Testing 9 types of air filters with a laser particle counter

The filter is the most important component of a respirator. The quality of air that gets into your lung is determined by the quality of the filter, provided there is a good seal between the mask and your face.

The design of Totobobo mask makes it quite easy to replace the filter with any flat, air permeable material. It is tempting to use a “filter look-a-like” such as cotton wool or a piece of felt to replace the used filters. Does it work? I am interested to find out how effective these different alternatives are.

A Dylos laser particle counter is used to collect objective data. The particle counter is affordable and performs consistently. I’m using the Pro version which is able to detect two fine particle sizes: 0.5 and 2.5 microns. The particle counter has an air-inlet and a exhaust vent. Air drawn into the counter goes through a passage and there is a laser diode and an optical counter along the air passage. If the air inlet is covered by a piece of filter, air drawn into the unit will be filtered and therefore produce a lower particle count reading. The performance of the filter can be calculated by dividing the ambient particle count value with the “filtered” value. For example, if the particle count drops to half, the protection factor of that particular filter is 2.

Laser Particle Counter

Dylos Laser Particle Counter

The method is simple and easy to understand. First of all the ambient particle counts is measured against two different particle sizes – 0.5 microns and 2.5 microns. Different types of “filters” are used to cover the air inlet (edge seal with tapes) one after another. Reading of each filter is obtained after 2 minutes when the value stabilized. Finally, the ambient particle count is once again measured without any filters. The first reading without the filter and the last reading are within 5% tolerance.

Laser Particle Counter reading without filter

Laser Particle Counter reading without filter

Protection factor (P) of each filter is calculated by the formula:

P = Qi / Qn

Where Qi is the value of ambient particle count without filters and Qn is the particle count of the filter concerned. Higher P value indicates better protection factor.

The Totobobo filter (low cost anti-pollution filter) produces a high protection factor of 5564/587 = 9.5 (for 0.5 microns) and 184/10 = 18.4 (for 2.5 microns). This figure indicates a high filtration efficiency of over 90%, effectively raise air quality from “VERY POOR” to “FAIR”.

Totobobo mask filter test

test reading of Totobobo filter

Surgical mask is commonly used, the problem is getting a good seal. In this test, we seal the surgical mask filter media over the inlet to measure the filter effectiveness. As expected the protection value is not very high: P=2.9 for 0.5 microns, P=4.8 for 2.5 microns.

surgical mask filter test

test reading of surgical mask filter

As a reference, the filter of a Respro City mask is also tested. The result is disappointingly poor: P=2.2 for 0.5 microns, P=3.1 for 2.5 microns. I am shocked since this product was reported some years ago to be among the “top performers”. I checked several times to make sure the readings are correct. It may be possible that my sample is dated (over a year old). I will update this record if new evident is found to prove this results wrong.

Respro City mask filter test

test reading of City mask filter

Magiclean is cheap and commonly available and it’s working principle is based on static charge in the fibre, a characteristic of Totobobo filter. I expect it shell produce reasonable good performance. However, the result of 4 layers of Magiclean wiper is the worst out of the lot: P= 1.6 for 0.5 microns, P=2.2 for 2.5 microns.

Magiclean dry wiper sheet test

test reading of Magiclean dry wiper sheet

Facial pad is probably the most commonly used “replacement filter” since some of them available in similar dimension as Totobobo filter. The result of the test is: P=2.4 for 0.5 micron, P=3.5 for 2.5 micron. This is slightly better than the result from the Respro filter.

Facial pads test

test reading of Facial pads

Finally a benchmark test was performed against the well know 3M 8210 N95 filter. However, as the air resistance is too high, we need to enlarge the total surface area and we manage to seal the enlarged area around the inlet. The performance of this filter is remarkable: P=12.8 for 0.5 micron, P=33 for 2.5 micron, highest of the lot. The only problem with this material is the air flow resistant. It is more difficult to breathe through.

Protection factor (higher better)

Totobobo anti-pollution filter
Totobobo 94% filter
Totobobo 96% filter
Surgical mask
Respro City filter
Respro Techno Gold filter
Magiclean wiper
Facial pad
3M 8210 N95 (enlarged area)
0.5 micro
2.5 micro
test results of 9 types of filters

test results of 8 types of filters: higher figure = better performance

Note: these tests are intended to compare relative protection factors between different filters. These test results are not calibrated to various manufacturer’s standard or NIOSH.


3M 8210 N95 and Totobobo are the only two filters provide real protection factor above 9.5. These results is consistent with the finding of a study done in the Chinese University of Hong Kong (Prince of Wales Hospital). However, the air resistance of the 3M 8210 filter is too high and is not suitable to be used with Totobobo mask.
All other alternative filters are considerably less protective.

I managed to get my hands on a new Respro Techno Gold filter and tested it with the laser particle counter. The result is much more promising than the Sport filter. In addition, I also tested the Totobobo 94% and 96% filter to complete the picture. The 94% is the 2nd best performer for 0.5 micron and slightly less protective than the 3M 8210 filter for 2.5 microns. The Totobobo 96% is the best filter so far. Added a chart for easy comparison.

read more :
Cyclist, protect your lung when riding in the city
How to change Totobobo filter?