By Jim Rosenthal
As we have seen in numerous tests, Corsi-Rosenthal Boxes with MERV 13 filters have been shown to outperform HEPA Air Cleaners in particle removal – on all particle sizes. Given the difference in single pass efficiencies, this does not seem possible. After all, a HEPA is at least 99.97% efficient on all particles including those less than 1 micron (um). A MERV 13 filter is tested to be at least 50% effective on the particles less than 1 micron. The reason for this seemingly impossible result is that filter efficiency is only part of the equation. It is Filter + Fit + Flow. One must have a filter with reasonably high efficiency on small particles, the filters must “fit” in the system where they are being used and the flow through the filter has a major bearing on the effectiveness of the filter/air cleaner.
This article discusses the filter effectiveness versus filter efficiency in more detail. Plus it also gives a list of peer reviewed studies confirming the Filter + Fit + Flow equation.
The problem is that the ISO and ASHRAE filter tests – both for MERV 4-16 and for HEPA only measure single pass efficiency. Consequently, they do not reflect what happens in the space where the filters are used. But there is an area where particle counts in the space must be measured – cleanrooms. Cleanrooms are utilized in medical, biotechnology, pharmaceutical, nanotechnology, compounding, and various cleantech manufacturing facilities. They are classified by the number of particles in the air in the room.
Cleanrooms are carefully controlled environments that not only are made to ensure specified particle levels, they also control temperature and humidity. They are sealed from all outside particles. Workers in the cleanrooms wear masks, gloves and clothes designed to limit human particle generation. The cleanroom HVAC systems are designed to provide the most efficient airflow possible, and they all use HEPA filters. Here is the critical point – they all utilize extremely high air flow rates. Designers of cleanrooms know you have to move the air to clean the air. Here are the Air Changes per Hour (ACH) for the most common cleanroom classifications. The particle classification levels are per cubic foot.
| Cleanroom Classifications | Air Changes Per Hour | >0.5 um | >1 um | >5 um |
| Class 10/ISO4 | 500-600 ACH | |||
| Class 100/ ISO5 | 300-480 ACH | |||
| Class 1000/ISO6 | 180 ACH | |||
| Class 10000/ ISO7 | 60 ACH | 10,000 | 2,377 | 82 |
| Class 100,000/ISO 8 | 20 ACH | 100,000 | 23,771 | 820 |
These Air Changes Per Hour (ACH) are substantially higher than we would see in typical buildings. Most buildings operate with an ACH of 2-4. The goal for cleaner, safer indoor air is 6 ACH. Some IAQ experts have been pushing for 12 ACH. Yet 20 ACH is the minimum for the lowest classification of cleanroom. It goes up from there – Class 10,000 / ISO 7 requires a minimum of 60 ACH.
So that raises the question: What would happen if we dramatically increased the Air Changes Per Hour with Corsi-Rosenthal Boxes? Could we achieve anything close to the particle levels specified for cleanrooms? The most difficult particle reductions would be for the 0.5 um levels. Remember the efficiency for a MERV 13 filter is only 47% at 0.5 um. Of course, I have particle counters and plenty of Corsi-Rosenthal boxes – so we just had to test it and find out. Good results would also further enhance the theory that filter effectiveness is based on both Filter and Flow.
The first test was conducted in my office. It is 12’X14’X8′ for a total of 1,344 cubic feet. It has a drop down ceiling, carpet on the floor and is a bit messy. I stayed in my office during the test and did not wear a mask. I used two Corsi-Rosenthal boxes with 4 – 20X20X2 MERV 13 filters. One has a Utilitech fan, the other has a Lasko fan. I ran both at the highest speed. At that speed they produce about 500 cfm each for a total of 1,000 cfm. This means that the room has about 40 Air Changes Per Hour. We used a LIghthouse Handheld 2016-IAQ particle meter. Here are the results in particles per cubic foot:
| Time – 3:30 | 3:45 | 4:00 | 4:15 | 4:30 | 4:45 | |
| 0.3 um | 312,200 | 182,300 | 132,800 | 117,900 | 116,000 | 112,900 |
| 0.5 um | 36,300 | 16,700 | 8,700 | 6,300 | 6,700 | 4,900 |
| 1 um | 4,300 | 2,200 | 800 | 600 | 200 | 200 |
| 2.5 um | 200 | 200 | 200 | 200 | 0 | 0 |
| 5 um | 0 | 100 | 0 | 0 | 0 | 0 |
| 10 um | 0 | 0 | 0 | 0 | 0 | 0 |
Within 30 minutes the particle counts in the office reached Class 10,000 / ISO 7 cleanroom levels! Wow! And this was achieved in an unsealed room with a person inside not following any of the appropriate cleanroom protocols. This certainly dispels the theory that a MERV 13 filter will never capture a sufficient level of sub-micron particles to match the performance of HEPA filters. The <1 um particles are not bouncing around continually passing multiple times through the less efficient filter. They are gone!
Maybe we just got lucky. To be scientifically valid a test or experiment needs to be repeatable. So let’s go to a different room and see what we can achieve. This room is 15.5’X16.5’X8′. It has a wood floor and painted wall and ceiling. Unfortunately, one of the doors does not close all of the way and the room is in the same building with a factory. We used the same Corsi-Rosenthal Boxes for this room. They were run at the highest speed and produced about 30 ACH. Starting particle levels were normal for this space. Here are the results:
| Minute | >0.5 um | >1 um | >2.5 um | >5 um |
| 1 | 1,376,300 | 92,000 | 34,300 | 100 |
| 2 | 1,055,500 | 47,700 | 29,900 | 0 |
| 3 | 622,800 | 25,400 | 18,100 | 0 |
| 4 | 377,000 | 13,500 | 9,300 | 100 |
| 5 | 225,800 | 7,200 | 5,500 | 0 |
| 6 | 135,400 | 4,700 | 2,300 | 0 |
| 7 | 83,900 | 2,500 | 1,600 | 0 |
| 8 | 49,900 | 1,500 | 1,000 | 0 |
| 9 | 33,700 | 900 | 400 | 0 |
| 10 | 22,700 | 500 | 300 | 0 |
| 11 | 15,200 | 400 | 200 | 0 |
| 12 | 13,000 | 200 | 200 | 0 |
| 13 | 9,600 | 300 | 100 | 0 |
| 14 | 9,900 | 200 | 0 | 0 |
| 15 | 7,800 | 100 | 100 | 0 |
| 16 | 7,900 | 100 | 0 | 0 |
| 17 | 6,900 | 300 | 0 | 0 |
| 18 | 6,600 | 400 | 0 | 0 |
| 19 | 7,200 | 200 | 0 | 0 |
| 20 | 8,300 | 300 | 0 | 0 |
After 13 minutes, particles in the room reached cleanroom Class 10,000/ ISO 7 levels! Here is a graph showing the precipitous drop in 0.5 um particles. It is impressive.
I just conducted another test in my office as I was writing this piece. Particle levels reached Cleanroom Class 10,000 / ISO 7 levels in 15 minutes!
So, what is the point? Am I suggesting that we need to start using Corsi-Rosenthal Boxes with MERV 13 filters to create cleanrooms? Of course not. But these test results do open up many possibilities where you want to quickly “clean” particles from a space. This would be an excellent way to purge a waiting room or examining room in a doctor’s office. Conference rooms could be “cleaned” after each use. Nursing homes have a multitude of possibilities from individual living spaces to common areas like card rooms, exercise rooms and other places where people congregate. How about nurses offices in schools? The potential uses are many. It would just take 15-30 minutes and a couple of Corsi-Rosenthal Boxes to create a much safer, healthier environment.
But the biggest reason these tests have value is that they confirm the need to think of filtration in a different way. I don’t know of anyone who would have said that MERV 13 filters could be used to produce particle counts equivalent to cleanroom standards. (I certainly would not have.) I encourage others to conduct the same tests. I am confident that they will have the same results. We need to test in real life situations and quit assuming that single pass filter tests give us the information we need to make the best filter choices.