Filter Ratings - what they mean (and don't mean)

[Rod Nikkel]

The question I was trying to answer was how efficient these filters are right away, since our brand new filters were passing a lot of dust. Two weeks later, this is what I have found out:

Straight out of the box, the 80/20 blend filters (the Wynn 9L300BL or the Camfil Farr 211736-1 HemiPleat) only capture about 50% of the 1 micron dust, and less than 30% of the .5 micron dust. This link on the Wynn site http://www.wynnenv.com/filter_efficiency.htm shows the actual results when the filters are tested when brand new, and I have received information from Camfil Farr about their filters which show very similar results. This means that until the filters are used or pre-seasoned, they pass a LOT of the very dangerous dust. The rating given to them is MERV 10 (despite what the Camfil Farr brochure says about them being MERV 12 - that is old data). Here is a good link that helps to understand what a MERV rating means: http://allergyclean.com/article-understandingmerv.htm

So then, why do they claim 99.99% efficiency at 0.5 microns as Wynn does on the link above and Camfil Farr does here http://www.farrapc.com/literature/pr.../hemipleat.pdf ? This last link gives the answer in the added words "by weight". This is from the old ASHRAE 52.1 test. I quote from a Camfil Farr Technical Service Bulletin: "A 5-micron size particle has a relative weight of 125. 125 1-micron size particles have a relative weight of 125. If 126 particles (one 5-micron & 125 1-micron particles) are fed to this filter, the one 5-micron size particle is captured and the 125 1-micron size particles pass through. This filter may be deemed efficient at removing 50% contaminant by weight. However, the efficiency of this filter by particle count is 1/126 of 0.088% or less than 1%." (italics added) So the newer test checks actual particles while the older one tests weight. So when using this test, it can look like most dust is being caught when actually most of the smallest (more worrisome) dust is passing through but almost all of the larger, heavier particles are being caught. This is still a very common way of describing the capacity of filters in the industry, but then people in the industry should know what it really means. The rest of us are mislead because we don't understand what it means and expect it to mean 99.99% of the particles down to 0.5 microns (and the Wynn site doesn't even use the words "by weight").

Now this is BRAND NEW. The Wynn links shows (and Camfil Farr has sent me the testing on theirs) that as the filters are loaded their efficiency increases. The test puts 30 gms of material into the filter (doesn't seem to depend on size of filter and they don't say what size particles are in that 30 gms) initially, and then repeats that 4 more times without cleaning the filters. As there is more material in the filters, they work better. Neither give data on the increase in resistance, and therefore the decrease in air flow, due to the loading. But this is what is happening in practice in our shops, and I know that our old filters did keep the air really clean as tested by the Dylos.

In fact, these companies (at least Camfil Farr) actually sell "precoat material" to season your filters with before you use them. This dust has an average size of 40 microns, and the instructions say: "If the material to be collected is small in particle size (predominantly under 5 microns), oily, sticky or otherwise unfriendly to the cartridge, longer cartridge life and higher initial efficiency can be achieved by pre-coating the cartridges." The idea is to coat the filters with larger particles that trap the smaller ones and, from the sounds of it, make it easier to clean the smaller particles out.

It finally got cold enough that we had to start using the filters. Within a couple of hours using the grinder the number of 1 micron particles coming out of the filter was between 3500 and 4500 as opposed to the 12,000 right at the beginning. Still far from good, but a whole pile better. And they will continue to get better as we get more dust into them. We don't have the ability to test to see increase in resistance or decrease in air flow due to the loading, but lots of air is still moving for sure. It would be interesting to know the difference though. And it will also be interesting to see what comes through after we clean them the first time - if they stay relatively good or go back to passing more fine dust again.

So the practical application of all this? Brand new filters of this type still spew a lot of fine dust for a little while. So wear your masks for a while when starting new filters and be prepared to do a good shop clean after they get seasoned enough to trap the smaller particles. Or go the extra bucks and buy the filters labelled MERV 15 (Wynn 9L300NANO) or MERV 16 (Camfil Farr PTU-Poly-Tech Ultra High Efficiency) that really do trap 95% plus of the 1 micron sized dust right out of the box. And look forward to spring when you can vent outside again.

[Paul Wunder]

Most useful information. Thanks for posting. Will look into the new filters.

[mreza Salav]

Very interesting read. I didn't know all these were after you had seasoned the filters long enough. What does a "load" mean in those charts? filling a drum? what size?

How accurate is the rating for their Nano filters?

Thanks for posting this.

[Rod Nikkel]

The load is a specific testing procedure, using 30 gms of dust on their test filter. (I wonder if their test filter is a specific size of the material, or an actual filter. They don't say.) Then they keep adding 30 gms and calling it another load. The filter is not cleaned.
The MERV rating is a standard measurement based on particle size, not weight. According to the charts on the Camfill Farr ones rated at MERV 16, the filter right out of the box will get about 94% of the 0.5 micron dust and about 98% of the 1 micron dust. Interestingly enough, the resistance of the two filter types is very, very close with the MERV 16 having slightly less resistance (and therefore slightly more air flow) than the MERV 10 ones most of us use. What I don't have any information about is what happens to that air flow with loading. Does it decrease a lot more with the finer filter when it gets dust into it? And how well do the filters clean, and therefore what is the lifespan of each before they present too much back pressure and need to be changed out? No idea on the answers to those questions.

[Rod Nikkel]

I've been reading the testing procedure more carefully and have found out some more interesting information. I'm looking at the chart on the Wynn site here http://www.wynnenv.com/filter_efficiency.htm . The initial testing is with a totally clean filter. The first "loading" is 30 gms or and increase in 0.04 WG (inches of water gauge) of resistance, whichever comes first. The second loading is till you get an increase in resistance to 25% of the "manufacturer's recommended final pressure drop". The third loading is to 50%, the fourth to 75% and the fifth to 100% of the "manufatcturer's recommended final pressure drop". In other words, the final loading on this chart is when the filter is a clogged as the manufacturer ever wants it to be! I have the same kind of chart on the Camfil Farr filter and it looks to be slightly better, but then that would depend on what they recommend the "final pressure drop" point to be. On the Camfil Farr one, that is 3 WG. I think this would mean 3" difference in measured static pressure, but I am not sure on that. I haven't tested the pressure in our system, so I don't know the percentage drop that would make in the amount of air moving. Maybe some of you who have that knowledge could help figure out how significant a drop that is. The Wynn site doesn't state how much it accepts as a "final pressure drop".
It is interesting that, in the literature I have, the filter is the same diameter as these others but only 27" tall, and they needed just over 2 1/2 pounds of dust to get to the 50% rating and just under 5 pounds of dust to get to the 100% rating. So the chances of us getting that much dust in our filters without cleaning them is pretty slim.

[Rod Nikkel]

I have the information on the MERV 16 Camfil Farr filters. They put their end point at an increase in resistance of 4 WG as opposed to 3 for the MERV 10 filters. It took 680 gms of dust to cause that much resistance versus 2218 gms to cause the 3 WG of resistance to the MERV 10 filter. So 1/3rd the amount of dust created 25% more resistance. Lesson - if you get the higher rated filters, you will need to clean them more often (and maybe replace them more often?) to keep up your air flow. But you would be catching almost all the fine dust right from time you installed the filters.