Currently in full hunker-down mode


Source: COMMENTARY: COVID-19 transmission messages should hinge on science. March 16, 2020, Lisa Brosseau, ScD, author, on line at the University of Minnesota Center for Infectious Disease Research and Policy.  Used without permission here, under a claim of fair use.

In the illustration above, Person A has some viral disease but doesn’t know it yet.  He’s “pre-symptomatic”.  As he talks, he generates a stream of airborne particles of various sizes.  Those particles can carry that virus with them.  Some of those (“droplets”) rarely travel beyond the current 6-foot “social distancing” range.  Others (aerosols, particles under 5 microns in diameter) can remain suspended in the air for hours, and can travel much further.  In effect, Person A is showering Person B with a spray of tiny virus-carrying drops.  These are not visible to the naked eye.  Plausibly, if B inhales a large enough number, the disease will be transmitted to him.

Person C is off to the side, wondering why neither A nor B is wearing a mask, when the US Centers for Disease Control (CDC) has said that everybody needs to wear a mask in public.

EDIT:  To which I now add: Or in any space, outside the home, that you share with others, such as a place of work.  See Post #631

With the current COVID-19 (SARS-CoV-2) pandemic, this plausible mode of disease transmission is something you need to be aware of.  And something you probably should take precautions against when you are out in public, near others.

The CDC has asked us all to wear cloth masks when in public.  This is mostly to prevent you from spreading disease (Person A).  A well-made homemade mask can substantially limit the spread of those droplets.  But the CDC advice is really not aimed at preventing you from catching the disease (Person B).  Homemade cloth masks probably have some ability to filter tiny aerosol particles.  But it’s a good bet that the average mask isn’t very good at it.  And, for sure, there’s no way for you to know whether or not your mask does that to any significant degree.

If we could all buy a proper aerosol-filtering mask (N95 or Chinese KN95), and wear that when in public, that would (mostly) solve this problem.  But as long as the shortage of masks continues, authorities have told citizens NOT to buy such masks, but instead leave them for health care workers and first responders.  And, at present, the wholesale and retail channels for such masks are tightly controlled, and you’d have a hard time getting one anyway.

So, in the meantime — until that shortage eases, which I expect will happen soon — there’s a role for creating homemade masks with known ability to filter aerosol particles.  The only materials in the home with known filtering ability are those marked HEPA, MERV, or MPR.  (See Post #593 for comparison of filtering standards).

There are problems with using HEPA-rated material due to high back-pressure and possibly unhealthful materials. Practically speaking, the only way I know of to create a breathable mask of known filtering ability is to use the material out of high-end disposable home air filters (furnace filters, air purifier filters).  Those filters are designed to filter such particles without unduly restricting air flow, and their filtration ability is known (via MERV or MPR rating).  I am not the only person to have figured this out, and this method was used on an amateur production basis during the initial crisis-level mask shortage that occurred in certain parts of Washington State early on in this pandemic.

This blog is not about producing homemade masks for health care workers.  Hospital workers would only wear those, for real, as an act of total desperation.  You can see that in the CDC guidance in this area, where hospitals are instructed to resort to homemade masks only when their sole alternative would be to go bare-faced.  In essence, they are instructed to use anything but a homemade mask.  You can see it in our largest local hospital system, Inova, which clearly tells citizens that it does not want homemade masks.  You can see it in a front-line physician’s discussion of his own high-quality “duckbill” mask below.  And now that a safe protocol for re-use of N95 masks has been developed, and the FDA has removed its ban on importation of plentiful Chinese-made KN95 masks for hospital use, it’s unlikely that any hospital is going to get that desperate (see Post #598).

Mainly, this is about finding a way to produce a large number of effective masks, and get them into the hands of residents of my home town.

Secondarily, this is about publishing means and methods so that anybody who wants to replicate this can do so.  But even as I write, there are plans for furnace-filter masks popping up all over the internet.

In addition to having a known, quantified ability to filter aerosol particles, a major advantage of a furnace-filter mask is that you can get adequate aerosol filtration without heavily restricting air flow.  This allows for more comfortable use, and allows for use of lighter-weight fabrics.  And, most importantly, the lack of high back-pressure means that your breath goes through your mask, not around your mask.  It limits the extent to which mask filtration is compromised by air leaking around the edges of the mask.

I have settled on a design for what I term a “shopper’s mask”.  Something good enough to be worn for an hour or two a week, and plausibly capable of lasting for months at that use rate.  My current status is that I’m waiting for materials to arrive.

The plan is to start from boxes of cheap “single-use” masks (unsuitable for hospital use, and still available (with delay) from Amazon).  I’m going to duct-tape pleated pieces of  high-MPR Filtrete (R) or high-MERV air filter material to the front of them.

I could go on from there, but that’s pretty much it.  Such masks are still available on Amazon, but with significant wait times.  An alternative would be to take somebody’s homemade surgical-style cloth mask, and do the same.  I calculate that two layers of MRP 1900 Filtrete ought to give roughly an N75 level of protection.  It should catch 75% of 0.3 micron particles.

This mask will be made entirely of plastics.  That’s because I don’t plan to wash it, and natural materials might eventually (e.g.) grow mold.  (Washing reduces the effectiveness of true N95 masks, which suggests that it might do the same to furnace-filter masks.)  I plan to disinfect as hospitals now do for N95 masks — drop it in an open paper bag and let it sit.

I’m also working on a “worker’s mask”, a sturdier, respirator-style mask that would seal up against your face, and could be worn daily for weeks without disintegrating.  I have one workable model — not very good, see below — and it looks like those will take a lot of time, per mask, to produce.  Probably not worth the effort, given that I expect the N95 mask shortage to be over soon (see Post #598).

Sewn masks

I am not sewing masks.  You can find plans and instructions for that all over the internet.  If you do sew a mask, I suggest that you make one with a “filter pocket”, in hopes that you can add a piece of Filtrete 1900 (or higher) or MERV-13 (or higher) air filter material.

That said, most sewn masks attempt to replicate a standard pleated “surgical mask” style.  By eye, these seem to take a fair amount of skill, and result in a mask with a lot of stitches (i.e., holes) in it.  But in their favor, the standard pleated over-the-ear surgical mask is a time-tested design.

I’d like to highlight two alternative sewn mask designs that I came across on  I like using Instructables because it provides a common format for comparing masks, has feedback from users, and there’s a count of the number of people who have made each mask.  In short, Instructables is set up for a task exactly like this.

I admire both of the masks below for their simplicity, and for their ability to incorporate a piece of air filter material.  Both of these are reproduced without permission, with citation as to source, with a claim that this is fair use.

Here’s what I believe to be the best sewn mask I’ve come across so far.  FWIW.  It’s certainly the slickest-looking.  When I saw another MERV-mask maker mention this, I knew it was a good design.  Note that she has a design with a pocket for a filter.  So, if you have a high MERV filter at home, you could plausibly get the best of both worlds with this.  I find this an admirable design due to its simplicity, depth of face coverage, and attention to critical details like the nosepiece.  If you don’t have elastic at hand, think something like kitchen cotton twine would probably work, but you’d have to tie it to the right length after fitting the mask to your face.

Source:  Instructables, as cited below.  Image used without permission, for which I apologize.

  • Mask 1:  DIY Cloth Face Mask.  By
  • Cloth:  Tightly woven cotton or cotton/poly.  Cotton recommended.
  • Cloth Layers (not including filter):  2
  • Ties:  Elastic, over ears.
  • Metal noisepiece:  Yes, continuous wire, sewn in, floral wire or other thin wire.
  • Design:  Single piece of cloth with elastic feed through “drawstring”-type channels at side.
  • Methods/skill level:  Knowledgeable craft sewer.
  • Steps:  14
  • Number made on Instructables (as of about 4/1/2020):  73
  • Interesting quote:  “Surgical masks and some cloth masks will block 7 micron particles.”

Here’s a mask that a physician made, just-in-case.  If nothing else, you should read what he has to say about the circumstances under which he would actually use this mask, in his hospital.  Even a good attempt a a cloth mask with high filtration rate would only be used as a last resort.  They’ll be re-using their existing N95 masks for a long time before they will resort to home-made masks.

And that’s a physician talking about a mask where a) he knows who made it, b) he knows how its made, and c) it contains materials that actually will filter out aerosol particles.

Source:  Instructables, as cited below.  Image used without permission, for which I apologize.

  • Mask 2:  Face Mask Duckbill With Filter, by Chris Holmes
  • Cloth:  T-shirt material.
  • Cloth Layers (not including filter):  2
  • Ties:  Elastic, around head.
  • Metal noisepiece:  Yes, lengthy band of metal flashing at top edge of mask..
  • Design:  Single piece of cloth in “duckbill” pattern.
  • Methods/skill level:  Amateur sewer.
  • Steps:  5 (but some steps involve more than one action).
  • Number made on Instructables (as of about 4/7/2020):  3
  • Interesting quote:  “So maybe you don’t need a N95 (Please stop hoarding unless you are intubating a patient with COVID-19). ”

Gallery of masks I have made.  So far.

First mask for distribution:  A “mask liner” using a cheap single-use commercial mask and Filtrete.  (The one pictured here uses a piece of printed fabric because I didn’t want to waste Filtrete on a prototype).  This is intended to be worn under the user’s existing cloth mask, to up the filtration level of that existing cloth mask.

Materials:  Cheap single-use mask (not suitable for hospital use), Filtrete MPR 1900 or higher, thin and breathable material for the top layer, staples, tape, a length of stiff wire.  I used a “floating row cover” for my cover material.  That’s extremely thin spun-bonded polyester.

Tools:  Stapler, wire cutter, needle-nose pliers.


Measure your mask.  My masks were 6.75 inches side-to-side, and 6.5 inches top-to-bottom when un-pleated.

Cut the Filtrete that size.  Cut the cover cloth roughly two inches larger in each dimension (so add about 2″ to whatever your mask size is).  Center the Filtrete on the cover cloth, wrap the cover cloth around all the edges, and clothespin the corners to keep it there.  The dimensions on the Filtrete need to be as close to correct as possible.  If you err, oversized is better than undersized.  The dimension on the thin cover material are not critical so long as it’s just somewhat bigger than the Filtrete.

Turn the cheap commercial mask so that the outside (blue, typically) is facing up.  Lay the wrapped Filtrete on it with the Filtrete facing down.  Staple the lower (chin) edge with five staples, taking off clothespins as you go. Staple the corners on the upper (nose) edge, taking off clothespins as you go.  Optionally add more staples to the upper edge.

Try to staple through the obvious “edge” portion of the mask, not the fabric, if possible.  Use the stapler right-side-up for this, so that the rough piece of staple end up on the outside of the mask.

Using your fingers, create three small pleats in the Filtrete/cover, so that it now lies flat.  Weight those down with something, to keep them in position.  Position is not critical — you don’t have to match the cheap mask.  But you’ll ideally want to position those so that you don’t have staple through the mask elastic at the next step.

Push one side edge of the mask over the edge of the table, and put one staple in each pleat.  Turn it, and staple the other side edge the same way.

You’re done, assuming you are going to wear this under an existing mask that has a good nosepiece that seals tightly against your face.  If so, position this mask liner so that the nosepiece of your main cloth mask holds this tightly against your nose and cheeks.

If the cloth mask you will use this with does not have a good nosepiece, you can make one for this mask as follows.  (But, really, you ought to make a better nosepiece for your cloth mask instead.)

For the nosepiece. Cut your stiff wire about 2″ wider than the mask.  Using needle-nose pliers, put loops in the end of the wire, so that you can guide the top elastic band through those loops.  Do that, then use three small pieces of high quality tape to fix the nosepiece to the mask at both ends, and at the bridge of the nose, along the top edge of the mask.

Unfold the mask, bend the nosepiece into approximate shape, put the mask on, and fine-tune the nosepiece shape and mask position to get the best possible seal on the nose and cheeks.

First try:  Nuisance dust mask converted to MERV-8 filtration.  Requires a MERV-8 or higher furnace filter, a low-quality dust mask, piece of cord, tape, and large scissors.  Seems to work fine.  Obviously needs higher-MERV material to be useful for filtering aerosols.

Christopher Hogan, Ph.D.