Inaccurate, unreliable partical counting can hamper your ability to make smart oil suitability decisions. That can cost your company considerably in tearms of time and money.
It's no mystery to Maintenance professionals that clean oil promotes enhanced equipment performance and reliability. There is, however, something that many of them do not know. Today's most commonly used particle counting tests for determining oil cleanliness—Filter & Count Method; Light Blockage Method—too often yield inaccurate and inconsistent results.
As detailed in this article, the inconsistency and lack of precision with these current practices can lead companies to waste considerable amounts of money and time developing maintenance plans based on inaccurate and unreliable information. But, by incorporating innovative methods to address sources of errors—including air entrainment, water content, and additive effects—particle counting precision and accuracy can be greatly improved. Maintenance professionals can then make better oil suitability decisions, garner a stronger return on their lubricant investment and enhance their equipment performance.
What is cleanliness?
Oil cleanliness can be defined as a measure of the level of dirt, other insoluble or hard particles in fresh or in-service oil. There are a number of factors that can impact a lubricant's cleanliness, most notably contamination and harsh operating conditions, such as extremely high temperatures, pressures and operating speeds. To this end, maintenance professionals are implementing oil analysis programs with the desire to gain accurate readings on the cleanliness of their oil to support oil suitability decisions.
Oil cleanliness is typically determined by particle counting. The Filter & Count (ISO 4407) and Light Blockage or Extinction (ISO 11500) tests are the most widely used methods of particle counting. Thus, the results of these tests were used as the foundation for the discoveries detailed in this article.
Filter & Count Method–ISO 4407…
Particle counting using the Filter & Count Method is conducted exactly how it sounds. Oil samples pass through a fine-patch filter that captures particles that are greater than four microns (> 4μ) in size. After the sample has passed through the filter, the particles on the filter patch are counted and measured under a microscope. The Filter & Count Method is considered to be the most accurate method of particle counting because the test is not normally affected by fluid color, air or water.
There is, however, no precision statement for the test, so the accuracy of the measurements is unknown. In addition, agglomerated particles, particle coincidence (excessively high particle counts that prevent accurate detection by the instrument), sometimes emulsified water and even air bubbles can contribute to false readings. It also is important to note that this test is extremely laborintensive and expensive. Furthermore, human error and variability can contribute to inaccurate readings and low test precision.
Light Blockage (Laser) Method–ISO 11500…
The Light Blockage Method is the most commonly used particle counting test. For this type of test, a laser is focused on a capillary through which oil flows. As particles pass through the laser, the beam is partially blocked and the transmitted light is measured by a photocell detector. The amount of light blockage is related to the number and size of particles in the sample. Similar to the Filter & Count Method, there is no precision statement for this test, so the exactitude of the measurements is unknown. In addition, compounds such as air, water and some additives which refract or impede light, can cause false readings. This method also cannot effectively measure dark fluids.
Once these tests have been completed, oil analysis providers utilize ISO Cleanliness Code – ISO 4406 (see Fig. 1). This chart has been used to establish a standardized code for quantifying oil cleanliness. In essence, it is a counting tool. The ISO range code number, or simply the ISO code, represents the number of particles found per milliliter of oil, and a single ISO code increase represents (roughly) a doubling of particles in the fluid. Under ISO 4406, an ISO code is determined by measuring and grouping particles into three categories based on their size in microns (> 4μ, >6μ, and >14μ). (To put the size of the particles in perspective, the width of a human hair is about 40μ.) As an example, the results outlined in Fig. 2 indicate that the ISO cleanliness code of the oil is 21/17/12.
Maintenance professionals typically use particle counts—along with other in-service oil analysis results—to make oil suitability decisions, based on the fluid's cleanliness rating and, consequently, its expected tendency to cause wear and premature failure. Recognizing the importance of oil cleanliness, equipment manufacturers have started to include limitations based on particle counts in their warranty specifications. A growing number of companies also include particle counting guidelines in their internal maintenance practices to ensure a strong return on their equipment investment. While monitoring oil cleanliness is geared toward improving equipment cleanliness and thus enhancing equipment reliability and life, both Maintenance professionals and Lubrication specialists have to be cautious when making decisions based on the results from these particle counting tests.
Comparison Study 1: Filter & Count Method
If samples of the same oil are tested using the Filter & Count Method with different operators, then all of the ISO cleanliness ratings should be exactly the same, right? Well, this hypothesis was dispelled when we evaluated the following four lubricant samples with increasing levels of particles at one lab with three different operators.
Cleanliness values were assessed based on counts of particles with sizes greater than 5μ and 14μ.
For clarity, Fig. 3 shows the results for particles >14μ; particle counts for both >5μ and >14μ showed the same pattern. These results indicate that even though the Filter & Count Method was conducted on samples of the same oil, the values that were generated varied by as much as two ISO codes between operators, or roughly a factor of four in terms of particle counts.
Comparison Study 2: Laser Blockage Comparison
In a second experiment, the reliability of the results of the Light Blockage Method was investigated. Filter & Count tests also were conducted for comparison. Here, 100 ml samples of a lubricant, formulated with a medium amount of additives, no polymers and a silicon antifoamant, were taken from one batch and distributed to four different particle counting labs. Each lab was given the same instructions on how to handle the samples and run the Light Blockage tests. Particles with sizes greater than 5μ and 15μ were evaluated for each test. Fig. 4 plots the results generated from the four Light Blockage tests. There was a fluctuation of as much as two ISO codes generated from these tests.
Comparison Study 3: Filter & Count vs. Laser Blockage Methods
In light of the differences between the ISO cleanliness ratings generated by both the Filter & Count and Laser Methods, several comparative tests were conducted to determine if any correlation between the results could be established.
In each test, samples for multiple batches and package styles of the same lubricant were evaluated. To ensure that the results were unbiased, these tests were conducted by a third-party commercial lab and no special instructions were given. The lab only was directed to evaluate ISO cleanliness using both the ISO 4407 Filter & Count and ISO 11500 Laser Blockage Methods. For clarity, only counts of particles larger than 14μ are shown.
First, the Filter & Count Method and Laser Method were performed on 19 samples of a lubricant that was formulated with a medium amount of additives, a silicon antifoamant and no polymers. This formulation is representative of a commonly used hydraulic fluid. The average ISO Code of the lubricant was 11.
The results of the test generated a mean delta of the ISO codes between the two test methods of 1.7, with variations in results as high as four ISO Codes. In many of these cases, Maintenance professionals would be inclined to change an oil that still could be serviceable. There appears to be a bias toward the Filter and Count method giving a higher particle count than the Laser Blockage method.
To investigate this theory further, the same evaluation was conducted on 17 samples of oil that was formulated with a medium level of additives with polymers and a silicon antifoamant. The average ISO value of this lubricant also was 11.
For this series of tests, the mean difference of ISO Codes was 2.5. In contrast to the previous sample, the results of the microscope tests were lower than the laser method more than 82% of the time. Additionally, it should be noted that the Filter & Count and Laser Blockage Methods failed to give the same test results more than 87% of the time, with differentials as high as five ISO codes. Moreover, the measurements differed by greater than one ISO code more than 50% of the time.
These significant variations establish that there is no distinct correlation between the reported cleanliness of the oil and the method of testing. Not only was there no correlation between the two methods, the accuracy and precision of both methods are clearly in question.
Comparison Study 4: Rate of Cleanliness
A comparative study also was conducted to see if comparing the results of the Filter & Count and Laser Blockage tests would unveil a discernible pattern when analyzing oils of varying cleanliness levels. The samples included clean samples (IS0= 10), samples of medium cleanliness (ISO= 12) and relatively dirty samples (ISO= 15, 16, 17). Similar to previous tests, only particles larger than 14μ are shown.
When the 26 different samples of the cleanest lubricant were tested, the mean difference in test results was 1.7. The results of testing 23 lubricant samples with medium cleanliness generated an average difference in oil cleanliness rating of 2.1, with variations as high as 4 ISO codes between results. Finally, the average difference between the oil cleanliness values of the 15 samples of the dirtiest lubricant was 1.9.
What do these results indicate? They indicate that there does not appear to be any consistency or pattern in the oil cleanliness results of oils with varying particle counts. Thus, Maintenance professionals need to be cautious when making oil suitability decisions based on particle counting results.
Sources of variation
The lack of standardization in the testing methods' practices and instruments can explain some variation in the test results. This, however, is only part of the story. There are several other factors that lead to inaccurate results. These include:
During any particle counting test, sludge, emulsified water and fibers can all be interpreted as similarly sized particles. Additionally, black particles that absorb light and shiny particles reflecting light can affect the results of the oil cleanliness measurement. There are means by which to improve oil cleanliness testing, though.
Let's now explore ways to increase test accuracy and precision, so that you can depend upon particle counting results more when making decisions about the health of, and investment in, your lubricant.
Minimizing test variability
As we've seen, significant sources of variation can occur from air bubbles, antifoam additives and water entrapped in the oil, and these "phantom particles" generate some of the largest spikes in ISO cleanliness values. With this in mind, many oil analysis companies are developing innovative ways to improve the repeatability and reproducibility of particle count tests.
The effects of air…
Because air has a different refractive index than oil, air bubbles can be measured as hard particles by Light Blockage particle counters. It has been discovered that by pre-treating the oil sample with an ultrasonic bath or a combination of ultrasonic bath and vacuum, inaccuracies in particle count as a result of air bubbles can be minimized. The ultrasonic treatment helps to remove the bubbles and provide more accurate particulate values.
Although there is not a significant change for particles greater than four microns, it is clear in Fig. 5 that by using this method, the particle count of the particles greater than six microns and particles above 14 μ has been signifi- cantly lowered after ultrasonic treatment. Thus, a more accurate assessment of an oil's cleanliness can be determined by removing entrained air.
Using an antifoamant…
Foam is created by the combination of air and a lubricant and can compromise the performance of a product, possibly leading to equipment problems. Many lubricant manufacturers include antifoamants in the formulation of their products. Although these ingredients can improve equipment performance, they also can contribute to inaccurate particle counts when a sample is tested.
Fig. 6 illustrates how an antifoamant substantially increases the cleanliness rating of an oil and how this interference can be avoided through the use of a diluent—a miscible liquid or solvent used to dilute and lower the viscosity of the sample. By obtaining a more accurate value, Maintenance professionals will be able to make better decisions about whether or not they need to change their oil.
The effects of water…
Water tends to have a deleterious effect on lubricant and equipment performance, potentially leading to frequent component failure. When testing new or used oil, the water in a lubricant can increase the particle count in a laser particle counter.
Fig. 7 details the increase in the oil cleanliness rating (indicating "dirtier oil") when water is added to a lubricant. Therefore, water is shown to affect the "hard particle" count, which we know should not be the case. Similar to the case of antifoamants, when a diluent is added to the same lubricant, the oil cleanliness rating is reduced, thus generating a more accurate particle count.
While there are currently no ASTM standard test methods for measuring the cleanliness of lubricating oils, innovative modifications to the Laser Blockage method can greatly increase the accuracy and repeatability of the results by eliminating test interference from entrained air, water and antifoamant additives. Armed with better in-service oil analysis results, Maintenance professionals can make better decisions about the suitability of an oil.
Bernie Koenitzer and Clint Smith are technical service advisors with Imperial Oil Ltd.
Alex Bolkhovsky and Dr. Tim Nadasdi are products technical advisors with ExxonMobil Lubricants & Specialties.