Contamination Control

Dr. Robert M. Gresham, Contributing Editor | TLT Lubrication Fundamentals October 2013

Want to say goodbye to the unwanted nasties that can wreck machinery? Start by changing the culture at the plant.

 

KEY CONCEPTS
• Lubricant contamination can occur at any of several stages in its production and application.
• The degree to which a lubricant is contaminated is as important as its viscosity.
• There are three major facets to controlling the influx of unwanted materials in the lubricant.

As a chemist, I always think of contamination as arising from chemical contamination, often resulting in unintended chemical reactions between materials that should be present and those that should not.

In the lubrication world, this certainly can be an issue as well. Examples might be at the formulator’s shop where additives are contaminated with various other chemicals that lead to an unstable formulation or some other negative outcome. Certainly in the end-user’s shop, a given lubricating oil might be contaminated with another oil of greatly different properties, leading to machine failure. I’m sure you can think of a few other examples.

However, there are other contaminants, not relying so much on their inherent reactive chemistry with other chemistries in the lubricant system, that can result in chaos in our machinery. Indeed, these materials may not react chemically or they may react in tribochemical reactions on or with the substrate or mating surface(s).

We often say that viscosity is the most important property of a lubricant, and it is. Equally important is the degree to which that lubricant is contaminated with some kind of foreign particulate material. I’m referring to those little, tiny, often abrasive particles swimming around in our lubricants blissfully grinding away at the surfaces of bearings, seals, shafts, etc. (see illustration).


3-body abrasion

These third-body wear particles can arise from initial manufacture of the lubricant—they arrive free to the customer with no extra charge from your lubricant supplier. Also, when you are starting up new equipment, these particles can result from the manufacture and assembly of the equipment itself.

Additionally, these third-body wear particles can arise from improper use and maintenance of the lubricant in your shop—that is, you are allowing them entry into your equipment. Finally, the lonely little devils create additional wear particles as a result of their wear processes on your equipment. Clearly, we must be certain that our new equipment, lines, etc., are clean, thoroughly clean, before ever charging with lubricant.

Before we consider putting lubricant into our pristine machine components, we must be certain the incoming oil is clean as well. Significantly, it is often the smaller particles rather than the larger ones that can cause the most damage. And, what do we mean by clean? Let’s take a closer look.

THREE CONTROL FACETS
There are three major facets to contamination control:

1. Establish an acceptable level of contamination via the ISO code for each machine—easy to say, not so easy to do.
2. Perform oil sampling and analysis to determine the level of contamination.
3. Select filters and filter placement to remove contaminants, whether generated in the machine or from the outside.

For now, we’ll take a look at the first two and save filters for another article.

First, not all machines need the same level of contamination control. Thus, depending on the specific machine type (hydraulics, gears, bearings, etc., and also the specific model and manufacturer, etc.) there is a minimum allowable level of contamination. Remember, it is impossible to remove or eliminate all possible contamination.

Thus, for a new program where we are going to start controlling contamination, anything that shows improvement from historic levels is good. After that, armed with trend data, we should be able to steadily improve to some optimal level, after which pouring more money yields little gain—kind of like the old 90/10 rule.

To accomplish this goal of contamination control, we need to instill a culture of change in the plant. Everyone needs to be focused on eliminating entry of contaminants into the system. This starts with incoming quality assurance requirements on the lubricant, moves to techniques for adding lubricant to the machines (without introducing contaminants) and ends with the control and elimination of contamination as the machine operates, which usually involves maintenance of sealing systems and filtration.

The next issue is—how do we know if we are doing the job? This requires a proactive maintenance program for routine sampling and lubricant testing. Particle counting techniques is the way this is done. There are a wide variety of analytical machines that serve this purpose. They also vary widely in the range and sensitivity of their measurement.

Automatic optical particle counting is a common method. However, it is not effective for high-viscosity and colored fluids, and emulsified water causes interference with particle readings. Thus, it is important to select the right analytical instrument for your fluid. Interestingly, like most things in plants, if you just measure it the problem magically improves—I think that goes to the culture change issue. In any event, after we measure the particle count, what do we do with the data?

The ISO 4406 Solid Contamination Code is the most widely used method for characterizing particle counts in oils. As far as I know, the current standard is still the 1999 version that employs a two-number system, although the new standard will use a three-number system.

Often a particle distribution more or less follows a bell curve skewed one way or another. (However, I have also seen bilobal distributions and other anomalous distributions.) Most people are currently using a three-number system now: R₄/R₆/R₁₄. The R is a so-called range number representing the range of the number of particles corresponding to a certain size. With three range numbers, you get a snapshot of the distribution curve in three places. By selecting the right particle sizes, you can also get a better picture of the number of particles that are the most deleterious to your particular machine.

The range number is codified in the noted chart. In this case, the first R represents the number of particles greater than four microns, the second the number greater than six and the third the number greater than 14 microns. Using the chart, if we have a target cleanliness for a given machine of 16/14/11, this would mean we would seek a distribution of 320-640 of 4-micron particles, 80-160 of 6 microns and 10-20 of 14 microns per milliliter of sample.


Number of particles per milliliter

So the trick is to determine the acceptable level of contamination and characterize it via the ISO code (such as the 16/14/11 example above) for the application. Work with the OEM, your lubricant and filter suppliers and your oil analysis laboratory to determine acceptable ISO cleanliness code limits for each piece of machinery by application.

Once determined, how do we achieve it? After the culture change which, believe me, will have a significant impact, we need to install filters in our system. Alas, all filters are not the same, and they are a relatively high maintenance item as they must be monitored to be sure they are working effectively and changed when clogged or otherwise incapacitated.

Additionally, it is critical that filters be located where they can do the most good—seems obvious, but in the real world some pretty strange installations can occur. And that’s what we’ll look at next time.


Bob Gresham is STLE’s director of professional development. You can reach him at rgresham@stle.org.