Hacker News Reader

Hacker News

Comment by lloydarmbrust | Hacker News Reader

Parent Full thread

lloydarmbrust

Thanks for noting, this was our old methodology and I will get that updated ASAP.

We are now removing the subjective feedback and only talking objectively:

1) Does the mask pass the claimed PFE (particulate filtration efficiency) standard that it claims to pass?

2) Does the mask pass the claimed breathability (as stated in the standard they are claiming, ie N95, KN95, surgical)?

3) Does the mask meet other parts of the standard?

KN95, N95, KF94, Surgical, FFP2 and others all have hundreds of pages of standards that they have to meet. Most people don't know what these are but we make masks so I've read all of the standards.

Most masks or respirators fail in #3 and we are working on being specific as to why those are failing but we have about 400 masks posted and the notations are being transposed as I write.

View on HN

With regards to #3, I found it a bit incongruous that you failed some KN95 masks (like Boncare) for incomplete labeling, while allowing a mask to pass that met a standard that appears to be entirely of their own devising (the Lutema M95i [0]). This seems to be a bit unfair, given that they both performed well in the PFE and breathability tests. It seems like if Boncare decided to call theirs an “MN95” mask, it would have passed too.

0: https://lutema.com/pages/frequently-asked-questions

A few things. The PFE test I'm doing is a 2min test and doesn't tell you what the mask will do under load (all day use). It's really a simple way to tell if the mask sucks but not if the mask is good (if that makes sense).

If someone doesn't label the mask correctly...are they even reading the standard?

The gb2626-2019 standard is 57 pages and in it you have to do a lot of very technical testing. One test requires several days of testing in a lab and is very expensive, but it's designed to make sure the mask performs under actual use (called the loading test).

If the manufacture isn't willing to put some simple words on a mask, do you think they did this complicated several-day test? Probably not.

IMO it's the brown M&M or canary in the coal mine test.

Thanks for the response. I understand your reasoning for failing the KN95s, but then what about the Lutema mask, which uses a proprietary standard? Lutema includes some lab test documents [0] but I can't find any technical information about what results exactly is required to qualify as an "M series" mask. They mostly seem to compare with with KN95 masks, but without really saying what makes them different [1] (note that the testing info on that page are for their surgical mask).

It seems like the "M series" designation is mostly a branding effort to make their masks stand out from the competition, and less about specific technical improvements.

I did look to see if you had a video on this mask but there doesn't seem to be one, would love to hear your take on this unusual designation.

And regarding the Boncare (et al) masks, I suspect the omission is at least partially motivated by style; it makes the mask less conspicuous for mainstream use. The labeling on most (passing) masks tends to be rather obtrusive.

0: https://lutema.com/collections/made-in-usa-masks/products/ad...

1: https://lutema.com/pages/m95-vs-kn95-masks

The Lutema does not claim to have an internationally recognized standard, which is why it "passes" because there is no way to fail it.

The pass/fail column can be considered more of a "is this manufacturer lying?" field. Perhaps it needs another value if it doesn't claim any standards.

No way to fail it but you also can't pass a non-existent test. N/A for cases like this would be more accurate.

If a product claims to be X and doesn’t meet the X standard, it’s not an X and it’s entirely fair to mark it failed.

Otherwise, everyone would just mark their mask however the hell they wanted, benefit in the market, and as long as they passed the MN95 standard, they’d get a passing mark on their “N95” masks.

Reply on news.ycombinator.com