Many industry recommendations for oil cleanliness are based on particle counts quantified by ISO 4406, Method for Coding the Level of Contamination by Solid Particles. In this standard, numbers of particles greater than or equal to 4, 6 and 14 microns in size (effective diameter) are classified on a scale, or ISO code. The ISO codes range in size from 1 to 28, where an ISO code of 1 corresponds to less than one particle per milliliter and an ISO code of 28 corresponds to millions of particles per milliliter. An increase of one ISO code corresponds to a doubling of the number of particles.

The most common way to measure particle count is through the use of an optical particle counter. These instruments shine a laser through the oil towards a photodetector, and when something scatters or blocks the light, it is counted as a particle. If any size-based ISO code is above the maximum specified for a given application, the oil is considered contaminated. Automated optical particle counters are most commonly used because they are easy to operate and can be adapted to a wide range of online and offline process configurations.

Additives as Particles

Lubricants such as hydraulic fluids contain additives that are critical to the function of the fluid. These additives are crucial to the performance and useful lifetime of the oil. However, when contamination is measured using an optical particle counter, some additives can affect the path of the light as it shines through the oil, resulting in erroneous particle counts.

In a lab at the University of California, Merced, researchers designed and built a test rig to measure additive-induced particle counts. The rig consists of a fluid reservoir, a motor and pump to circulate the fluid, an aluminum filter housing and a HIAC ROC in-line particle counter. A three-way valve allows the fluid to be circulated through the particle counter only, or through both the filter and the particle counter. There is also a port that allows samples to be drawn during circulation. Using this setup, we tested the effects of a typical dispersant-inhibitor additive package, viscosity modifier and foam inhibitor additives in the lubricant formulation-both with and without filtration, and in real time-on optically-detected particle counts.

Figure 1 shows the results of a representative test in which additives were introduced one at a time into a base oil while particles were measured using an optical counter. The ISO codes for the base oil alone were 15, 13 and 10 (reported as 15/13/10) for the 4, 6 and 14 micron particles, respectively. This corresponds to a relatively clean fluid before filtration. (Otherwise clean base oils may have moderately low particle counts due to low-level contamination by dust or silt, fluid opacity or the presence of water or entrained air.)

However, introducing the additives increased all of the ISO codes substantially. The most significant increase was caused by the foam inhibitor, which resulted in a 4 micron ISO code greater than 28, which corresponds to more than 2.5 million particles per milliliter. The full counts were >28/23/18, far above the recommended limits. This observation is typical of polydimethylsiloxane (silicone) based foam inhibitors, which are the most commonly used anti-foam additives in lubricant formulations. However, it is important to keep in mind that the oil is not actually dirty; it just contains additives that are interfering with the optical detection method used by the particle counter.

This can be a major problem, particularly when trying to follow recommendations for oil cleanliness. For example, a cleanliness level equal to or better than ISO 18/16/13 is recommended by the American Gear Manufacturers Association for in-service wind turbine gearboxes. This corresponds to maximum particle counts of 2,500 per mL for particles larger than 4 microns, 640 per mL for particles larger than 6 microns and 80 per mL for particles larger than 14 microns. (Additional original equipment manufacturer cleanliness specifications are shown in Table 1.)

Filtration Effects

Generally speaking, contaminated fluid can be cleaned by filtration. Can filtration then also be used to decrease additive-induced particle counts? We tested this by filtering the additized fluid through a 4-micron synthetic filter. As shown in Figure 1, filtration causes a sharp decrease in additive-induced particle counts. However, filtration may not be the answer to the problem, because decreasing particle counts in this case actually means removing some additives from the fluid.

Filtration can have a particularly negative effect on the performance of foam inhibitors. Foam performance is characterized by ASTM D892-13, Standard Test Method for Foaming Characteristics of Lubricating Oils, which calls for forcing air into the oil through a stone or metal diffuser and measuring the resultant foam. We performed part of this standard test, the Sequence I foam tendency, on the oil samples before, during and after filtration. The results are shown in Figure 2. We see that there is a direct correlation between the optically detected 4 micron ISO Code and the foam performance, i.e., the ISO code decreased and the foam tendency increased with filtration. Therefore, although filtration can be used to decrease additive-induced particle counts, this approach could adversely affect the performance of the oil, particularly foaming.

Counting Carefully

In summary, high particle counts measured when new oil is tested with an optical particle counter could simply mean that additives are present. This is particularly an issue with polydimethylsiloxane foam inhibitors, which can lead to particle counts far above recommended values. Applications that require the most stringent cleanliness include hydraulic and turbine systems, as well as high pressure systems and industrial servo valves. Machined tolerances in these applications can be very tight, and it is critical to differentiate foam inhibitors from foreign contaminants that can cause significant mechanical wear.

If the source of the particle counts is not determined to come from additives, the equipment operator might try to filter the oil to address the perceived problem. Filtration will decrease the particle counts, but the process may remove additives, which can adversely affect fluid performance; again, this problem is most severe with foam inhibitors, where higher foam can lead to hydraulic inefficiency and cavitation wear.

One approach to the issue of additive-induced particle counts is to measure particulate contamination by a different technique, such as ASTM D7647, Standard Test Method for Automatic Particle Counting of Lubricating and Hydraulic Fluids Using Dilution Techniques to Eliminate the Contribution of Water and Interfering Soft Particles by Light Extinction, which is designed to eliminate the counting of water droplets and other soluble soft particles. This method uses specified solvents to mask the light-scattering effect of these particles, which can then limit particle counts to actual contamination.

However, that can be an expensive option. A simple alternative is to measure the particle counts of a new, ideally clean, fluid using your own equipment. Then, those results can be used as a reference later if particle counts increase due to real contaminant particles. The most important thing is just to be aware of the potential effect of additives on optically detected particle counts and interpret cleanliness measurements accordingly.

Further Reading

Lantz, S., Zakarian, J., Deskin, S. and Martini A. (2017): Filtration Effects on Foam Inhibitors and Optically Detected Oil Cleanliness, Tribology Transactions, Vol. 60, No. 6, p. 1159-1164.

Sander, J. Busting the Ghost Particles, LubesnGreases, Vol. 21, No. 12, December 2015.

Ashlie Martini is a professor of mechanical engineering at the University of California, Merced, where she runs a research lab focused on tribology and lubrication engineering. She is a former member of the Society of Tribologists and Lubrication Engineers board of directors and chair of the Tribology Frontiers Conference; she now helps plan and
organize STLEs annual meeting. Contact her at amartini@ucmerced.edu.

Scott Deskin has worked as product formulator for Chevron Lubricants since 2008, currently focusing on automatic transmission and tractor hydraulic fluids. He has an M.S. in chemical engineering and worked in the plastics industry for three years before joining Chevron in 2002. Contact him at Scott.Deskin@chevron.com.