The order in which oil analysis tests are run affects the quality of results. Achieving accurate measurement requires great care in sample handling at each step in the process, including handling at the laboratory.
Numerous factors can influence the quality of oil analysis decisions (Figure 1). The location from which the sample is drawn, the sample container, sampling hardware and sampling procedure all affect the degree to which the oil sample actually represents the oil in the machine.
Likewise, the sample must be carefully agitated at the laboratory to restore it to its representative condition. Then, properly skilled technicians must analyze the sample shortly after agitation using well-calibrated instruments to avoid resettling of contaminants and wear debris. A failure at any point in the process can compromise the data’s quality.
While these general comments apply to all oil analysis tests, some oil analysis parameters are more sensitive than others to the various possible infractions. Generally speaking, the homogenous properties of the oil - those measurable properties that result from material dissolved in the oil or from the oil molecules themselves - are much less sensitive than the measurable properties of material suspended in the oil.
Achieving accurate measurement of suspended materials on the other hand, requires great care in sample handling at each step in the process, including handling at the laboratory. An oil analysis report is actually the compilation of 80 or more values extracted from 10 or more different tests. The sequence in which these tests are run can significantly affect the quality of the information. It is necessary to run the most sensitive test first, saving the tests that are comparatively unaffected by handling for last.
Despite the importance of proper sequencing of oil analysis tests, infractions are not uncommon. Due to lack of knowledge, laboratory layout, laboratory information system (LIMS) setup and a host of other reasons, tests are not performed in the optimum order.
Sometimes, the reason is simply “because we have always done it this way.” The following shows common oil analysis tests divided into four tiers, starting with the most sensitive tests (Tier 1) and proceeding to the least sensitive tests (Tier 4). The order in which the tests are run within a defined tier can vary depending upon the situation.
Whether the optical or pore-blockage method, particle counting is the test that is most sensitive to handling errors. For this reason, it is in a class by itself and therefore, should be performed first whenever it is on the test slate. Simply opening the bottle a time or two, particularly if the lid is mishandled, can ingest sufficient debris to significantly alter the particle count.
Even drawing a fraction of oil from the sample bottle using a clean device can introduce a large amount of contaminant into the oil, and should be done with great care. Likewise, the particle count should be performed as quickly as possible after agitating the sample on a paint shaker. If the test slate calls for screening for water with a crackle test, and the lab employs an optical particle counter, it is appropriate to perform the crackle test first, just prior to performing the particle count.
This is recommended because it is necessary to determine if free moisture is present, which precludes the use of the optical particle counter without first chemically masking the water or dehydrating the oil.
In the normally climate-controlled atmosphere of the laboratory, water can evaporate from the sample, causing the measured water concentration to decrease. Free and loosely emulsified water may tend to settle out, so testing shortly after agitation is important. Some oils, particularly ester-based oil, tend to be hydroscopic and absorb water from the atmosphere, so sample bottles should be kept capped at all times.
The crackle test identifies the presence of free and emulsified water. Therefore, it is important to perform the test shortly after agitation and before water has had a chance to migrate from the sample to the climate-controlled air in the laboratory.
When FTIR is the primary screen for water contamination in the oil, the test should be performed early in the sequence for the same reasons discussed in the water by Karl Fischer and crackle tests. When FTIR is not used as a screen for water, it can be performed during tier four.
Ferrous density tests, which measure the concentration of magnetic debris in the sample, are less affected by handling-induced ingestion than particle counting simply because the sample is more likely to ingest dirt than ferrous wear debris in the typical lab environment. However, because the test is measuring the concentration of comparatively large and high-density particles that can quickly settle out, it is advisable to resuspend the particles in the sample using a paint shaker before performing this test.
Like ferrous density tests, these tests target wear debris, and the risk of wear debris ingestion within the laboratory environment is comparatively low. However, also just like ferrous density, ferrography and/or patch microscopy should be conducted soon after agitation due to the propensity of large, high-density wear particles to settle to the bottom of the sample.
Optical microscopy analysis of wear particles is usually an exception test that is performed in response to an abnormal particle count, ferrous density and/or atomic emission elemental spectroscopy reading. If analytical ferrography/patch microscopy is first triggered by elemental spectroscopy, which is a third-tier test, it is important to perform the ferrography before proceeding to the fourth-tier tests.
In this case, it may be necessary to first resuspend the particles in the oil using a paint shaker, depending upon how much time has elapsed and the oil’s viscosity.
Like particle counting, atomic emission elemental spectroscopy is in a class by itself, but in the third tier. Because the test is highly biased toward dissolved and submicron particles, debris induced during handling, normally larger particles of dirt and various organic materials, is unlikely to have much effect on the test results.
Likewise, if a particle can quickly settle in the sample, it is probably too large for the spectrometer to measure; so performing the test a few minutes after shaking the sample should not dramatically affect the results. However, the larger particles in the oil do have some influence on the results from elemental spectroscopy, so it is best to perform this test in advance of tackling the tests in tier four.
Organic and other acids that increase the oil’s acid number (AN), or additives that might affect the oil’s AN and/or its base number (BN), are typically well-dissolved in the oil. In addition, in the lab environment, the oil is unlikely to ingest contaminants that would adversely affect results. Neutralization number tests may be run relatively late in the analysis sequence.
When not used to screen water, FTIR may be tested in the final tier. The only undissolved property (other than water) that is commonly tested with FTIR is soot in engine oil. Soot particles tend to be very small and well-suspended by the oil’s dispersant additive package, so the risk of erroneous data is relatively small. When FTIR is used as the primary test to screen for water, it should be performed during tier two.
Flash point is employed in used engine oil analysis to detect fuel dilution. Fuel is soluble in oil, so the property is not sensitive to poor agitation. Likewise, it is unlikely that fuel will be ingested into the sample from the typical lab environment. As a note of caution, heating the entire sample of oil can vaporize the lighter fractions of the fuel, raising the measured flash point. Therefore, care must be taken not to heat the sample prior to flash point testing.
Unless the sample is diluted in the laboratory, viscosity is almost impervious to problems related to agitation and handling. It can safely be saved for last in the sequence.
This is a general guide for sequencing the used oil analysis tests most commonly used in the laboratory. There are sure to be instances where adjustments are required. Also, the fact that a particular test is less sensitive to handling and agitation practices than another should not be used as an excuse to handle the sample casually. It is ideal to test well-agitated samples quickly, using the suggested sequence to reduce the likelihood of error.
If you are interested in a test that is not identified here, employ the following rule of thumb: tests that measure materials suspended in the oil are more sensitive than tests that measure dissolved components of the oil itself. If you still aren’t sure, send Noria an e-mail or give us a call - we will be glad to help you if we can!