Should You Use Ferrography In Your Sampling Program? Part II

By RAY GARVEY, Engineer, Emerson Process Management

Last week Brett Winberg wrote an article titled, “Should You Use Ferrography?” In that article was a good explanation of one of many methods for performing ferrography, but the question of whether or not you should or should not do it was left open. In this article I would like to answer that question.

The short answer is, “Yes, for most industrial machinery, you should use either Ferrography or another form of Wear Debris Analysis.” This technique, together with particle counting and ferrous density measurement are the additional things that cause industrial oil analysis to cost from US$30 to US$70 from most commercial oil labs; and it is worth it. You can expect better than 500% return on investment!

The question, "Are you wasting time doing oil analysis?" may sound strange coming from me; but I really am quite serious. During the past thirty-six months I have visited dozens of plants and asked about the oil analysis being done. I have found that many industrial plants are simply wasting time doing oil analysis.

These plants have little or nothing to report when asked about cost savings resulting from their oil analysis - so WHY are they doing it? In fact the opportunities for cost savings at those plants are excellent. Unfortunately, they are not finding problems because their oil analysis is not adequately searching for problems with particulate contamination and with mechanical wear.

The example shown in this case history shows how easy it is to get false negative results from conventional oil analysis that does not include particle counting with size distribution or Wear Debris Analysis (WDA). In this case the oil supplier provided "free" oil analysis that included AES spectroscopy that was unable to detect the serious particulate contamination and initial wear indications revealed by proper debris analysis. Why? Very simply, the AES does not view particles >5-microns. The big concentration of particles you will notice is in 5-micron to 15-micron size range. In fact, this sample was so contaminated that it looked visually cloudy for an hour after shaking but would eventually become clear again.

If your industrial oil analysis does not include particle counting with size distribution on all samples and WDA when appropriate, you are probably not getting the most from your sampling program. This type of oil analysis is excellent for condition based oil changes, but is inadequate for finding contamination and abnormal wear problems. Industrial contamination and wear problems require frequent oil analysis and they require tests that account for larger size particles. Not surprisingly, it is through measuring contamination and wear that most industrial oil analysis cost savings are found.

Manage your subscription to LubeTalk, using the SUBSCRIPTIONS area of this newsletter. You can also signup by visiting lubetrak.com. For more information, call toll-free 1.866.LUBETRAK (1.866.582.3872)