Introduction: Snorkel PPE Mask Hack for Caregivers | Coronavirus | Covid19

Quick tl;dr:

  • you can hack a mask to make face protection that, from what I can tell, is easier to wear and to wash than existing designs (an opinion, not thoroughly tested)
  • it takes a bit of time to do the first one but this design can be improved, jigs made, etc.
  • it will be faster to hack than most 3d printing, but will require more labor
  • it will be easy to adapt to other filter materials


  • it looks crazy
  • people can't hear you that well
  • untested, untested, untested

On March 20, 2020, I noticed an article about people trying to hack the full-face snorkeling masks that became popular a few years ago. Several hacks have since been publicized which use the masks. For example:

There have also been fake 3D printed masks that do not work, destroy working filters, and are dangerous. For example:

This is a design for a mostly-untested alternative that uses existing, 15€ hardware and a disposable mask. The goal is to take two things that work, and combine them into a comfortable system that is more usable. If you have thoughts, warnings, or ideas for improvement, please leave comments.

If you haven't tried one, these snorkel masks are somewhat miraculous in that they form a very comfortable tight seal around the perimeter of your face, but remain fog-free.

Seeing the number of doctors and nurses in Italy and Spain who have tested positive is depressing, because these heroes are sacrificing / being sacrificed to help others. I live on a remote island, and while my laboratory has lots of fabrication equipment, management of the institution have closed the facilities and not responded to requests for access. I was forced to look for a design that was easily and quickly made in my living room with available materials, like hot glue, a dremel tool, a drill, and some tape.

I'm a hacker not a medical engineer, but for my work I've spent a great deal of time wearing PPEs, and they can suck. I think a lot of that suckiness leads to "misuse" that causes problems. I know a bit about how to fit a disposable mask, but by the end of many work sessions in a shop I have had clear marks on my face, or the inside of mask, where dirty air was coming through between my mask and my nose. I hadn't noticed a gap, but had been breathing through it long enough to leave a trail. With dust or solvent it's a problem; with the novel coronavirus it's potentially fatal.

I thought that a mask with a full face seal, which can be reused by washing the entirety in soapy water, and that isolates the filter section so it doesn't have to be touched often, would perhaps have advantages. Most importantly, it is:

  1. Full face
    1. fog-free
    2. soft on the face
    3. very well sealed
  2. Easily removed and washed
    1. without disposing of the filter
    2. without touching the filter
  3. Might also accommodate home-made filters if P3 or N95 filters run out.
    1. furnace, vacuum, and other filters might work.

Rather than Netflix and chill during my lockdown in Portugal, I thought I might try to hack a way to add filtration the snorkel masks. Only problem: all the sports stores were closed. Contacting the corporate headquarters of a great chain, I managed to contact a manager, who was an absolute mensch. The next day we met in a parking lot outside the shuttered store, and he gave me four masks to experiment with. It felt like a drug deal, but was in fact kind of the opposite.

This work is presented as a reference!
I am a [talented] amateur!
Don't try this at home!


  • 1 full-face skorkel mask
  • 2 tupperware containers
  • hot glue
  • closed-cell packing plastic
  • latex tubing (or similar)
  • a disposable filter mask


  • hot glue gun (or silicone, or possibly epoxy)
  • dremel tool (or boxcutter)
  • printer to make template (optional)
  • sandpaper (optional)

Step 1: The Unmodified Snorkel Mask


Don't mistake this for the work of an Amazom review farm, but do allow me to marvel as this mask is, to my old eyes, a masterpiece in contemporary CAD/CAM and injection molding techniques. It mixes very hard plastic and silicone into an organic unit that: sucks fresh air into the eye area; directs it to the nose/mouth area; then expels it. It is really, really comfortable, and has no less than six valves and two secret chambers.

Various mask designs offer varying degrees of protection to the wearer, the public, or both. Thin surgical masks offer some protection to public, and perhaps a little to the wearer. An n95/ffp2 mask with a valve offers a lot of protection to the wearer but less to the public, as the valve's output isn't filtered (though it sprays less than a cough). An n95 ffp2 mask without a filter technically offers both, but in my experience these masks are very quick to get water logged and unusable.

This mask has valved output, so it is _NOT_ filtering the output. The output is quite dispersed though -- it's easy to cough inside the mask, but this gets transformed into a gentle puff out of the (3) output valves, with few direct droplets.

I conducted an amateurish test on an unsuspecting mirror (pictured), lit with an offset flashlight. I coughed with the mask, then cleaned the mirror, then coughed without the mask. Both with and without mask were three coughs from a hand's width away, and (intentionally) spitty and dank. The difference is pretty marked. I am completely mortified by the unmasked cough. The majority of the droplets seemed to be heading at a 45° angle – the angle that travels the farthest – and they look as if they were big and fast enough that they could have felled a small grouse. I think it's safe to say that the mask, even without filtration on expelled air, will somewhat reduce the chance of infecting others.

The structure of the mask may be described in three main sections: the eye section, the nose section, and the snorkel. The snorkel has three pipe sections, each valved to allow only one-way passage of air. The middle pipe acts as the input for fresh air, passing from the top of the snorkel to the top of the eye chamber. The two outer pipes lead from the nose/mouth area to the top of the snorkel, and output exhaled air. The eye and nose/mouth areas are fully sealed from the outside (even with my two-day scruff) and moderately sealed between one another. When the wearer inhales, it creates a vacuum in the nose/mouth area, which pulls from the eye area through two one-way valves. The eye area receives fresh air from the snorkel. This air is, in turn, drawn through the valves into the nose/mouth area. Exhaled air cannot travel back to the eyes, but rather is conducted both through a valve through the faceplate, below the mouth, and also through two channels to the outer pipes/valves in the snorkel. The result is that the mouth/nose area is humid, foggy, and filled with droplets. The eye area, however, is dry and fresh.

The comfort of the mask, which I am hoping to exploit, is a combination of 1) the thick and compliant pads that rest at the outer edge of the face, distributing the load through pillowy silicone, and 2) the airflow. Six valves are used to provide this airflow, and while the lower valves are delicate silicone butterfly wings, the upper valves are even more so: very, very thin acetate. I don't know how long they will last. I think the silicone (1) will last the duration of this pandemic, but the snorkel valves (2) may not. The snorkel can be replaced, however.

Step 2: Mounting Snorkel to Tupperware Lid.


Our goal is to mount the snorkel breaching through a tupperware lid from the outside, with its top protruding on the inside, but not too far. This involves some compound curves, and will vary from container to container. My container is a flat 800 ml with a rubber gasket and four locking latches.

For this step I'm using hot glue -- in final assembly a silicone may work, and look better. Epoxy might also work.

Cut a hole approximately the size of the tip of the snorkel (see template.pdf file), at an angle where the full container won't hit the top of your head, but the tip of the snorkel doesn't stick out into the container so much that the filter edges can't sit fully flat on the inside of the lid. With my head/tupperware, I exposed three of the horizontal vent cuts in the top of the snorkel; this is enough for flow but allows the protrusion to be less than the interior of the filter. My angle was parallel to the ground if the mask is lying face down, but again, depending on the size of your head and your container, that may vary.

You can overcut somewhat -- hot glue is your friend here. Since the glue won't make a chemical bond to the tupperware (it bonds better to the snorkel) you can overglue above, at, and below the join. Rough sanding the tupperware where the bond will be might help. This will trap the tupperware mechanically. I know this joint looks like... crap hot glue.

Tack the join in 3-4 places first, then assemble the tupperware and carefully put on the mask to make sure you have head clearance. If so, you're ready to finalize the glue seal between the snorkel and the tupperware lid. This seal MUST be air tight.

All vent cuts outside the tupperware must also be sealed. Hot glue works.

TEST: At this point you should be able to put on the mask, and be UNABLE to inhale. If you hear wheezing or even a bit of airflow, stick the end of the snorkel in your use a length of hose as a stethoscope to . Don't forget to take the mask off before you asphyxiate.

Step 3: Making Lip for Compressing O-ring/filter


Here's where the second (sacrificial) tupperware comes in, though this could easily be made from any flexible plastic. I'd rather use a u-profile or z-profile flexible plastic strip, often available in hardware stores, but I don't have any, so I'm using the lip from a second container.

Your goal is to have an internal lip, mounted on the main tupperware container, that compresses an o-ring made of latex tubing against the perimeter of the outside of the filter mask, which in turn compresses against a gasket and the tupperware lid. This creates the seal so that all air travels into the mask through the filter.

[NB: the seal of the tupperware container is not important; neither is the the oring to the filter mask. All these are on the outside of the filter mask. What is critical is 1) the filter-gasket-tupperware lid, and also the 2) snorkel-tupperware lid. These are on the inside of the filter.]

Cut the second tupperware container below the lip, as pictured.

Flip the lip upside down, then put it in the first tupperware container so that its top is about 2mm below the edge. Trim the length of the lip so that its edges join precisely (though it doesn't need to be completely precise, as it's pushing down on an o-ring and some slop won't affect the seal). I then used a dowel to make the height difference a consistent 2mm all the way round.

Hot glue the bottom of the lip to the tupperware container. Do a few "tag" spots, make sure the spacing is still correct, then go nuts. This needs to be pretty strong.

Step 4: Making the Inside and Outside Seals -- a Gasket and an O-ring


I made a gasket (that seals between the tupperware lid and the mask) of think closed-cell packing material. The o-ring (that compresses the filter to the lid) is latex hose I had lying around; it's 3/8" (≈9.5mm) diameter. I simply lay it into the lid, and cut the overlap so that it's a bit larger than comfortable, so that the ends will be forced against one another when it's under load. This makes the gasket a bit "squirrely" to put in, but I'm sure the pressure is strong. All these materials are washable.

Lay the gasket into the lid. Lay the filter over the gasket. Note that these are the two important layers. If the gasket seals to the lid, and the filter seals to the gasket, we're golden. Everything else is just for pressure to ensure this seal. Lay the hose around the filter, making sure it is completely covering all edges.

When you snap the tupperware into place, it should flatten the o-ring somewhat and make a nice seal all around. Luckily, you can see that the gasket is tight against the clear tupperware, and note if there's anywhere that might be leaking.

This step will need to be repeated every time the filter is replaced, but hopefully this can be every few days? Don't ask me, I'm not an expert. But all parts can be washed in soapy water, and I believe the gasket and tubing can be swabbed with Ethyl (though NOT isopropyl) alcohol.

Step 5: Make Vent Holes in Main Body of Tupperware Container.


This type of plastic can easily crack when drilled. I used tape on both sides, a Brad point bit, and some patience. I did fail on one hole nonetheless. Go slow.

Step 6: Try It Out.


You are now ready to try the entire assembly. Breathing should be easy; no fog should appear in the eye section or the middle of the snorkel.

Step 7: Testing


Hah I fooled you THIS IS STILL UNTESTED but here's what I did:

  1. I put a HEPA filter in a vacuum cleaner hose
  2. I poured a lot of coal dust on to a table
  3. I put the end of the vacuum cleaner onto the snorkel attachment with a P2 filter (not happy to ruin this filter)
  4. Suckage through the face mask
  5. Compared hose area to non-hose area of filter
  6. Opened snorkel to see if there's any sign of entry into the filter

Did I remember to clear the inside of the filter I'd been using for days before the test? No. Do I know that this brief test would reproduce the "contrails" of contamination I get when using a filter that isn't perfectly snug? No. Does the vacuum pull significantly more vacuum than breathing? Yes. Did the vacuum cleaner suck one of the delicate valves out of the snorkel? Yes.

What I can say is the the Hepa filter showed no obvious contamination, nor did the area of the filter mask that was past the point of compression by the o-ring/gasket. I saw none of the feathered patterns of flow on the inside of the mask that I see from an ill-fitting mask. Of course, visible coal dust is not the size of a virus. My apartment does not have adequate testing to say anything conclusive.

Unscientific conclusion: from spending a few days with this, I'd personally use this over a disposable filter mask directly on my face and safety goggles, partly because safety goggle are terrible to wear for extended periods. I believe the filter is giving a good seal, and the face is _very_ well sealed. I think my face and lungs are more protected than with a half-face mask. I think the apparatus looks nutty and might scare people.

Please suggest improvements, or tell me why I should remove this altogether from the public record.

Step 8: Alternate Filter Materials


coming soon....