Ashes to Ashes, Dust to Dust: The Use of Test Dust in NSF/ANSI 42 and 53
By Rick Andrew
Dust is not something that most of us spend much time thinking about. Occasionally, we need to remove it from our furniture. If we ride motorcycles or off-road vehicles, we sometimes get covered by it (and even eat it). Otherwise, dust is not typically at the forefront of our attention. This also holds true when we consider testing of water filters. Testing these filters with contaminated water seems intuitive, but who would consider using dust as the contaminant? Most water treatment dealers do not get calls for help from homeowners with dusty water problems!
Despite this seemingly odd connection, the use of dust in testing water filters dates back many years and has served its purpose very well. Let’s take a look at this unique use of something that most people consider a nuisance (if they consider it at all): dust.
Why use dust?
Dust is certainly not a common drinking water contaminant, so why use it for water filter testing? The answer is that although dust itself is not a water contaminant, it does have properties of two commonly occurring contaminants: particulate matter and turbidity. Dust can be a source of both.
However, the main purpose of dust in terms of water filter testing is to provide a source of clogging to test mechanical reduction properties of filter systems. These mechanical filtration properties are most stringently tested when pressure drop is high and flow rate is decreased due to clogging, so the purpose of the dust is to clog the filters.
Yes, the commercially available dust is relatively expensive, because it is painstakingly processed to meet precise specifications. This processing involves meticulous separation and sorting to create consistent particle size distributions. These particle size distributions are described by ISO standards, which are summarized in Table 1 and Figure 1. The exception here is the nominal test dust, which is a custom dust for the water treatment industry. The original reason for these ISO standards for dust was not for testing of water filters at all, but rather testing of automotive filters, including air filters for engines. In fact, this dust was originally called Arizona Road Dust due to its origin in that state; then it took on the name AC Test Dust from the automotive supplier. The water treatment industry capitalized on the development of the auto industry’s dust specifications by adopting their use in NSF/ANSI 42 and 53.
Dust has different roles
When conducting nominal particulate reduction testing under NSF/ANSI 42, laboratories use the specified test dust to clog the filter. They also count the dust particles in the relevant size range using a laser particle counter to determine the efficiency of the filter. The particle counts in the influent stream are compared to those in the filtered effluents to calculate a percent reduction; 85 percent reduction is required for manufacturers to make the claim of nominal particulate reduction. (Unlike some other turbidimeters, a nephelometric turbidimeter monitors light reflected off the particles and not attenuation due to cloudiness.)
Likewise, the nominal test dust is used to measure the turbidity caused by these particles in the influent and effluent streams. (In order to make the claim of turbidity reduction, filters must reduce an influent of 11 ± 1 NTU to d ≤ 0.5 NTU.)
Asbestos reduction and cyst reduction testing is different. For these, the nominal test dust is used to clog the filters and create pressure drop; however, actual asbestos, live Cryptosporidium or polystyrene microspheres are used as the test particle (as appropriate to the test being conducted). This means that the laboratory will change influent streams throughout the test, depending on whether they are about to collect influent and effluent samples or whether they are clogging the filter to the next sample points. The sample points are based on a percent reduction of the initial, clean system flow rate, as they are for nominal particulate reduction and turbidity reduction testing. To make claims of asbestos reduction, 99 percent filtration efficiency is required; for cyst reduction, 99.95 percent filtration efficiency is required.
The test dust is used for clogging instead of clogging the filter exclusively with asbestos, live Cryptosporidium or polystyrene microspheres because of the cost. Although test dust is relatively expensive, just imagine how many live Cryptosporidium organisms (each about five µm in size) that it would take to clog a 10-inch carbon block filter to a 75 percent reduction in flow rate!
Dust specs matter!
Whether the specific test dust being used can influence results is a question regularly raised in water filter testing. To find the answer, over the years, several different studies have been performed.
One of these, conducted a few years ago by the NSF laboratory and several others, investigated whether the specific clogging dust used when conducting cyst microsphere reduction testing under NSF/ANSI 53 made a difference in filter performance. Filters from the same production lot, designed and manufactured to be on the razor’s edge of passing or failing the test, were distributed to participating laboratories. The laboratories tested them according to the standard using the nominal test dust and also tested more filters from this lot with an identical protocol except that ISO Fine test dust was used to clog the filters.
The study results indicated that the specific test dust used influenced the results significantly. Testing with the nominal test dust was determined to be a more stringent test. Several explanations were proposed by the study participants, which centered around the smaller average particle size of the nominal test dust allowing it to penetrate the filter better and disrupt any electrostatic influences in retention of the cyst microspheres.
In 2007, a second study effort dealt with the issue of more than one test dust being acceptable for use in the same test. Under NSF/ANSI 42, either ISO Fine or ISO Coarse test dust could be used to test for nominal particulate reduction, class III. However, these test dusts have considerably different particle size distributions. The ISO Coarse test dust tends to clog the filters faster due to the larger particle size and to create a different filter cake than the ISO Fine test dust due to the particle size distribution difference. This can cause differences in test results. The task group addressing this inconsistency proposed to specify the use of only ISO Coarse test dust for this test. This change will publish into NSF/ANSI 42 soon and will allow for a more consistent test both within and across laboratories.
Time to hit the dusty trail
One of my favorite training sessions with new certification project managers is the one about test dust. The appeal for me is that in every day life, dust is incredibly mundane. Yet for the purposes of water filter testing, it is a highly studied, technically precise, very important tool. I really enjoy sitting down in the training room with fledgling project managers and telling them that today’s training session will be about dust. The looks on their faces are priceless. Once we get into the details and go back into the laboratory to look at the various types of dust, however, the session generates significant interest and a new appreciation for… dust.
I hope you, the kind reader, have had a similar experience as you’ve spent these last minutes reading an article about something until now you may have thought of only in the context of an annoyance that you wipe off your furniture every week…or two…or at least once in a while.
About the author
Rick Andrew is the Operations Manager of the NSF Drinking Water Treatment Units Program. Prior to joining NSF, his previous experience was in the area of analytical and environmental chemistry consulting. Andrew has a Bachelor’s Degree in chemistry and an MBA from the University of Michigan. He can be reached at 1-800-NSF-MARK or email: Andrew@nsf.org.