After publishing this article: https://www.texairfilters.com/the-usa-should-not-adopt-iso-16890-for-air-filter-testing-heres-why/ on Linkedin, we had a very good discussion about the pros and cons of the US adopting ISO 16890 as our air filter test standard. This article is taken from that discussion. I have omitted some comments and the names of participants, have shortened the comments of other participants (although I have tried to accurately express their views) and have added some “notes.” My comments during the discussion remain virtually unchanged.
One change is the use of the terms PM and ePM. As you will see in this discussion, the use of ePM in ISO 16890 is very confusing. In ISO 16890 the term ePM is used to denote particle size in the test ie. ePM1, ePM2.5 and ePM10. On the other hand PM1, PM2.5 and PM10 are used by many researchers and agencies to show the mass or weight of the particles in a given volume of air – usually expressed as micrograms per cubic meter. It was correctly pointed out in our discussion that I was using these terms interchangeably. I have edited my comments to reflect this difference between PM and ePM. Where others have used these expressions, I have kept them unchanged.
Comment: So the question is IF we only worry about E1 and stop reporting on E3 would we save budget dollars? After all the health effects are in the E1 and the amount of particles by count in E3 is almost not noticeable. True, by weight it’s up there but it is very easy to pull the bowling ball out of the equation and focus on the E1 particles that have meaning.
The other issue we have is how air filters are installed, if they are not sealed then you are not getting the reported MERV value.
JR: Adopting ISO 16890 does not help us address either of your concerns. I agree that E1 removal rates tell us more about what happens in an indoor environment. But most of the filters we use in the US would fall in the ePM10 (maybe ePM2.5) ISO 16890 classification and we have no new data on small particle performance. We would be better off continuing to enhance the MERV tables to reflect E1 performance. The inclusion of a 20% E1 level for MERV 11 filters was a start.
No one could argue with the importance of proper filter installation. However, changing our testing procedures to ISO 16890 is not going to help this at all and is not progress.
Comment: When they came up with the 3 Ranges I said cool, now we can show folks how air filters are performing. But all the 3 ranges have done is make it harder to understand because engineers still think a MERV 14 is a 95% filter and as long as 95% is being removed why should they use MERV 15 or 16 that are also a 95%?
JR: Your point on engineers still using 95% filters is a great example of the difficulty we will have if we adopt ISO 16890. The 95% comes from ASHRAE 52.1 which has been expired for years. We haven’t even been successful in training users about ASHRAE 52.2 let alone trying to train them on ISO 16890.
Comment: you are wrong. ISO 16890 is more clear for the end user and would shift the focus to fine sub-micron particles that are the ones that get deep inside the body. Your assessment that most filters in the US are PM10 is exactly why we need ISO 16890, a PM10 filter should not be acceptable. MERV is both confusing and misleading and no alterations to the MERV tables will address that.
JR: The MERV of ASHRAE 52.2 may be confusing but it is not misleading. ISO 16890 is confusing and is clearly misleading. By connecting filter test results with desired indoor air quality PM1, PM2.5 and PM10 levels, this system is intentionally misrepresenting what happens in indoor environments.
Filtration efficiency is just one factor in determining indoor air particle counts. Indoor air counts are also determined by activity in the space, particle deposition, air exchange rates, and outdoor to indoor particle transfer. The fact of the matter is that these things often “overwhelm” the ability of the filter to remove particles. This article illustrates this point:
Comment: Case in point – MERV 11 is not a high efficiency filter, it removes far less than half of all particulate in the air. Multiple studies have been done now, including one sponsored by NAFA done by Drexel University, that shows using MERV 11 and recirculated air exponentially worsens indoor air quality. But because 11 is on the high end of the MERV table it seems like like it would be high efficiency and most manufacturers call it High Efficiency.
The Drexel University study does not show “using MERV 11 and recirculated air exponentially worsens indoor air quality.” What it does illustrate is the point made above – other factors often overwhelm a filter’s ability to remove particles.
Another case in point is the school study mentioned in my original blog post. The indoor particle counts in the 7.5um to 10um size range during the day when the building was occupied were 11.3 times outdoor particle levels on very polluted days. They were 50 times higher than the same building when it was not occupied. The fact is school children create lots of particles. A MERV 8 pleated filter is very efficient in removing 7.5um to 10um particles. Do you really think that a ePM1 filter would show any better results?
The more times I read this comment the more concerned I am about how ISO 16890 is being sold to the filter community and how it will be used to sell filters to users. By using ePM1, ePM2.5 and ePM10 as filter classifications in the test standard there is an implication that filter users would achieve these levels of particles in their buildings. In supporting documentation I have been reading this being touted as the big benefit of ISO16890. “Filter users will now be able to select the level of air quality they want or need.”
This is a false assumption and, if used to sell filters, it is deceptive.
You cannot predict indoor air quality with a single-pass filter test. There are a number of things that happen indoors that create particles like people, office machines, cooking, etc. . In a busy building it is often “particle chaos” and the particles just don’t line up like in a test duct for easy removal .
Comment: ISO 16890 passed by unanimous vote, that does not happen often – if this standard was not beneficial there would’ve been AT LEAST one dissenting vote.
JR: The fact that ISO 16890 was approved unanimously does not mean anything other than there has been a well coordinated effort to encourage this result. Actually, this unanimous vote is one of the biggest reasons I am opposed to the USA adopting the ISO 16890 standard. There are over 100 voting member countries in ISO – one country – one vote. This is a dangerous situation for the US filter industry in that we loose control over standards that effect our market. Given the relative size of the filter business in the United States vis a vis the vast majority of countries in ISO, this is not a risk a reasonable and impartial person should be willing to take.
(Note: I cannot overemphasize the importance of this point. By and large ASHRAE 52.2 is a pretty good test method. Sure it has it’s “warts.” But do we really think that ISO 16890 will not have it’s own “warts?” As filter media and filter manufacturers learn about ISO 16890 testing, the same types of issues will arise whereby tests will be used to try to obtain a competitive advantage. Are we naive enough to think otherwise? With our current system of ASHRAE 52.2, we have the ability to adjust our standard. By adopting ISO 16890 we will be at the mercy of an international body that does not care about our filter market or our filter users. In short, it’s a bad deal.)
Comment: ISO 16890 does not give any recommendations for filter usage nor for indoor air quality. It is just a filter test standard that compares filters like by like. The measurement of the fractional efficiency curves and the test rig are almost identical to ASHRAE 52.2, but ISO 16890 gives better transparency to filter users. Like in Europe, I see in this paper and in the discussion here that obviously some people are concerned of this transparency. If you sell a low efficient filter, it is much better to print MERV 8 on it (which does not tell anything to anybody) than to print ISO ePM10 50% on it. This USA and Europe have in common.
JR: Yes, a MERV 8 filter might be “low efficiency” in your world. But I think this illustrates the fact that in the USA we look at filters differently than you do and that we need to address this in air filter testing standards.
There is a clear multinational effort to connect the adoption of ISO 16890 and clean indoor air. Just do a Youtube search on “ISO 16890” and you will see clips in dozens of languages touting a connection between ISO 16890 and health. These clips close with a picture of an adorable baby breathing. The message is clear – vote for the adoption of ISO 16890 or the health of cute babies will be at risk. I don’t think so.
Comment: Because ISO 16890 will report filter efficiencies in the same terminology used by the EPA, OSHA, and the WHO I think that alone will be enough to lead to its adoption here in the USA.
The use of ePM1, ePM2.5 and ePM10 does not really correspond with the terminology used by WHO, EPA, OSHA, etc. The PM used by these organizations refers to the mass of particles in the air less than these thresholds. The use of these same thresholds for filter testing and classification is not the same thing. It is only clever marketing. These thresholds could have just as easily been named Voodoo1, Voodoo2.5 and Voodoo10.
Comment: However, let’s take a moment to assume your arguments get heard and there are enough votes to go against ISO 16890. Here is what my tea leaves say will happen next: The largest air filter manufacturers in the USA are now global entities. Most of these companies, are proponents of ISO 16890. If they choose to start labeling their filters per ISO 16890 and ignore 52.2 (or 52.3 if that happens) it will put ASHRAE in a difficult spot.
JR: At the end of the day, companies will do what is best for them in selling filters in the United States. If they decide to label filters with ISO 16890 classifications, so be it. However, I do not think they will do it – any more than they would label filters solely in the metric system.
Comment: the NAAQS monitor and have thresholds for PM2.5 and PM10. ISO16890 does have a lower bound of 0.3um but the ePM2.5 takes into account all particles between 0.3 and 3, similarly ePM10 takes into account all particles 0.3-10 so there is a direct correlation between what the EPA uses and 16890 it was specifically designed to do so. However the E1, E2, and E3 ranges in 52.2 are independent of each other, in 16890 your performance on fine particulate (which is most detrimental to human health as these particles can enter the bloodstream) always matters, just as it would when monitoring PM2.5 or PM10.
I think we just illustrated the confusion caused by using the PM air quality terminology in ISO16890. PM as used by the agencies you mention is derived by sampling the air. As I have explained above, indoor air particle counts are dependent on activity in the space, outdoor to indoor particle transfer, deposition, air exchanges of the HVAC system and filtration efficiency. By using ePM1, ePM2.5 or ePM10 in the air filter classification, the air filter user is going to be confused about what actually is going to take place in their building.
No, we do not agree that what is meant by PM1, PM2.5 and PM10 as used by the EPA, WHO, etc. is the same as what is meant by ePM1, ePM2.5 and ePM10 in ISO16890. WHO and the EPA are using these terms in reference to the weight or mass of the particles for a given volume of air – usually expressed as micrograms (um) per cubic meter (m3). ISO 16890 uses these terms as thresholds for measuring particles captured by a filter in a single pass test. The use of the same terms is confusing and misleading.
Of course, I feel that removing particles under 1um is important. (Note: That is exactly the point made in my original article and the biggest advantage of electret pleated filters. They are 5 to 7 times more efficient on PM1 particles than mechanical only pleated filters.) My point is that the filter is just part of the equation. The articles, videos and interviews supporting ISO 16890 imply (and state) the filter is the answer to give the desired level of indoor air quality. The other factors I have mentioned are just as important.
Comment: I don’t understand what you are saying when you refer to the use of air conditioning in the USA and air filters.
Let me explain a little more about what I meant by the effects of air conditioning on air filter selection. When you have a heating and ventilation system with no air conditioning, you can slow down airflow by using a more resistant filter and you probably will not have too much of an effect on the system. However, when you have HVAC (with air conditioning), airflow is critical. Without it the system will not work properly – if at all. We have millions of installed HVAC units in the United States that are designed with this fact in mind.
This leads to a difference in air filter perception. As I explain in my article, when you say “air filter” in Europe, one probably thinks of a pocket filter. When you say “air filter” in the USA, one probably thinks of a pleated filter. This difference in perception has and will continue to make it difficult for us to agree on air filter testing standards.
My point in arguing against the adoption of ISO 16890 by the USA is that it really does not help in our quest to improve indoor air quality. It is just another filter test. It is not an advancement – just a way to move the deck chairs on the Titantic. I would rather put our efforts, time and money toward new test methods and procedures that give us the tools to improve indoor air quality. As mentioned in my article, testing using recirculation would be a good step.
Comment: One more thing which I like to address in your article and which is clearly wrong is your sentence: “Another important component of ISO 16890 is that it requires all filters to be “discharged” before testing”. ISO 16890 requires the measurement of the fractional efficiency curve before and after IPA treatment (so-called “discharging”, but I never use this word), and then the average of both is calculated.
I don’t understand your objection to my saying that all filters are required to be discharged in ISO 16890. In the next sentence you say that the filters are required to be given IPA treatment. That seems like discharging to me. (Note: Now I see the objection. I used the word “before.”) The fact that in ISO 16890 you average the “discharged filter” efficiency and the new efficiency does not change this.
It is interesting to me that no one has addressed my proposal in my article of having two tests – one for high efficiency and one for medium efficiency. I guess the strategy is that it is better to force one size fits all – when it really does not.
Comment: I get more and more the impression, that you haven’t read ISO 16890. If you would, you would know, that the symbols PM1, PM2,5 and PM10 are only used once in the introduction to explain exactly what you mention in your comment: the difference between the use of PM1 by EPA, WHO, etc. and the size range used in ISO 16890. Everywhere else in the standard ePM1, ePM2,5 and ePM10 are used as symbols for efficiency values to the related size ranges, what is also clearly and openly explained in the standard.
JR: I have read ISO 16890 and I have read supporting articles and literature and watched supporting videos and clips. I don’t think you can separate the actual standard from the supporting statements about the standard. At the end of the day the argument being made for adoption about ISO 16890 is about indoor air quality.
I am not an engineer and have not served on the ASHRAE 52.2 Committee. But I am a particle counting nerd and have two particle counters operating in my office continuously and three others that I use at various times to measure both inside and outside particle counts.
Since 2002 I have been giving 5 to 10 presentations per year to groups of 5 to 500 on air filtration, air filter selection and indoor air quality. I have given these presentations to prospects, customers, ACCA, NADCA, RSES, INDA and others. My only rule is to never talk about my company or my products. The presentation is neutral.
After covering 52.2 I explain the most important take-away for indoor air quality is to focus on the the big changes. Don’t worry about whether to use a MERV 7 or a MERV 8. If you are going to improve indoor air quality at all move up to a MERV 11. Or if you have the budget and desire do the retrofit necessary for a MERV 13 or MERV 14, do that. ISO 16890 does not change that message. In fact it makes it more confusing in that you would have 40 levels of filtration instead of 16. You are just splitting the hairs in more pieces. A new air filter test standard will not change the filters being tested and result in lower indoor PM1 levels. (Note: The use of more levels implies some added degree of precision. Actually, it is an illusion created by specificity. I would challenge anyone to tell me what level of ePM1 filter is being used in an HVAC system in the USA by taking a particle count in an occupied building.) At the end of the day implementing ISO 16890 in the USA will be expensive, time consuming, confusing and is unnecessary.
Comment: I think we have reached an impasse in this discussion.
JR: You are right. We have reached an impasse. I have been reluctant until now to point out the obvious, since I feel this has been a good discussion to bring out the pros and cons of the USA adopting ISO 16890. The obvious is that certain companies have made it their mission to support adoption. Coincidentally this position reduces their expenses and improves their competitive position in the marketplace. It is hard to refute the argument that their support is self-serving.
Part of this campaign has been an effort to connect exposure to PM1 and ISO16890. I am all for the health of babies and allowing people to breathe easier. I am clearly in favor of reducing PM1 exposure through filtration. The most effective way for doing that in the USA is through the use of pleated filters with higher PM1 capture efficiency. ISO 16890 tries to negate this fact to the detriment of our filter users.