Reusable Mask PPE Alternative (MalaMask)

Reusable Mask with one time use filters, tested at N95 specs


Creative Commons Attribution Noncommercial Share Alike

Commercial use is not allowed, you must attribute the creator, you may remix this work and the remixed work should be made available under this license.

Learn more or download attribution tags


Please note this mask project is formally defined as a "Reusable Mask with one time use filters, tested at N95 filter specs". Be aware, to be full NIOSH approved a full mask + filter must be tested together to get the rating. We are focused on getting masks to the people who need them ASAP and are putting resources toward the aspects that are needed short term for someone to make an informed safety decision. Our main focus is filters that could be put in any mask that seals to the face and only allows air flow in and out through the filter.

The project does not just involve 3D printing masks but also using commonly found materials to create a respirator filter that is tested at the N95 requirements (0.3 micron tolerance) as a temporary solution until supply can meet with demand.

A lot of research has gone into this project. See the resources below.

Project Home Page:
Filter Research:
Mask Design:
Project folder:

Silicone Mold Stand by BreakSteel:

Silicone Casting How-to:

Materials and methods




See all issues or open a new issue


34da8dd3cf908c59dfc9e14800520b60?default=blank&size=40Cerulean Ronin added this to the Crafts collection ago
Mini x6nezqxo 100x100MalaMaker commented ago

Posted a V2 of the silicone molds to add a little more comfort around the nose bridge area.

Mini hd 1280x1024 abstract 3d medaltations 13580JosΓ© RamΓ³n Tezanos added this to the Covid collection ago
Mini x6nezqxo 100x100MalaMaker commented ago

Hey everyone I uploaded a new version of the mask shrouds that has size markings on the nose bridge. No major changes beyond that.

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

i bent it out of shape when removing the warm mold-half off the bed.

Mini x6nezqxo 100x100MalaMaker commented ago

@Marcus Wolschon This is the feedback I need and appreciate it! I have some 00-35 coming and I will try it. The tackiness is the reason I bumped up to the Dragon Skin 20... I will say the Dragon Skin 10 is like pillows against the face compared to the 20 but everything clings to it.

I have a heated enclosure so warping isn't an issue for me... I have a post on youmagine with my printer setup but I know not everyone has an enclosure. I will look into adjusting my mold with the feedback you gave. Do you mind if I ping you through email with the changes? In the mean time I see this project pushing to a simpler linear seal because it is a quick print, easy pour, and simple install.

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

I've been casting parts from silicone for....quite a while.

As a 3D printed part, my try of your 2 part mold had some tiny thermal warping

that the bolts could reduce but they of cause also bend the material, forcing other, tiny gaps to open.

It did require to be sealed around the outside to not slowly start leaking after 10 minutes.

The reservoir seems to have helped with the bot holes in the middle, that I could not seal.

Given that you have bolts to keep it together, I found that you forgot about bolts to force it apart again. ;)

Interesting video. The idea with the insertable slit is interesting if you still want that round shape.

I have a beard and found that a round shape to contact the skin doesn't work very well. A flat shape works much better for a air tight seal.

I also used a much firmer silicone then 35 Shore(00) because that's less sticky and doesn't attract dust as much as

Mini x6nezqxo 100x100MalaMaker commented ago

The 2 part mold is really intended for a silicone caster that I am working with and really he may not even use it... I'm not a silicone expert at all. I can tell you how to do aluminum casting but silicone I am a 2 week expert. The caster has been giving me information that I have been bench testing and then posting to the project when I find something.

I have been working with a group in Northern Virginia and they are using some of these linear silicone seal molds that people have on here and thingiverse, using sil-poxy to attach it. It is actually working great and may yield a future recommendation for DIY / at home builds of this mask. Smooth-on has a really good video on making gang molds of masks and they use this linear seal...

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

Oh, I see. That subdirectory is not for the mold of the seal as published here but an injection-mold for a complete mask.

Thanks for the thicker version. I already printed and coated mind and the single piece mold for

the seal currently has the silicone curing. (The contour will probably be a bit too short in the first revision.)

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

On that Google Drive link, you have placed the files for Contour* in the subdirectory for the mold.

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

You have not answered the question.

"Why do you use a 2 part mold anyway?"

Mini x6nezqxo 100x100MalaMaker commented ago

@Marcus Wolschon The model has been available this whole time. See the link that is labeled "Mask Design" in the description. Also I have made a 1.5mm thick walled version of the mask shrouds if you would like to try them. This is not something I am going to just openly publish on this platform or thingiverse but you can find it under the project folder. If you look under "Models -> STL -> Mask - ... -> Thick_....stl" , you will find each size with a thicker shroud.Β 

Project folder:Β

This is a fine tuned project because these masks have to fit and seal properly! This is not a novice project. I have documented things to help in the learning curve, please read. Understand that collectively there are 500+ work hours that have been put into this by the contributors. We are trying to help, not trying to cause anyone problems. The models are a very complicated design because it has to fit very complicated anatomy.

This is to everyone and I am saying this because I am being trolled on another platform. Please be courteous in your comments. Please do not be condescending. This is not political... leave politics out of this! Please understand that I know everyone has an opinion and I respect your opinions. I am doing this project to help fill a gap to help keep people safe and give them data and facts that they can make an informed decision. I am simply trying to help my fellow human being regardless of who you are, where you are, and how you live! period! We are all in this thing together!

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

Here is a trival 1 part mold I through together.

The extrusion pass is very roughly meassured because you didn't publish your original design files (very poor style).

So I had to so some rough meassurements on the exported STL.

A531cdef4f7baf439ea9742a3e492ef5?default=blank&size=40Hindimied added this to the Cora Cora collection ago
730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

Why do you use a 2 part mold anyway?

Since the upper side is flat, you could have used a very trivial one part mold with an open top

and nothing could possibly go wrong.

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

Try sealing the mold with something easy to remove, so it doesn't leak.

(e.g. a leftover drop of silicone from another casting some hours earlier)

Mini x6nezqxo 100x100MalaMaker commented ago

Marcus Wolschon... The mold had a ledge on one of the molds right at the top surface. The slicers never would have tried to print it because of how thin it was but I cleaned that up so the ledge doesn't exist.

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

I tried subtracting a cleaned up 3D scan of my face.

But it wurned out that it's MUCH easier to heat up the mart "Contouer" with a filter assembly attached (so the thread doesn't warp out of shape) and a gasket

and push it to your face to have it perfectly conform to your face.

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

What has been updated on the small silicone mold?

Mini x6nezqxo 100x100MalaMaker commented ago

An updated small silicone mold has been posted.

2fb7ab92bef88047ddaba3be65c908c5?default=blank&size=40Scott Ferguson commented ago

@MalaMaker I printed V3 and it's stronger, but my printer still prints it as a series of adjacent squares. When held up to the light, you can actually see empty space between each square.

Would solid bars 1/8" wide, with 1/8" space between them be that much more prone to physical contact with the outside world? Or perhaps I should solve my problem by buying that resin printer I've been wanting. :-)

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

"Filter Body" should not be perfectly symmetrical but link with a freature of with the "Mask Shroud".

So it doesn't turn freely with no way to hold it when trying to thread on "Filter Ring".

Mini x6nezqxo 100x100MalaMaker commented ago

@Scott Ferguson The v3 prints really nice. I think I am going to replace it with the current version.

Mini x6nezqxo 100x100MalaMaker commented ago

That is something I have considered but they need to be secured in place so people can't touch the filter. If people do re-use the filter which I know is happening with standard N95's, touching the filter can contaminate and obstruct the the filter materials. That is why I have the honeycomb design... to keep people from making contact with it after the filter is installed. This mask was reviewed by a mask expert in R&D at a major N95 mask manufacturer and they noted that keeping people from touching the filter is good and notated the contamination and obstruction through contact. V3 is what I am leaning toward. Test printing it now.

"Thank you for uploading the revisions. I think V3 would be a much more durable design for my printer. Have you considered making the openings a series of wide, narrow slots, so the filament is laid down in long, contiguous bars instead of a series of small squares?"

2fb7ab92bef88047ddaba3be65c908c5?default=blank&size=40Scott Ferguson commented ago

Thank you for uploading the revisions. I think V3 would be a much more durable design for my printer. Have you considered making the openings a series of wide, narrow slots, so the filament is laid down in long, contiguous bars instead of a series of small squares?

"@Scott Ferguson... I am working on an update to the body and the honeycomb. I have found that the current honeycomb is difficult to print with the heavy retracts. I have a v2 and v3 I'm test printing now in the google drive project folder if you are interested."

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

I'm printing with nGen for years and it never, ever delaminated.

From week long printed masters in 40 and 20 micron layers to super fast stand-ins in 0.6mm to strong and never leaking mother molds in 0.3.

It's more well tempered in it's temperature range then even PLA.

Sterilization over here happens by bathing the parts. Heating would damage the straps at the very least.

I'm using the project as a starting point to experiment with subtracting a 3D scan of a face from the mask with and without a silicone seal. It worked well in the past for custom fit electrooculography rigs and now I'm looking, if it can be set up as an automated pipeline for many faces.

Mini x6nezqxo 100x100MalaMaker commented ago

@Marcus Wolschon We are saying 255c just to guarantee layer adhesion but if you can get a print at a lower temp out of PETG that gets a good seal... great. These are recommendations, I understand everyone's printer prints differently. FYI with nGen... don't use it for the masks... we already tried it and it delaminated all over the place. Great for a visual print and looks awesome but not good for a usable part. PLA is another filament you could use but it can not be dry heat sterilized, that is really the only restriction to that filament.

Let me know if you have any other questions. I appreciate the feedback.

This project is more about the filter research than it is about the mask but the mask has had a lot of work and design considerations put into it. We want people to be safe, this is the driving effort to all of us designing and printing PPE.

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

"@Marcus Wolschon see the print settings on the Project Home page."

Ah, damn. "Filament: PETG"

I just removed the PETG from my printer because after 2 days of trying I still can't get that damn stringing under control. Already got back to nGen and will never, ever return to PET-G.

255Β°C is the very upper limit of the PET-G sold by Colorfabb and would made the stringing even worse.

Mini x6nezqxo 100x100MalaMaker commented ago

@Marcus Wolschon see the print settings on the Project Home page.

This is not a vase mode print, the print is designed to be a double shelled mask and we are passing N95 vapor fit tests in a hospital fit test with the mask and filter combinations we are having lab tested. It is critical that this mask be a double shell or you will run into sealing issues. Recommended nozzle sizes are 0.4mm-0.5mm anything bigger than that the slicers will do more of a vase mode to the mask.

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

This seems to list the standard threads.

NATO is "STANAG 4155" 40mm, 1/7”=3,63mm

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

Looks like most gas mask filters use a 40mm thread.

The difference to 68x3mm or 62x4mm should be large enough, so an adapter to a standard size that others are also using is possible.

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

First print failed because "Mask Size Contour Large" seems to be a vase print and the walls don't come out anywhere near air tight.

Since I have no model to modify for thicker walls,

I'm improvising and trying again with my 1.0mm nozzle but this time 150% flow rate. I may have to file done some parts later, so that it still fits.

Mini x6nezqxo 100x100MalaMaker commented ago

@Marcus Wolschon... At this time I am not posting the project files. The thread profile between the body and the ring is ISO Metic 68x3. The thread profile between the body and the cap is ISO Metic 62x4. There is a filter body long, that was actually the original design but the nose was shortened because there was wasted space. The original long design can be found in the google drive project folder.

@Scott Ferguson... I am working on an update to the body and the honeycomb. I have found that the current honeycomb is difficult to print with the heavy retracts. I have a v2 and v3 I'm test printing now in the google drive project folder if you are interested.

730a061596c420c8915d9f20bd5d89e1?default=blank&size=40Marcus Wolschon commented ago

Could you please not only publish the exported STL meshes but also the original project files? Thank you.

Does the filter attachment use any standard filter thread/bajonett? If so, wich one?

You have "Filter Attachment Short *". Is there also a "Filter Attachment Long"?

2fb7ab92bef88047ddaba3be65c908c5?default=blank&size=40Scott Ferguson commented ago

I can't thank you enough for publishing this for all to use. I have printed this on an extrusion printer using PLA filament, and I have a question: The two filter screen elements appear to print correctly, but they are quite weak. It prints as a mass of tiny squares which don't bond strongly with one another. Do you have any suggestions which print parameters I might alter to strengthen this part?

401749b2db7871a0974b685ed8ca4152?default=blank&size=40John Carbahal added this to the Tools collection ago
65ea4687e323b1164d21a93369af1f7b?default=blank&size=40mmxx commented ago

Great work!

Bc6a8a8a4d8c8b0a5d5e49e49717c0c2?default=blank&size=40jabab added this to the covid-19 collection ago
Mini x6nezqxo 100x100MalaMaker commented ago

New models with lips on the edge of the mask are posted. This will help keep the silicone attached.

Mini x6nezqxo 100x100MalaMaker commented ago

Quick heads up, I pulled down the contoured masks and silicone molds temporarily because I found that they need a lip for the silicone to seat correctly. Models will be back up later today.

Mini screen shot 2013 07 15 at 6.49.36 pmErik de Bruijn added this to the COVID19 collection ago
Mini x6nezqxo 100x100MalaMaker published this design ago