Solution for the AirMini Travel Noise

[JimIve]

I've created some 3D printable adapters to allow the standard airmini attachments to work with the Q-Tube muffler.

The AirMini uses a high speed smaller fan to achieve the same pressures as an AirSense 10.  The problem with this
is that the fan generates more noise and combined with the airflow, increased noise is generated in the pipe
(put the pipe to your ear to hear).  This noise also exits from any vents (but it's not the humidifier chamber generating it).
Using the Q-Tube muffler results in a 10db improvement which is about 1/2 the noise level.

In comparison to a standard AirSense 10 with a P10 mask, it is about 1/2 again, so for Noise levels:

AirSense 10 with P10: 1X (using this as the baseline)
AirMini with AirMiniP10 and Muffler: 2X
AirMini with AirMiniP10: 4X

Having the humidifier in the AirMiniP10 mask or not made little difference.

Credit for the original idea this is based on Q-Tube Muffler / Adapter however this design is completely from scratch.

The differences between the original idea and this is:

• It allows the masks that come with the AirMini to work with the Q-Tube Muffler
  
• Male AND Female to 22mm Male Adapters
  
• Female Coupler for 22mm to 22mm
  
• Rubber gasket for Female Adapter (i.e. same approach as airmini to get a good seal)
  
• Rubber gasket to prevent the clips breaking on the Male Adapter and provide retention
  
• It's a parametric design using OpenScad, i.e. you can change it if you need different tolerances or adapters
  

Printing instructions and disclaimers can be found in the .scad file.  .stl's are included.  PLA and NinjaFlex are used.

Future improvements....  I plan on designing a 3D printed muffler to improve noise reduction further if there's interest.

The intention of this is to allow the use of the standard AirMini masks with the muffler.  However it will also allow the
use of other masks (unintentionally).  Warning: Other masks may not have the required Carbon Dioxide exhaust or produce the same throughput/pressure with the AirMini's high velocity fan.  The AirMini has a small high velocity fan, which is why it only works with certain masks.  I.E. You use this design completely at your own risk and you could suffocate or not get your prescribed therapy.   Consult your DME.





3D Printer Zip File

[JimIve]

And here's the noise comparison.  Noise levels where recorded 1cm away from exit vent on P10 (humidifier exit vent on AirMiniP10) at 8PSI whilst breathing normally and are of the Inhale phase (the loudest).

[ardenum]

if youre designing a custom muffler do a kickstarter or something. I personally don't care about the noise i always sleep with silicone plugs.

[2dogs]

Hi Jim,

Thank you for putting up the .stl files for the AirMini. They were just what I was looking for. I am a novice to 3D printing but our local library had one and I have printed both adapters. I did one in ABS and one in PLA because that was what was on the printer at each time it was availble. I have to redo the ABS in PLA because of a crack at the bottom of the slot on the machine adapter. Until it fractured, I was wondering why the slot lengths were different. I now understand.
This project has got me going in 3D and I have ordered a build kit for an Prusa i3 so that I can learn about it and do more in 3D printing.
Can I ask, how did you create the initial 3D model? Did you create it from measurements in a CAD program or did you scan it in somehow.

I was also wondering how you were going with getting unencrypted data directctly off the airmini board.

Thank you again for your initial posts

Simon

[ardenum]

Probably Fusion 360, I use Solidworks myself.

[JimIve]

Hi Simon,

Glad this got you into 3D printing, it will be a lot of fun, and building it yourself will be even more fun :-)

The reason for PLA instead of ABS was because of off gassing.  PLA is more natural (based from corn) so less likely
to produce volatiles, where as ABS, while minimal when in hard form, if I was using them very often, I would want to avoid it
going into my lungs, otherwise occasional usage during travel I wouldn't be too concerned.

I've used this setup now for a couple of trips, and it's I'm sleeping fine (same as my regular AirSense P10) as opposed to as
shipped, where I pulled it off after 2 hours due to the noise.

Well spotted about the different lengths on the slot, it was to make it less likely to break at that point, i.e. make it so the filament direction didn't line up with the same point on both slots.  Also, the Ninjaflex insert prevents some resistance to prevent the clips
bending too far (i.e. past the break point).

I modeled it using OpenSCAD.  It's more of a programmers 3D modeler, but is worth the time taking to learn, if you're not a programmer, if you can program then it's easier to learn than Fusion 360.  Fusion 360 is good for pretty stuff (curves) and a behemoth of a product, can be slow.  OpenSCAD is better for practical stuff, where you need to change the fit as it's
parametric by nature.  I used calipers to measure the ResMed attachement to replicate that, and then modeled a negative around that for the opposite end, comparing with the inlet on the machine.  Took a few iterations to get the fit correct.

Did an initial pass at getting the data off the eeprom, but ran into some issues (data didn't appear valid, and couldn't find a trace that would hold the processor in reset)...so as I was traveling I put it on the back burner.     So I'll take another pass at it (likely today), and see if I can trace down a reset line (one of the problems is that the processor is a BGA, so all the connections to the traces are hidden without an XRAY).  If that fails I'll desolder the chip and read it that way.

[JimIve]

So I'll take another pass at it (likely today)

See update on this post: http://www.apneaboard.com/forums/Thread-...n-Teardown

[ardenum]

jeez @ OpenSCAD , looks terifying. Also you could have it send and printed in Polyamide 12, its what they use for Narval CC. Polyamide 22 is same material only powder form to use on SLS. I had my Narval knock-off printed in Polyamide 12 on a DLP HP printer.

[JimIve]

jeez @ OpenSCAD , looks terifying

I supposed it depends on if you're can program or not.  To me, it's cleaner, easier to maintain designs and just works.  Fusion360 and solidworks to me are bloatware in the extreme.

Parametric is also important.  It's also possible to provide parametric designs to others with OpenScad and Thingiverse.

A .stl only gets you so far.

But if you can't program, something else would be an easier choice unless you're prepared to put in the effort to learn.

[Frequenttraveler]

So is the Q-Tube the muffler you made or something I can buy? I don't understand this 3D printing thing, I just need to be able to travel with my airmini and have less noise so I can sleep!