Skip to content

Material Filter Testing

nasocializes edited this page May 8, 2020 · 63 revisions

We got a particle counter!

We got a Fluke 985 Particle Counter, which is a re-branded version of the MetOne HHPC+ Handheld Air particle counter used by hospitals and clean rooms to evaluate air quality. It is a 6-channel filter that simultaneously measures particle concentration of 0.3um, 0.5um, 1um, 2um, 5um, and 10um particles. It can be used to estimate the filter performance of different materials used as PPE, as well as verify if the labeling on PPE is accurate (or not).

I wanted to test the quality of PPE that was being sold by suppliers on Alibaba to verify their performance. One sample performed far below the claimed KN95 rating yet another did Ok. We are told KN95 exports should now be more reliable (post several April restrictions).

Our latest results are viewable on this interactive auto-updated website. It contains a range of shop towels, halyards, a merv 13 air furnace filter, vacuum bag, ...

Jason Tongbai started a similar effort a couple weeks prior with more KN95 mask samples. Here are his results

DISCLAIMER: We are not testing professionals. The method we use does not match any NIOSH, CDC, FDA, or certified procedure. This is not medical advice. These are estimates of performance using the equipment we have access to. We are sharing the results of tests we did with the samples we had. The true protective effectiveness of materials against infection is not something we can measure or claim. What may be most helpful is how each material performs relative to the other. Additionally, a poor mask fit to a person's face/nose allowing air to pass around the material will significantly degrade effectiveness of even a good filter.

Context

We use 2 N95 masks we have as our baseline.

OSHA states:

N95 respirators provide the lowest level of acceptable protection against COVID-19. Therefore, OSHA now recommends that employers supply healthcare personnel who provide direct care to patients with known or suspected coronavirus with other respirators that provide higher protection than N95s. This includes N99 or N100 filtering facepieces, reusable elastomeric respirators with appropriate filters or cartridges, or powered air purifying respirators (PAPR).

Material Filtration Test Setup:

The filter material testing setup. The filter material testing setup. Zero count filter

Zero count filter

Methodology & Caveats

  • We are not testing professionals. None of the methods or tools used here are certified by NIOSH, CDC, FDA, or any other authoritative agency. This is simply sharing the data we have collected with the tools, methods, and samples that we have at hand.
  • Each sample was gathered using 2 cycles of 2L measurements. The meter reports Count/L for each of the 6 sizes.
  • The meter was last calibrated on 2016-06-24. And the zero count filter was checked at the beginning of each session. We don't have the ability to re-calibrate the meter ourselves.
  • An ambient air measurement with no filter was taken before and after each material sample. The average of the before and after samples are used as the baseline to compare the filter measurement. The ambient air quality fluctuated during the process, and this was used to help normalize the readings.
  • The inlet of the meter was covered with the material sample, and secured with several turns of a rubber band holding each sample tightly around the lip of the inlet adapter.

Breathability Test Setup:

Coming soon.

Results so far: Check Here

Results are here on this site that links to the Google sheet, which updates dynamically as new materials or tests are collected.

Notes

  • According to this source, it seems we care the most to filter out the 1-10 micro meter sized particles which the top 3 mask filters do well.
  • Some materials tested consistently and some didn't showing that our setup can still create consistency therefore it's likely it's the material that is inconsistent. (particles in the room do vary.)
  • A material's performance, especially non certified material, may differ across different sheets or different parts of the material.
  • 2 mask samples were tested at the seams and the seams seem to make the material's filter capability stronger or the same! (over 10 measurements)
  • We have to be careful that there is enough particles at the 10um. We've added a humidifier in the breathability test rig setup.

Materials in pictures (Outdated. Check the website for the latest results)

Percentage Filtered (Outdated. Check the website for the latest results)