A look at NanoHack 3d printable respirator mask

Disclaimer: I’m not a medical professional or a scientist or an engineer. I could be completely and totally wrong. And while we’re at it, this is true for nearly everyone you talk to. Follow my advice at your own risk. But these are extraordinary times so…

What is the NanoHack?

Simply put, the NanoHack is 3d printable personal respirator design. Its home is here: https://copper3d.com/hackthepandemic/

Does it work?

I sought to test it out. The day before I was made aware of this particular design, I had already been contemplating making a mask design myself. But I put that aside to evaluate the NanoHack.

The answer as to its veracity isn’t a simple one. Some short answers that don’t quite get to the crux are:

  • If you just print it out of PLA, fit it and put it on: no it doesn’t really work as a filter, as is. But maybe it’s better than nothing?
  • If you print it out, get a really good fit and know enough to actually get a proper filtering material into the modular filter holder: it can theoretically work. But the design makes it unlikely to really work.

Why does this matter?

The other night, the Rachel Maddow Show did a segment on hospitals and their need for personal protective supplies. There were multiple public calls for action documented within. This included people sewing masks and also; a note that a Boston official had made a call out to people with 3d printers to print out a particular mask.

I dug up the actual source material and found this: news report that in turn led me to further twitter posts and the like. And eventually I was able to confirm that the design was the afore mentioned NanoHack from Copper3D.

Now, I want to stop here for a moment. At this point, I was extremely skeptical. Because it became clear to me that a hospital administrator was panicking and asking for people to make a thing that was far outside his area of expertise to evaluate. I assume, in hopes that it was some kind of magical (Clarke’s third law) solution to a very real logistical disaster.

It was also clear to me, that regardless, a whole lot of people were going to jump on this plea, because it comes from the right place. And also, that almost all of those people reporting, retweeting and pleading are completely unqualified to be able to be as minimally skeptical as I was.

Look, most ‘3d printing experts’ are pretty terrible. They’re hobbyists. And if the worst that happens when they’re wrong is that someone’s man-cave burns to the ground, maybe that’s okay (so long as their man cave is not attached to their home). 3D printing forums are full of some of the worst self proclaimed experts I’ve encountered. So, I fully expect large numbers of very reputable 3d printer folk to jump on board without getting it right.

Also, it was clear to me that I’m not a doctor or scientist (I only have a MS degree in an unrelated field) and that I too could be wrong.

But it does matter. If even one person prints one of these things and gives it to a medical professional to use as is, it matters. So, I decided I needed to print it and evaluate it the best I could.

The filament (material)

Copper3D makes it pretty clear that they originally meant the mask to be printed in their own PLActive. But as you might imagine, it’s sold out. Very few people would happen to have a spool of that product sitting around.

Now again, I immediately thought the worst (such is my nature). I thought they were telling everyone to buy their PLActive filament, to save the world. But after reading through their site and supplementals multiple times, I came to the conclusion that they were clumsily getting a pitch in for their filament; while suggesting the design could be useful when made of other filaments (materials). And I agree. I just wish they were more concerned with explaining these things, than with getting their pitch in.

If you look at their marketing materials for PLActive, you see that its primary use is more for objects that should be resistant to bacterial and viral infection, like prosthetics. The idea for printing the majority of the mask from this material likely is that it’s better to be using an anti microbial material than not. But any kind of respirator clearly is going to need to be cleaned constantly. You’re breathing through\on it. An anti microbial material alone is not sufficient to negate the need to clean it. And further, I’d point out an anti-microbial material probably isn’t enough to relax the cleaning schedule either.

Working around the lack fo PLActive should be simple. You need to be diligent at washing your mask anyway. I therefore, chose to use simple PLA filament (which is biodegradeable), which doesn’t have the copper additive that Copper3D claims is the active aspect of PLActive.

Of note here is that it’s fairly well accepted within the 3d printing community that fused filament (FF) thermoplastic prints are generally not ‘food safe.’ The reason given: all the ridges and nooks and crannies of the FF process can theoretically become breeding grounds for bacteria. Some good advice I’ve seen is:

  • 3D print a prototype coat and smooth it. Make a mold of it. Cast it out of something that’s more appropriate.
  • Seal the print with something that’s food safe, that’ll also fill in some of the micro-structure.

Since the issue of ‘food safety’ is probably congruent to the issue of keeping the respirator disinfected, it’s good advice to keep in mind.

The design files

Getting the files was easy enough. They provided STL files. I would have liked for them to also provide some actual CAD files to make it easier to work with the design. They do call it the NanoHack, which implies a need to (hack) change the design. The mask is in three parts.

There’s the large flat mask with a threaded filter cylinder. The base of the filter cylinder has a vent in a lattice pattern.

The second part is the cap for the filter cylinder. It is threaded as well, and has four inlets at the top. They are angled. Seemingly to protect the filter cylinder’s contents from direct spray from the outside.

Third, there is what appears to be a filter disk. It’s a waffle cut kind of design.

The filter

[Deep breath]

Let’s start with the most important part of the NanoHack website. It’s step 5 of the assembly instructions:

5. Screw the “cover” on with one or two additional filters. If you use a filter, you can use a circular piece (single or double) of a conventional mask (nonwoven propylene) or add a round facial cotton pad for makeup. Half pad can make breathing easier.

There’s two really important things to glean here.

First, they seem to want us to believe that the filter cylinder, with one or two of those honeycomb filters, is somehow appropriate for personal protection. This is laughable. Even if those honeycombs were printed from their PLActive material, they’d be woefully inadequate. A material that deters microbial growth on its surface over a period of hours, will not filter your air when it’s got a bunch of 1-2mm holes in it for the air to pass through.

Second, they suggest that “if” you use a filter, you can use a couple of different materials to make it. I’ll get into filter materials later.

Let’s start with the obvious bit. They start by saying you can put 1 or two of their filters in, then they say if you want to use a filter, don’t use their filters, but instead use a filter. I do not think they know what the word “filter” means.

Let’s give the best possible reading here, in favor of Copper3D. I think, maybe, what they mean to say is: The mask can be minimally useful to you just by being a physical thing on your face that you breath through. And additionally, to the extent that you can add some geometric complexity to the path the air takes, you can get some protection from droplets and such. And should you want to do that, you can drop 1 or 2 of our honeycomb objects in the modular filter cylinder to add some geometric complexity beyond the basic mask. But, if you want a real filter, you’ll need to make one that’s effective, and install it in the modular filter cylinder in an effective manner.

They point out two materials: non-woven propylene and facial cotton pads.

Again, I’m going to point out the device is called the NanoHack. And this is where the “hack” part begins. Though they’ve done a horrible job of making that clear.

I’d just like to link here to some actually useful source material on filter materials. DIY face mask. The referenced article cites its scientific source material.

If you take those two bits in context of one another, you get what is meant. That a disposable filter can be made from the same material as surgical masks. And a reusable, launderable filter can be made from 100% pure cotton. In theory anyway. The afore linked article shows why the cotton would be a little less effective but would still be effective enough to use.

The print

Doing the print was fairly painless. PLA on my PRUSA MK3s MMU2s went smoothly. It took about 4 hours. That print time could probably be optimized easily.

I opted for support only for the cap. The other parts didn’t need any support. The supports dropped right off.

The end of the thread in the filter cylinder dropped a little. Later on I hit it with a heat gun to keep it from being a problem. One could also just cut that part away with a knife. I don’t think support material would be worth it to fix that one little thing though.

The fit

One of the primary features of this mask is that it prints flat, and is then “fit” to the wearer. The fit process can be done with a hair-dryer or hot water. I happen to have a heat-gun, which is really well suited to it. But I suspect you can re-melt the plastic with all kinds of DIY tricks.

If you go by their 3d renderings, you’ll assume it’s more of a technical folding to get the mask into shape. But what really is missing is an explanation of just how much it’s actually a process of creating a personal molded fit. You are doing something that is sometimes antithetical to a 3d printing enthusiast. You are sculpting it by hand for each owner.

30 minutes into the process I’d gotten pretty good at heating up the part I wanted to adjust, bending it into rough position, and putting it on my face (without getting burned) for fine adjustment. You can get a very tight fit this way. And the fit is absolutely critical.

You will want to test the fit constantly, by lightly pressing it into your face from the front, and breathing. Then, check for where the air is coming out. With my glasses it was easy to tell when it was leaking from the nose bridge, because my glasses would fog up. Another way to test, is to just put your hand over the open filter cylinder. If you can feel hot breath, you’re probably sealed, if not, find the leak.

There is however a limit to how well it can be made to fit. And also, your face is not a piece of unmoving granite. It changes shape with expression and movement. So, while a good fit is critical, its not actually possible to assume a perfect fit. This is the biggest problem with this design. It’s made from ridged plastic and assumes a nearly perfect seal. But that is impossible.

Rather, the only way it really can work, is to get the best seal possible, and work with the filter design to get that imperfect seal to be good enough. More on that later.

I then used a 3d printing pen to close up the seams of the printed design. If you don’t have a 3d printing pen, you can probably use a bunch of things. A hot glue gun comes to mind. Caulk might work. PVA glue (Elmers glue or wood glue) probably would work (but is water soluble and might need to be done again after washing.)

I used some rubber bands at first as straps. Then I moved on to attaching some packing string loops with the 3d printing pen. I have some thick ribbon I might try next. You’re sort of on your own for straps. They made nice big slots to work with.

Evaluation of the stock mask

Okay, so. The mask, printed, assembled and fit. No hacks. No options. How does it work?

Not so great.

So, as I was getting at earlier, the fit cannot possibly be perfect when you’re working with ridged plastic because your face moves and plastic doesn’t. And if you assemble it and fit it really well and try to breath through it, you’ll realize pretty quick what’s going on.

The cover and the honeycomb filters restrict airflow. The geometric complexity of the air path offers resistance. This is probably a good thing if all you are depending on to protect you is that geometrically complex air path. But if the resistance offered by the filter assembly is greater than the resistance offered by the edges of you mask, your breath will simply escape from the edges of the mask, rather than the filter assembly. And this is what happens most of the time. Even with a really great fit, you have to focus on keeping your face relaxed just right to keep from opening up any seams on the edges. It’s easy to hold your hand in-front of the filter cap to tell if your hot breath is going through the filter assembly or not. You can alleviate this some by making a tighter strap. But there is some point at which you are fighting a losing battle.

Of course, if you don’t care about your germs escaping into the world, and only care about yourself; you might point out that when you breath in, it could still be working. That’s easy to test too. Just cover the holes in the cap and breath in. If the mask suctions down to your face, you’ve got a seal. If not, the air is coming in from around the sides of the mask just as it was escaping during exhale. Can you relax your face and apply just the right pressure to make this work occasionally? Sure, but during average use, it just doesn’t stay sealed easily enough.

Getting crafty

So, I happen to have sheets of adhesive craft foam lying around. So I decided to cut some strips and try and make a bit of a gasket. The adhesive was only good enough to hold things in place while I affixed them more permanently with the 3d printing pen. Again, here a hot-glue gun probably works just as well.

With a reasonable gasket in place, the seal got better but still was not perfect. Or should I say, was still not good enough to satisfy me that most of my breath was going in and out of the filter cylinder during normal use.

I did actually try pulling out the honeycomb filter. There was less air resistance but still not little enough to allow the light face seal to be strong enough. And I realized the cap itself was too restrictive and was the main culprit.

I did try putting a cotton pad in there as Copper3D suggested. It was nearly impossible to get any air to pass through the filter cylinder at that point. I tried cutting it down as they suggested. I never got it thin enough to have a chance of being effective.

I also combined knowledge from the afore linked filter material article to try and make it work. I printed a ring in the right size to stretch a piece of 100% pure cotton cloth to the width of the filter cylinder. I put that in the filter cylinder. And that showed a lot of promise. However, it was clear to me that the cap was too restrictive. And that if I wanted to go the 100% pure cotton fabric filter route, I’d need to redesign the cap at least to allow it a chance to work.


At this point, I reached a fork in the road. It was clear the stock mask wasn’t really useful.

I had some ideas on how to hack the mask and modify the design to actually be useful and successful. And I have moved forwarded with those efforts. But that’s a matter for a different article (soon). For now, I just wanted to share my evaluation of the mask, as originally published.

It doesn’t really seem to work. At best it’s ineffective. At worst, it might be fooling a whole lot of people into thinking it’s doing something it’s not. I think its imperative that people understand what it’s actually supposed to do and what it actually does.

It’s a platform off of which to hack something that might be useable in some situations. That’s all it is in my view.


But here’s the thing.

The afore linked article on mask filter materials makes it very clear that 100% pure cotton is reasonably effective. And further, that cotton blends are also reasonably effective (maybe more so) but are a little harder to breath through.

If you have the ability to just sew/tie a mask out of those materials, that’ll work. They are reusable/launder-able. So, before you go down the path of trying to construct a highly customized ridged respirator/filter-holder that may or may not actually work, consider if you’re making the problem more complex than it needs to be. Maybe thread and needle is more practical to get a filter thing you can breath through. The virus does not actually care how even your stitching is. What works, works.

Also, if you actually have surgical masks from which to cut filters for the 3d printed mask, why would you not just use the surgical mask itself instead? Using it in this mask as a filter doesn’t change its disposable nature. At best, you’d be able to stretch the use of the mask by cutting 4 filters from a single original mask. And maybe that’s something. Maybe it isn’t.

Also, consider that even if the mask does absolutely nothing to filter air, it can at least keep you from touching your nose and lips, and from direct spray from other people coughing and sneezing. And that ain’t nothing.

My best guess, is that the only way this 3d printed mask will be really functional, is if the basic geometry of the air-path proves to be a layer of protection on-top of a good filter material. That’s why I’m going to keep working on improving the design on my own. Because that may be possible. Will I be able to test its efficacy versus a sewn mask? No, I don’t have that kind of equipment or expertise. But if I can get a 100% cotton filter in-line and can be confident that the seal is good enough that my breath actually goes through the filter, then I will be at least fairly certain it’s as-good and leave the questions as to it being “better” for calmer times.