This design is an extension of https://3dprint.nih.gov/discover/3dpx-013599

I also created another model that is entirely 3D printable: https://3dprint.nih.gov/discover/3dpx-015194

This project is still active as of December 2020. All files are available on a github repository http://github.com/basicrespirator/basic

Design Caveats

This device is intended to eventually meet criteria for a surgical N95 respirator, but it has not been fully tested. I have released the design as-is and will continue modifications. The mask fits well and forms a tight seal when properly fitted, but the design of the filter assembly may need to be modified to increase the surface area of the filter and slow air flow. If you have any questions about the design, please email me at basicrespirator@gmail.com

Construction & Materials

The version pictured is sewn from coated Cordura© 1050D - a dense, water repellant ballistic nylon that is extremely durable and used in military tactical gear. A key advantage of using high performance fabric in a respirator design is that the flexibility allows for a contoured and continuous fit (even while talking) but the shape is rigid enough to insert a 3D printed (or plastic injection molded) filter assembly. It accepts a standard-sized 3.125" N95 disc filter manufactured by MSA, Honeywell and Moldex.

A heat-bonded strip of waterproof tape creates a seal around the filter opening. The filter assembly is designed so that when locking the cap, it compresses the tape and fabric to prevent leakage. The cap's locking mechanism is similar to a DSLR camera lens.

Ideally, the 3D printed parts should be made from nylon using MJF or surgical resin using SLA.  FDM printing can create microscopic holes that allow small particles to penetrate and make the parts harder to clean. And they cannot survive the high temperatures needed for sterilization. FDM printed parts are not recommended for use in a clinical setting but are fine for draft prints to test mask fit.

Sizing & Fit

There are 5 sizes of flat patterns that were designed to fit the 5 standard head forms created by NIOSH. They all accept the same 3D printed filter assembly and comes with 2 types of straps - adjustable elastic and an adjustable harness.  A hidden aluminum nosepiece shapes the area around the nose to prevent glasses or goggles from fogging. 

Seal

A neoprene (or hypoallergenic silicone) inner gasket around the interior edge of the mask forms an airtight seal on the face. The fabric flexes with the skin so the user can talk, yawn and make facial expressions without breaking the seal. A neoprene gasket in the center opening seals the filter assembly and a waterproof tape is heat bonded to seams to seal areas where the needle punctured the fabric.

Comfort

Since the mask is made from soft materials used in athletic wear, it conforms to the skin, wicks away sweat and does not cause bruising or sores during long-term use.

Exhaust

There is no exhaust valve in the mask, the filter is large enough so that the wearer can breathe comfortably.  I've worn it for up to 4 consecutive hours and performed strenuous activity without discomfort, but outgassing has not been lab-tested for professional use. 

Manufacturing

This mask was designed to be mass produced by manufacturers in the soft goods industry. The plastic parts can be 3D printed or potentially injection molded.

Testing

This design has not been fully lab tested to ensure it meets requirements for a surgical N95 respirator. If you have any questions about the design, please email me at basicrespirator@gmail.com

Sanitization

The fabric mask can be washed in detergent and hot water heated to at least 60˚C, per WHO recommendations for cloth masks. With the correct materials, it can also survive steam sterilization in a rice cooker. The INVISTA nylon fiber used in Cordura© fabrics has a melting point of 219˚C, so autoclave sterilization is possible, I just haven't tested it yet.

If using surgical resin or MJF printed nylon, the 3D printed parts can either be steam sterilized or placed in an autoclave. For testing purposes, I use FDM printed PLA or PETG parts that have been soaked in 3% hydrogen peroxide for 30 minutes.