Comparing KN95s, KF94s, and Mask Fitters

Fix The Mask — the technical.
7 min readAug 26, 2021

With the Delta variant of COVID on the rise, it’s no secret that you should wear a mask that fits close to your face. With all the buzz about KN95s, KF94s, and other #BetterMasks, it’s hard to know what mask is best for you.

A good mask is one that actually forms a seal around your face. Most people get recommendations for masks from their friends or social media. Unfortunately, a mask that creates a good seal for your friend may not do the same for you because you each have different face shapes.

How big a difference does face shape make when picking your mask? We had four of our team members test three different masks using a quantitative fit test. Here’s what we found.


Summary of effective filtration of KN95, KF94, and FTM Mask Brace on 4 subjects

There are four key takeaways:

  1. Effective filtration varied by up to 30% between team members depending on who was wearing a given mask
  2. Some people can get good seal with a KF94 (but not everyone)
  3. Ear loop masks like KN95s may not be a good option
  4. Any of these masks is better than no mask. Don’t let anyone tell you that N95s/KN95s/KF94s “don’t work” or “do almost nothing” if they don’t seal well, but do consider using a mask fitter to improve your fit.

You can see our full data here.

What are these masks??


A KN95 is a mask that has been certified to a Chinese standard called GB2626. This standard is similar to the NIOSH N95 standard. You can see a summary of the differences here.


A KN94 is a mask that has been certified to a Korean standard called KMOEL — 2017–64. This standard is similar to the NIOSH N95 standard. You can see a summary of the differences here.

These masks commonly come with “ear savers,” a plastic piece which secures the “ear straps” behind the head to improve comfort and fit.

FTM Mask Brace

The FTM Mask Brace is a mask fitter which goes over the outside of a surgical mask to create high quality fit. It has been awarded a Workplace Performance Plus rating by the CDC when paired with an Armbrust USA Surgical Mask and has been validated to improve fit by experts and in published papers.


We tested a PSA KN95 (the original link we used to purchase is no longer available), a Dr. Puri KF94 (L), and an FTM Mask Brace + an Armbrust Surgical Mask (aka a Better Mask High Filtration System).

Each team member wore each mask which was connected to a Portacount Pro on the N95 setting to complete a Fit Test. Both men had minimal facial hair at the time of the test (photos provided in discussion section).

During the Fit Test, each team member followed the OSHA fit testing protocol. This protocol indicates that the participant must do 8 activities while wearing the mask.

  1. Normal breathing
  2. Deep Breathing
  3. Turning head side to side
  4. Moving head up and down
  5. Talking (reading a predetermined passage called the Rainbow Passage)
  6. Grimacing
  7. Bending Over
  8. Normal Breathing (again)

This test was designed by OSHA to see how well a mask fits while doing typical work activities rather than just sitting still.

For the data reported in the graphic above, we converted the overall fit factor to the overall effective filtration (1–1/Fit Factor) for ease of communication.


Why was the fit so different between masks and between team members? We think there are three reasons:

  • Masks move around when you do
  • Ear-loop masks have bad fit
  • Face shapes aren’t the same

Masks move around when you do

You’ll notice in the full data that performance of each mask varies during each exercise. This is because the mask shifts around the face while the participant is moving.

The FTM Mask Brace is made of flexible silicone, which means even as you talk or move your head, the brace flexes to adjust to the movement. This is part of why the FTM Mask Brace performs so well in fit testing.

Ear-loop masks have bad fit. Head straps, ear savers, and mask braces improve fit.

Ear-savers pull ear straps behind the head to avoid pressure on the ears.

In this test, our team members wore the KN95s without ear savers. KN95s worn with only ear loops have been shown to have fit similarly to what we found in our data (Sickbert-Bennet et. al. O’Kelly et. al.). Early in our prototyping process, we found that adding a fitter (with nose cushions) to a KN95 allowed it to pass fit testing to the N95 standard (see sheet 2 of our full data).

NIOSH has also stated that a “respirator” which has ear loops is likely to be a counterfeit. We haven’t heard of any NIOSH certified N95s with ear loops. This is because you’d have to apply significant force on the ears to achieve a good seal, which is uncomfortable. This is why all of our FTM Mask Brace designs feature a head strap.

Overall, we think the NYTimes Wirecutter article from mid-August, 2021 said it best.

“There’s nothing magical about a KN95. The edges still need to form a good seal against the face for it to be protective.”

Face shapes aren’t the same

Andrew has a slightly angular face with a slightly pronounced nose
Katherine has a soft, oval shaped face with a slightly pronounced nose.
Rod has an angular face with the most pronounced nose of our team members.
Sabrina has a soft, round face and a small nose.

The impact of these differences is most pronounced in the KF94 test. Andrew, Katherine, and Rod all had a hard time achieving good fit. By contrast, Sabrina’s small nose and wide face is well suited for the boat-shape of the KF94.

Unlike the KF94, the FTM Mask Brace was designed specifically to accommodate faces like these. Our patent pending nose cushions allow all the tested faces to achieve a high fit factor because they close the gap created between the nose and cheeks, as well as on the sides of the mask.

But wait, aren’t KF94s and KN95s required to pass certain fit requirements?

Yes, KF94s and KN95s are supposed to only allow 8% leakage. Leakage is particularly hard to measure because face shapes are so different. What’s the “average face” that is allowed to have 8% leakage? How do you ensure most/all people experience low amounts of leakage?

The NIOSH N95 uses a method called a Conformity Assessment which evaluates fit on a variety of specified face shapes (known as the Bivariate Panel) that is meant to be representative of the US adult population.

KN95s are certified according to a standard called GB2626, and KF94s certified according to a standard called KMOEL — 2017–64. We’ve researched what we can find about these standards, but can’t find the test method used to measure leakage. Further, because they’re designed for East Asian populations, it is likely that the face shapes they are tested on are not representative of faces in the US. If you have more information about this, please shoot us a note at

The FTM Mask Brace was designed based on the NIOSH anthropometric heads which represent a variety of nose and face shapes and sizes based on data from thousands of American adults. We’re in the process of compiling the fit data based on the N95 certification protocol to submit for NIOSH N95 approval with the Better Mask High Filtration System.


Sample Size

Clearly, our sample size is quite small. We aren’t making the claim that the FTM Mask Brace is THE solution for universal fit for every person. We’re still working on manufacturing larger braces to pair with larger masks.

We also won’t tell you that our faces represent the average faces of every person in the world.

But that doesn’t undermine the fact that this data represents how each of these masks performed on each of our team members’ faces.


There is of course a chance that our KN95 was a counterfeit. After all, the CDC says that 60% of KN95s are fake. NIOSH also revoked approval for dozens of models of Shanghai Dasheng “respirators,” underscoring the prevalence of counterfeit masks.

We are sending the KN95 we used to 4CAir, a facility with NIOSH testing capability, to have its filtration performance measured and will update this post with the results.

Well why didn’t you test x, y, z, mask/fitter/filter combination?

Fit testing takes time. If there’s specific data you feel is missing, let us know at and we’ll share the feedback with our team for our next blog post.


As the Delta Variant has become more prevalent, the importance of mask fit is becoming common knowledge. But when you’re picking the right mask for you, don’t just pick your best friend’s favorite mask. Make sure the mask you’re using actually creates a seal with your face.

All of our team members who experienced bad fit with the KN95s and KF94s could easily feel the leakage around the side of their mask. If you notice leakage with your mask, it may be worth trying out something new like the FTM Mask Brace.



Fix The Mask — the technical.

The latest data and information on the Essential Mask Brace and masks in general.