3D Printed Respirator
With access to a 3D printer (or using an online printing service), a T-shirt, and a strip of foam, you can make a respirator that can help prevent the spread of COVID-19.
This mask is not meant to replace an N-95 for medical practitioners, but there is evidence from the World Health Organization and the National Institutes of Health suggesting a mask using a filter such as this one can make a difference. Download the files below and follow the instructions to make the mask.
The mask is comprised of 6major parts. Pictured here is the filter clamp, the filter, the lock and the mask. Not pictured is a strip of foam that is adhered around the opening of the mask to make contact with the face, and the head straps that attach to the mask. The filter clamp and the mask are the 3D printed parts - download them here:
And choose between:
These files can be used with a 3D printer at your school, business, home, public library, etc. Otherwise you can upload the files to a service such as Shapeways or Protolabs and they will mail you the parts. Some of these services may provide expedited service at a discount for COVID-related projects and I encourage you to ask them about it. Note: I generally find Shapeways more affordable, while Protolabs is often faster.
The filter is based on the mask described in this National Institutes of Health article, which describes some positive test results. Below are directions for our "quick filter". If you would prefer the filter that more closely follows the NIH direcitons, click here. Our quick filter design uses a similar filter construction and material, but with a mask that seals better to the face. This should force more of the air through the filter rather than allowing it to go around, or through the thinner parts of the mask described by the NIH. The filter is constructed from a 40cm x 20cm rectangle of cotton T-shirt material.
The T-shirt must first be boiled for 10 minutes and then air-dried. This shrinks the material and makes the holes in the material smaller.
I used these common Hanes T-shirts.
Mark a 40cm x 20cm rectangle on the shirt at a 30 to 45 degree angle as pictured. I found it helpful to cut out a template from a file folder first to help me place the rectangle onto the shirt. It is important to mark the rectangle at an angle because this will cause the gain of the knit to alternate direction in each layer of the filter.
Cut out the rectangle - if you cut through both layers of the T-shirt you will have two filters. Then fold it as shown here:
Seal to Face
It's very important that the mask seal well to your face. I cannot guarantee that this mask will fit any specific person well, but this step should help make the mask fit many kinds of faces. Apply this adhesive-backed foam around the opening of the mask where it contacts your face:
It can be purchased at McMaster Carr here. Apply it with the ridge pointing outward. For some people this foam may be too thick. I have also used a thinner window weather strip. This allows for larger faces or for your chin to fit inside the mask.
To ensure a good seal, apply silicone adhesive to the joint where the two ends of the foam meet. I have found that it is also often necessary to apply a small bead of silicone all the way around the edge where the foam and the plastic meet to make sure no air can squeeze through there.
Make sure that when you install the filter clamp on the front of the mask, the side with the ridge on it is facing the inside of the mask toward the filter. If you do not have a strong enough rubber band handy, a hair elastic may work as well for holding the filter clamp on. If/when I have a more "professional" solution, I will update the files and instructions.
For the straps, I chose to use some from two old pairs of swimming goggles. You can use just about anything you want, but it should be able to hold the mask quite tightly to your face to prevent any air from leaking out of the sides.
Please CONTACT me with suggestions or questions.
On my first prototype, the tabs that snap over the filter clamp broke off very easily, but a stiff, wide rubber band over that end of the filter does a good job to hold everything in place. I will work on a better solution and re-publish when I've got it.
I did a "smoke test" with the mask which is a less scientific version of a real fit test of a respirator. Most wood smoke particles are in the range of 0.4 to 0.7 microns. For the test, I let a paper bag on fire and held it under my face while wearing the mask. With the mask on I was able to detect a very slight smell of smoke. I removed the mask and the smoke smell instantly became overwhelming. This indicates that the mask is successfully filtering out the vast majority of smoke particles.
Thank you very much to Matt Malone of Barber Nichols for printing out the first prototype for me.
I will be working on a second version that uses MERV 16 filter material, certified to filter out 0.3 micron particles, so check back for updates. Before publishing I need to find a reliable source for the material and ensure there is nothing in it that will cause harm to anyone who wears the mask.