Blending lubricants is said to be an art form as much as it is a science. The selection of one of a dozen different additives for one of a dozen different types or blends of base oils requires a chef’s nose for effectiveness and some good testing capabilities to prove the recipes work.
Testing based on ISO and ASTM standardized methods is routinely performed to verify the usefulness of a new lubricant recipe. Over the years, many different tests have been developed precisely for this purpose. As new products are developed and the application of existing products changes with operating conditions, then the testing methods must change as well.
The test for oil filterability was developed in response to customer and industrial demands for finer filters to produce cleaner systems. This article looks at the development and use of the oil filterability tests, especially those applied to paper machine oils (PMOs).
The filterability characteristic of a fluid can be defined as its ability to pass through a filter without giving rise to undue pressure drop which will lead to loss of useful life.
From this, the reader can conclude that filterability is a user-driven requirement aimed at enhancing operating efficiency and reducing costs.
If a hydraulic fluid has poor filterability, the filter element will block up rapidly, perhaps frequently. Conversely, good filterability of the oil does not guarantee good filter service life. This presents the question, “Is poor filterability the fault of the fluid, or the fault of something else?”
Poor filterability is caused by unwanted material carried by the lubricant being removed by the filter. This uses up useful life. Rapid removal can plug the filter in seconds. Slower removal is progressive, taking days, weeks or even months. The net result is that the filter blocks quicker than the user wants.
Possible causes of poor filterability are summarized below:
High levels of dirt, or a level where the cleanliness control measures are ineffective, such as not maintaining existing filters or those in a newly constructed system.
Deterioration of components, such as seals which allow dirt (mainly process materials and chemicals in paper machine applications) to ingress into the system.
Debris caused by increased component wear.
Thermal stressing of the oil during normal operation which accelerates oxidation and promotes the formation of degradation products. This forms precipitates, carbon particles, gums, waxes, sludges etc. Cavitation and dieseling are prime causes, especially in aerated systems.
Ingression of liquid contaminants which react chemically with the oil and its additives to produce a precipitate. This can be either maintenance or process induced. Water in a mineral oil is probably the most common liquid ingested into paper machine lube systems. The tolerance of oils to water has increased over the years, but the chemicals that the water contains are of greater concern. This makes the formulation of oils an extremely difficult process, because it can require more than 30 different chemicals. Because papermaking is a highly competitive business, manufacturers are reluctant to identify many of the chemicals involved. What may be acceptable today may be unacceptable tomorrow as changes are made to the process and the chemicals contained in the water.
Preservatives and coatings, which are washed from the component during initial system operation.
Inadequate blending of the oil that leaves solid particles which are removed by the filter.
Insoluble additives. Although most oil additives are usually soluble and pass through the finest filters, some special additives do not. The most common are the polymeric additives used to improve the fluid’s viscosity/temperature characteristics commonly called viscosity index (VI) improvers. At high temperatures, these form long chained, branched molecules to increase the oil’s viscosity. At lower temperatures, generally less than 86°F (30°C), they clump together to form relatively large particles that can be filtered out. They can cause filterability problems even with relatively coarse filters; for example, those rated at ß12=1000, only to see the problem disappear as the oil heats up. Temperature effects are not much of a problem with paper machine oils, both hydraulic and lube oils. They tend to run at high temperatures, greater than 140°F (60°C), and for continuous periods.
Cross-contamination of lubricant types and brands may create problems given complex chemical formulations.
Filterability testing has been the subject of much discussion and research since Dr. Leonard Bensch introduced the Bensch Filterability Test in 1971.1 The driving force for the development of this filterability test was the introduction of 3-micron filters - revolutionary 30 years ago. Unfortunately at that time, oils and hydraulic systems were not designed for such fine levels of filtration. Consequently, the filters frequently became blocked and interest diminished rapidly. Subsequent interest, at least for a while, was shown by the more forward-thinking users who had the infrastructure and design to support the use of these filters. The benefits that the 3-micron filters would bring were ignored because their service life was unacceptable.
The test is based upon the volume that can be vacuum filtered through a 47mm, 1.2-micron analysis membrane at room temperature before blockage. This gave the Filterability Index (FI) as:
For a good oil, 1000 mL should pass, giving a FI of greater than 104. A time constraint was included, but rarely used in the author’s experience. This new test was investigated and guidelines were issued for FI values to give the following service life ratings: good (FI > 70), acceptable (FI 70 to 30), and poor (FI < 30).2
The Bensch filterability test generated considerable work and interest among user groups, oil companies and academics, resulting in the development of tests for individual or group requirements. However, it has taken nearly 20 years to develop, refine and agree on an internationally acceptable test. This is ISO 13357 and involves pressure filtering the oil at room temperature through a precise 1-micron analytical membrane filter.2,3 Two tests are described, in both wet and dry conditions. One test is more severe than the other. An upper limit of 100 cSt is placed upon the oil tested.
What are the options for determining the filterability characteristics of PMOs where the viscosity grades can vary from ISO VG 100 to ISO VG 460? If the Bensch method is tried, the very high test viscosity means that the end point is reached at or shortly after the start of the test. The ISO 13357 test uses pressure to force the oil through the membrane, generating high pressures. This is something to beware of, because the glass apparatus poses a health and safety risk.
The need for a filterability test for PMOs was recognized by Pall Corporation, and to respond to customer demands in a more scientific manner, it developed two filterability tests for PMOs. These have since been adopted by the oil companies as a development tool.
This was a modification of a test used in the early 1970s. It was developed to test oils under their working conditions, such as:
1. Same media grade
2. Same operating temperature or range of temperatures
3. Similar flow density and
4. Operation in a recirculatory mode.
The test is performed in a bench-sized test rig seen in Figure 1. This allows two separate tests to be performed.
In a standard test, a 1-gallon (4 L) charge of the test oil is added to a clean test rig fitted with a new filter similar to the one used in the field operation. The oil is circulated through the bypass and brought up to the test temperature, usually 140°F (60°C). Once stable, the flow is diverted to the test filter section and continues recirculating until either the filter element blocks or the differential pressure stabilizes. Values of differential pressure, temperature and flow rate are recorded at regular intervals.
In Figure 2, the differential pressure characteristics of four example fluids are given. At the onset of flow through the filter, the initial pressure rise is due to the viscous loss through the filtration media and is proportional to the grade of media used. Thereafter, the differential pressure should remain stable as in Curve A, indicating excellent filterability.
On the other hand, Curve B represents an extreme case of poor filterability where blockage of the element occurs in a relatively short time. In most cases, the differential pressure will continue to increase after the initial value as the element removes particulate contaminants. This usually takes about six to ten passes and should then stabilize (Curve C1). However, it may take longer (Curve C2) because of the presence of very small particles, called silt.
If there is a filterability problem, a sample of the oil is analyzed and the solids separated by filtration. The membrane filter is then examined using optical or scanning electron microscopy to determine the composition ofthe material.
Whether or not these characteristics will present a filterability problem depends upon the system and its fluid volume. A small increase in differential pressure would not affect element life in a small volume system, such as a truck transmission, because the source of the blockage will be quickly filtered out and may not appreciably affect the element life. The same characteristic, however, could be a problem in a paper machine lube system where volumes of 793 to 7,925 gallons (3,000 to 30,000 L) are common.
Example C1 shows an increase in differential pressure of about 1.5 times. Approximately 33 percent of useful life has been used up with the passage of one gallon (4 L) through a filter of given area. As a result, the expected length of service or the area of filter media required to clean-up the system can be estimated.
Filterability problems can occur with low temperature operating conditions. Problems associated with high VI oils are somewhat predictable. However, the presence of gels, waxes and lacquers may be less obvious in the oil whose solubility can be affected by temperature. The dynamic filterability test can easily evaluate such characteristics.
There are no standard tests for these characteristics. The tests are custom-designed. The following is an example of a procedure usedto check the filterability of an oil across a temperature range.
1. Add the charge of oil to the reservoir fitted with a filter of the same grade used in the system. Circulate through the bypass and raise the temperature to 158°F (70°C).
2. Divert the flow through the filter and circulate for 20 minutes to remove the dirt. Operations at this high temperature will allow temperature-dependent components of the oil to pass through.
3. Replace element with a new one and run until conditions have stabilized.
4. Reduce temperature to lowest possible; 50°F (10°C) is desired. Monitor the flow rate, oil temperature and element differential pressure as the oil cools.
5. Run at minimum temperature for 15 minutes.
6. Raise temperature to starting value, recording measured parameters.
7. Correct for changes in flow and plot the differential pressure against viscosity. Ideally, both curves should be proportional to viscosity and overlay one another.
The Filterability Index Test for Paper Machine Oils (FIT-PMO) was developed in 1993. It provides a simple screening to identify oils which would give a satisfactory life when used with modern filter elements featuring finer levels of filtration (less than 6 microns).4 It was designed to be a bench-top test, to use little oil, and to test at high temperatures. The size of the reservoir in paper machines means that it takes a long time for the passage of one complete volume (typically 25 to 50 minutes), such that in a filterability context, a single-pass test is more appropriate. The test is carried out on a single-pass setup as shown in Figure 3.
This arrangement uses a 47mm diameter, 3-micron nylon membrane filter, selected to ensure test repeatability and reproducibility. After conditioning the test oil by stirring for one hour at 145°F (63+1°C), the oil is pumped through the membrane filter. The initial differential pressure is noted and collected in the graduated vessel. The test continues until the differential pressure reaches 25 psi (1.7 bar) and the total volume passed is recorded on the Filterability Index (FI). It is common practice to limit the test to the passage of 4,000 mL. Filterability Index ratings are shown in Table 1.
Volume Passed (mL)
Filter’s Length of Service
4,000 (4 L or 1 gallon)
2,000 (2 L or 2 quarts)
1,000 (1 L or 1 quart)
Table 1. Guidelines for Filterability Indexes for PMOs
As with the dynamic test, the causes of poor filterability should be investigated by examining the membrane for the presence of abnormal quantities of particulate, or the presence of gels or waxes. Once identified, the source can be located and corrective measures taken. The test can be performed on new, serviced or water-aged oil.
Improvements in oil formulation, fluid systems design, and more effective contamination control techniques have made filterability problems caused by the oil less common as they were 10 to 15 years ago.
However, this has not solved all filterability problems. Short filter element life should now be seen as an indicator of other problems. This is an opportunity for mill personnel, filter supplier and the lubricant supplier to work together to resolve the problem and create an efficient process.
In the experience of the author, those mills that have adopted a practice of Total Cleanliness Control™ or proactive maintenance approach to their mill generally suffer fewer filterability problems. Typically, solving a filterability problem requires the cooperative efforts of the mill personnel, the filter manufacturer, the oil manufacturer and the system builder working together to rectify the root cause.
1. Bensch, Leonard. (1977). Verification of a Hydraulic Filterability Test, Paper P73-CC-8. Seventh Annual Research Conference. Oklahoma State University: Stillwater, OK, USA.
2. ISO 13357. Petroleum Products - Determination of the Filterability of Lubricating Oils - Part 1 (dry oil), Part 2 (wet oil).
3. Day, Mike J. (1997). Increasing Profitability Through a Policy of Total Cleanliness Control.London: IMechE Seminar on Cleanliness Control in Fluid Systems.
4. Pall Corporation. (1995). Pall Filterability Index Test for Paper Machine Oils, Pall FIT-PMO. Issue 4, Pall Corporation: Port Washington, NY, USA.