Impact on Filter Performance Ratings and Particle Counts

For many years, General Motors (GM) manufactured Air Cleaner Fine Test Dust, otherwise known as ACFTD or Arizona Road Dust. A few years ago, GM discontinued selling ACFTD, and things have not been the same for the filter industry since. ACFTD was used as a challenge medium to load filters, whether for air or liquid, as well as the calibration of particle counters.

While optical automatic particle counters were first developed for processes other than oil contaminant measurement, they utilized polymer spheres rather than ACFTD. Automatic particle counters (APCs) typically calculate the shadow area into a radius rather than a square or rectangular shape. This was problematic because these spheres swelled in oil and did not truly represent the particle size distribution in used oils. Therefore, ACFTD was originally selected for instrument testing and calibration methods.

Once GM announced it would no longer supply ACFTD, the National Fluid Power Association (NFPA) started a project to replace it with an equivalent dust. The main objective was to find a dust that was not only close to ACFTD in composition, but also more controlled than ACFTD. The main interests were in sizing and counting particles. There were also disadvantages with ACFTD: It contained too many particles smaller than five micrometers (µm), particle size distribution below five µm was questionable, and there was no known standardized calibration method for the dust. The project that the NFPA started was intended to take the new ISO Medium Test Dust (ISO MTD) and let the National Institute of Standards and Technology (NIST) define the size distribution for this dust. NIST completed its analysis about three years ago and gave the results to NFPA. The NFPA standards committee T2.9 and the International Organization for Standardization (ISO) committee TC131, in conjunction with working groups in other countries, have finalized the calibration fluid. It is now available as Standard Reference Material (SRM 2806). The main impact will be in the particle counter calibration procedure. To minimize confusion, ISO has established a new standard for the calibration of liquid automatic particle counters. ISO 4402, which uses ACFTD for calibration, is replaced with ISO 11171. Table 1 shows the difference in particle sizes due to NIST calibration.

Impact on the End-User
For some, this may be a simple change. For others, it will be a bit more costly. Still, such changes will be worth the effort, bringing a greater level of accuracy and fairness to the business of oil analysis. It will also help level the playing field for filter suppliers and particle counter manufacturers. The following changes will be observed by most:

  1. Filter media defined as ß3=200, ß5=200, etc. must be redefined as ß5.1=200 and ß6.4=200. Although the media remains the same and still offers the same protection to the hydraulic system, the media rating will change. Filter manufacturers will have to reprint their catalogs and customers will have to change their procurement documents. On paper, some filter media will appear coarser. Using the above example, what used to be 3-micron absolute (assuming ßX=200 is absolute) will now be 5.1-micron absolute (see No. 3 in Table 1).
  2. The ISO contamination code has also changed. The old standard for this code is ISO 4406:87. In this system, the contamination level of a fluid could be represented using a two-number system. The new standard for coding the level of contamination by solid particles is ISO 4406:99. It is now a three-number system. It was anticipated that the ISO 4406:99 would introduce a third category to measure the level of particulate greater than 2 µm to better address the fluid power industry. However, with the new NIST-calibrated dust and sizing, sizes 2 µm, 5 µm and 15 µm are now 4 µm (c), 6 µm (c) and 14 µm(c). This change does not affect many users because a previous ISO 4406:87 value of 17/14 should still measure out at 17/14 when using automatic particle counters calibrated by the new standards. However, documents referring to the actual number of particles measured rather than the ISO 4406:99 code, except publications in which the ISO code is referred to as 5 µm and 15 µm (still in use for microscope particle counting), must be changed to 4 µm (c), 6 µm (c) and 14 µm (c). The (c) is important, because it represents “certified” - the new method of calibration and reporting.
  3. The Multipass test (ISO 16889) determines a filter’s performance, measuring the ability of the filter to efficiently remove particles of test dust over a wide range of particle sizes. The test reports a series of Beta ratios for the filter as opposed to a percentage efficiency. The Beta ratio is defined as:
    ßx = Nu(x)/Nd(x)
    ßx = Beta ratio or filtration ratio at X micron
    Nu(x) = number of particles upstream at X micron
    Nd(x) = number of particles downstream at X micron
    Where X is the determining size of particle for the Beta ratio.

Figure 1. How the ISO Solid Contaminant Code Changed - Effective December 1999

Performance of a filter media per ACFTD and ISO MTD when plotted on this graph will yield two different profiles. Even though the media has not changed, lines showing the performance of the same media using ACFTD and ISO MTD will be different. This may cause some confusion in filter media ratings. Most filter manufacturers show the performance of their media graphically. The X-axis usually represents particle sizes in micrometers and the Y-axis represents filtration or Beta ratio. The X-axis is linear and the Y-axis is a log scale. Historically, filter manufacturers used a nominal or absolute rating, the latter giving the impression that filters with an absolute rating stop all particles larger than the quoted size. To overcome this, manufacturers are now required to show the particle size at Beta equal to 2, 20, 75, 100, 200 and 1,000. The slope or curve of this plot of size against Beta ratio will give a clear indication of overall filter performance, thus making for better evaluation.

It is possible that manufacturers of equipment or components who are advising their customers to use filters exhibiting ßX=200 will now have to change their recommendation to ßY(c)=200. Y may be greater than or less than X. Bear in mind that the equipment has not become more tolerant of dirt, but it is adjusted to account for the shift in dust from ACFTD to ISO MTD.

To differentiate between the old and new dust, ISO is proposing a change in nomenclature. For example, if the Beta ratio with ACFTD were ß5=200 or ß3=200, then the proposed nomenclature will be ß6.4(c)=200 or ß5.1(c)=75. Subscript (c) denotes the new dust. This will help alleviate any confusion. It is desirable that filter manufacturers publish both old and proposed ratings during transition until all concerned are properly educated. The real end-user benefit is the ability to easily differentiate between a good filter and a poor filter across the full particle size range rather than at one point; in other words, a poor caking filter can be determined more easily from the new data.

New NIST traceable calibration will also affect particle counters. Older model counters currently in use may not be able to count particles smaller than 5 µm(c). Those wishing to count smaller than 5 µm(c) will have to purchase new counters or count optically using a microscope.

This is an important step in particle sizing and calibration. For the first time in 35 years, the industry has a NIST-traceable Particle Counter Calibration Test Dust. ISO has approved the new calibration document (ISO 11171). Once participating countries begin using this document, all involved should have a common and accurate language of cleanliness.

Every change requires an adjustment period because transition is neither quick nor easy. The only solution is to educate as many people in the shortest possible time. To learn more about the impact of these changes, contact NFPA (or similar organization in your country), SAE, ISO or the filter and particle counter suppliers.

Further Reading

Bensch, L. “How the New ISO Particle Count Standard Will Affect You.” Practicing Oil Analysis magazine, May-June 2000.

Day, M. “Clarifying the New ISO Contamination Filtration Standards.” Practicing Oil Analysis magazine, September-October 2001.

Back issues of Practicing Oil Analysis magazine can be viewed online at

A Dust By Any Other Name
Arizona sand has been used for testing filtration, automotive and heavy equipment components for decades. A variety of names have been applied to Arizona sand including Arizona Road Dust, Arizona Silica, AC Fine and AC Coarse Test Dusts, SAE Fine and Coarse Test Dusts, J726 Test Dusts, and most recently ISO Ultrafine, ISO Fine, ISO Medium and ISO Coarse Test Dusts. Many military and industrial specifications require use of Arizona Test Dust and refer to one or more of the above names.

History of Test Dust.