For engineers just tuning in to the #8020Agang scene now, a thread. 1/n
Back in 2022, @ghhughes started posting an interesting kind of mask test on his youtube channel . Broadly surveying common public health masks using condensing particle counting equipment originally used mostly for firefighting and military masks. 2/n
A test instrument with such a wide dynamic range might seem like overkill, but it ended up capturing a phenomenon entirely missed by bitrex tests, N95 mode, and PFE tests - a Pareto distribution of typical exposure reductions - even within masks with the same certifications.
3/n
N99/N100 mode is a more informative test than N95 mode test. It measures the overall reduction of exposure, implicitly requiring *both* good filtration and fit. In a Wild West era of unregulated Amazon products, it’s the only way to know for sure. 4/n
The tests didn’t worry about whose toes they step on. They were adversarial, so they didn’t focus on what the mask *could do*. There were relatively few mods, since most people use most products in the default settings, and this includes masks.
5/n
• • •
Missing some Tweet in this thread? You can try to
force a refresh
If you’re an engineer and are looking for a major time commitment of a hobby, you can probably get a used 8020A up and running in half a year using just eBay. 1/n
You could set up a pair of machines and do some statistical analysis to make sure you’re confident in the results. If someone from #8020Agang is in your area, it’s a mad scientist meetup! 2/n
The only part you can’t get on eBay are probe installers, which I make for free for people who have already gotten an antique machine working. They’re pretty simple to make with ring magnets but I’m picky about tools. 3/n
Things you can fit test (or at least particle count) that aren’t masks, paprs, or hepas:
Your own car’s cabin filter. Mine hasn’t been replaced in 5 years. Outside air was 3K particles / cm3 at the time.
Curious to see what happens when I finally change it.
You can measure the particle count in a hospital room (~5K here).
I’m not sure people understand how unregulated this space is. The mask on the bottom is a vertical bifold with bread twist tie nosewire and happens to be a legit N95. Its typical protection(hmff) is 30 times worse than the *worst series of another respirator. This is still legal.
That wasn’t even a cherry-picked example. There’s no bottom to this. I can’t think of another example of types of products where you get this much variation.
ChatGPT has a long way to go before it can pack this much wrongness into one tweet.
Unpacking the multiple layers of wrong assumptions:
Layer 1:
An implicit misunderstanding of how electret filters work - thinking that they’re like sieves rather than the overlapped sticky spiderwebs that they actually are.
Layer 2:
You need to be so out-of-touch with the literature as to not realize which particle sizes even carry any SARS-CoV-2 genomes at all.