean Square. As a mathematical concept, RMS is the square root of the sum of the squares of the individual surface irregularity readings taken over a given sampling distance. More simply, RMS reflects the average depth of the irregularities a seal may encounter across a gland surface. That is, the higher the RMS number, the greater the depth of these irregularities and the greater the likelihood that they will impede or damage the seal.

For example, break-out friction (also known as static friction or stiction) results when seal material flows into these tiny metallic irregularities during a period of no relative motion. The more time that the seal and the gland are in contact, the greater the interface between them, and the greater the break-out friction. A time-lapse look at the seal’s progressive flow into the irregularities can be seen in Figure 88. A combination of rubber-to-metal adhesion and the shearing force generated by the irregularities must be overcome before movement can begin. Smaller surface irregularities (as denoted by a lower RMS number) will allow for less interface with the seal material and thus decrease break-out friction, running friction, and wear.

A word of caution is in order here, however: RMS measurements are good as far as they go, but be aware that they deal solely in depth, ignoring both shape and direction. It is entirely possible to have a number of different types of surface irregularities that would all result in the same RMS measurement but would affect seal material in vastly different ways. Some examples of this can be seen in Figure 89.

With this in mind, the optimal surface finish still depends on the application. Because they undergo no motion, most static seal surfaces need not be finished better than 32 microinches RMS. Some projects (e.g. low-pressure applications) may allow for surfaces as rough as 64 or even 128 microinches RMS. Due to increased friction and wear concerns, dynamic seals should have much smoother surfaces. Finishes of 8 to 16 microinches RMS are common for dynamic seals. As you might expect, smoother surfaces take longer to machine (and are more expensive) than rougher surfaces.

Keep in mind that there is not a fixed relationship between RMS measurements. In other words, a surface finish of 80 microinches is definitely rougher than a finish of 40 microinches, though not necessarily twice as rough. You should also be aware that, contrary to popular opinion, it is possible to have too much of a good thing; gland surfaces can be too smooth. The surface irregularities that contribute to frictional build-up are the same irregularities that entrap lubricating fluids. A finish of less than five microinches will essentially eliminate these metallic micropores, making the metal too smooth to hold on to lubrication. Friction will increase and the entire process will be for naught.

Experience has shown that traditional RMS measurements are not completely indicative of surface irregularities, so many manufacturers now use profilometers geared to generate “Ra” (roughness average) measurements. Ra is the sum of the absolute values of the peaks (above a median surface baseline) and the absolute values of the valleys (below this baseline) divided by the length of the sample (see Figure 90). For example, let’s say the peaks have a total absolute value of 23 microinches; the valleys have a total absolute value of 27 microinches. The sum of these values (50 microinches) divided by the sample length (we’ll say two inches) yields an Ra value of 25 microinches. Since Ra measurements take into account both the peaks and the valleys in a given sample, many designers consider Ra results to be more indicative of surface irregularities than simple RMS figures.

CONSTRUCTION MAIN PAGE