Mobile continuous gas mixing controller

Continuous gas blending for scuba breathing mixes.

Similar projects worth following
Various mixes of oxygen, nitrogen and helium are used as breathing gas for recreational diving. One way of blending these gas mixes is by using a continuous mixing method where oxygen and helium are added to plain atmospheric air at the inlet of a high pressure compressor. By measuring oxygen content at the inlet and regulating oxygen and helium flow using motorized needle valves the proper mix is created. The project is started with an oxygen control system, helium is added once the oxygen system is fully functional and tested. Safety features for gas and compressor shutdown are all included in one mobile case.

More details and progress will follow shortly.

  • 1 × Mix Stick Device build from standard plumming parts that mixes oxygen, helium and air using baffles.
  • 1 × Air filter To prevent the entry of foreign matter into the system, an air filter is added to the inlet of the mix stick.
  • 1 × Oxygen sensor A fuel cell type oxygen cell is placed at the end of the mix stick to analyze the gas composition. The analog output of the cell is used to control the oxygen flow by adjusting the needle valve.
  • 1 × Needle valve Flow control device to adjust the amount of oxygen added to gas going the compressor inlet. A stepper motor is added to control the needle valve.
  • 1 × Low pressure regulator Pressure regulator to reduce intermediate pressure to approximately 3 bars. The pressure setting of this regulator determines maximum oxygen flow.

View all 10 components

  • Manual continous flow set-up online

    ErikH06/01/2018 at 12:23 0 comments

    Right...  Long time no update. But good news, the manual version is up and running. The mixstick operates as it should and the normaly closed oxygen shutoff valve (2) works. A socket (1) is connected in parrallel to the electric motor and provides power to the shutoff valve when the compressor is running. This way the shutoff prevents oxygen from pooling in the mixstick/hose when the compressor switches off (which could lead to Bad-Things[tm]). Next up is a microcontroller controlled safety shutoff at 40% O2. Parts are ordered, so stay tuned :) When that works, automatic flow control is ofcourse next! Pictured below is the compressor (Bauer Junior 100L/min) and mixstick.

  • Spring is here ;)

    ErikH05/11/2016 at 12:46 0 comments

    Spring is here :) Time to get this page moving again soon. Still loads of other projects, but also found a compressor and other parts to get this project going. It might become a 2-step project though, as a wall-mounted system is more easy to optimize before building the mobile version..

  • on hold till spring...

    ErikH01/12/2015 at 08:26 0 comments

    Due to moving into a new home I'll be spending my time working on home improvement for a while ;) I'll be back in spring for this project.

View all 3 project logs

Enjoy this project?



bonsixo07 wrote 03/09/2022 at 14:35 point

this idea sounds cool. I love this project. I am also working on a gas grill blog. I hope this will be helpful for me.

  Are you sure? yes | no

Vladan wrote 10/26/2018 at 12:39 point

Could you please post some more details on the makes & models of the components used? Maybe some links?

Is your mixing stick homemade or?

Would be nice to have more detailed pics even in the manual stage :)

Good luck with the project!


  Are you sure? yes | no

ErikH wrote 07/15/2014 at 08:38 point
Sounds like a whole different kind of diving jlbrian7 ;) 86/14, yikes! With today's helium prices that would cost a fortune :P I already hate to get my tanks (twin 12l.) filled with 21/35 around here. That's already 50€/68$ for one dive.

For scuba, the gas is blended either as 'best mix' or 'standard gas', depending on whether you prefer cheap gas or easy deco calculation. Most popular gasses would be 32% nitrox (32% oxygen/68% nitrogen) for up to 30m/100ft. Deeper diving would require 21/35 for up to 45m or 18/45 for up to 60m (oxygen/helium and rest is nitrogen). These mixes are easily made by decanting the right amount of helium in empty tanks and then topping up with 32% nitrox. For decompression 50% nitrox or 100% oxygen are most common.

  Are you sure? yes | no

jlbrian7 wrote 07/14/2014 at 16:57 point
Checked in on your page because I hadn't seen any updates, and was hoping you didn't get a mix wrong, but it looks like there are some very knowledgeable people on here. SCUBA is obviously very different from surface diving. I spent most of my time on an inspection boat and the deepest we could go was about 285', but the only mixes we used were an 86/14 Heliox that we would switch over to at 150' and a 50/50 nitrogen-oxygen mix that we would go on at 90' on the way up for in water deco. Our gasses were purchased already mixed in 800cuft. bottles and we would just regulate them down to about 150psi. We would use a ziplock baggie and an O2 sensor to make sure the helium did not leak out, but the sat boats would have large vessels of helium and oxygen and mix them to best fit the working depth with sat techs constantly monitoring the situation. Talk about alot of valves that looked like a high stress job.

  Are you sure? yes | no

ErikH wrote 07/14/2014 at 08:30 point
Hi Paul,

don't worry, I know what I'm doing :) Continious blending by adding oxygen (and helium) to the inlet of a SCUBA compressor is a tested method. Safety precautions are needed, as a too high oxygen content wouldn't be very good for the compressor (to say the least...). An automatic shutdown of the compressor and oxygen flow will occus if the O2% will be higher than 40%. Setpoint will usually be 32%, so adding ~16.5L of O2 to every 100L of air. O2 will be injected just above athmospheric pressure, but comes from a 50L/200bar cylinder. The mixer design is, simple, but has been proven sufficient and does not add too much of restriction to the compressor inlet, which could harm the compressor.

The system is more complex than partitial pressure blending, where first pure oxygen is decanted from the large storage cylinder to the smaller scuba cylinder and air is filled on top. However, as you already mentioned, High pressure O2 has it's risks and in my system, even oil contaminated scuba cylinders would be safe to fill, as they never see more than 32% O2.

I'll add more info on the project in the next couple of days, so stay tuned if not all is clear ;) Alternatively you could Google 'continuous flow nitrox blending'.


  Are you sure? yes | no

Adam Fabio wrote 07/14/2014 at 03:05 point
Thanks for entering The Hackaday Prize Erik! Can't wait to hear more details on your gas mixing system. Give us some updates soon!

  Are you sure? yes | no

paultcmgp wrote 07/13/2014 at 06:28 point
Adding additional compressed gas components to the ambient air is possible, providing you have high enough pure gas pressures to cause the flow to go the way you want it to. Normally, these kinds of mixtures are produced synthetically, meaning the master mixing bank of cylinders is evacuated by high vacuum, then the He or other minor components quantities are added first. The oxygen goes in next, with the nitrogen (roughly 80%) last. The mix vessel has to be homogenized, either mechanically or thermally, as gas stratification occurs at these pressures (~3000 PSI). The highest grade of gas mixtures is performed gravimetrically, where the individual components are weighed into the tared cylinders using large precise digital scales.

Doing the mixing by adding to the compressed air can be attempted, but careful use of gas concentration and pressure sensors is crucial. I see on your schematic diagram a baffled pathway for the gas. This may provide some mixing, but you may find that a venturi or perforated educator tube inserted beneath the valve collar will give better homogeneity. Supplemental cylinder rolling will be necessary for absolute mixing in either event.

I don't want to deter you, but dynamically mixing as you describe is routinely done by gas divider/mixing panels ($4k-$15k), but usually with the final gas stream at near ambient pressures. Doing what you describe with the final mixture at compressed cylinder levels (roughly 200 atms.) will be very "challenging", and probably cost prohibitive.

You have to start out with gas supplies of at least the same pressure as your final product, as using a compressor on the mixture will introduce a fine mist of petroleum used to lubricate the pumps unless VERY expensive and specialized equipment is used. Doing a simple mixture by vacuum evac., followed by individual gas additions measured by partial pressure gauges on a mixing manifold is the most cost effective way to do this, and it is done this way professionally.

The electrochemical sensors used to measure O2 at ambient pressures do not work well at high pressure. You will need a paramagnetic O2 analyzer (>$5K) or a gas chromatograph (>$15K) to accurately analyze your final mixture. The final He concentration will be nearly impossible to accurately determine without a GC fitted with a thermal conductivity detector.

Finally, compressed oxygen can by explosive if introduced to any flammable gases, so starting gas purity (no methane, propane, or other hydrocarbons) is essential. Also, even with accurate needle valve controls, final O2 concentration MUST be performed by analysis, as higher than normal oxygen concentrations are deadly to humans at high pressure. Low levels of O2 in a SCUBA tank are obviously not healthy either.

I know I sound like I am taking you to task about this, but I have worked in this field for many years, and I merely want you to know the challenges you will face in taking on a project of this nature. Learning to control needle valve ports and mixing technology is valuable knowledge in its own right, but compressed gases are intrinsically dangerous, and proper precautions must be observed at all times. Good luck in your project---dr-zin

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates