DIY SCUBA Tank Boost Pump (for Mixing Gases)

No love for that sweet old NES controller? Fine, here’s a bonus hack for you. This home-built boost pump is designed to take in a volume of gas (like oxygen) and compress it with a hydraulic style air powered cylinder to raise the pressure. With the increased pressure, the O2 can be added to a SCUBA tank that contains high pressure air to create what’s known as NITROX. The idea is to increase the percentage of Oxygen in order to reduce Nitrogen absorbtion in the blood – increasing a divers safe time at depth underwater.
The compression cylinder is built entirely of brass with Sirvon seals. The drive cylinder and 2:1 lever produces 3,000 lbs of pressure on the pump cylinder, all from 120psi of pressure from a standard shop air compressor.

38 thoughts on “DIY SCUBA Tank Boost Pump (for Mixing Gases)

  1. This is quite dangerous. The design is using shop-air for the first stage, which is the first dangerous aspect due to the lubrication oils and water vapor present in the output line. Also, being a nitrox certified diver myself as well as a chemical engineer, the air mix is crucial down to the percent… 32% and 34% are drastically different when you have it strapped to your back when diving.
    Anyways, a main fact about filling your scuba tanks is to keep all contaminants out, since there isn’t much knowledge on what they will do to you when you’re down 35+ feet, but a guess is that it isn’t good and I wouldn’t recommend it…

  2. high concentrations of 3000psi oxygen in conjunction with some stray oil means kaboom. it takes a weird mix of high level knowing AND not knowing what you’re doing in order to attempt stuff like this.

  3. ah, I see now that he’s using a source tank of O2 rather than pumped atmospheric… sorry, 2:30am.
    regardless, i still wouldn’t trust the “drive” cylinder to fill tanks… I think the $12 fill at a dive shop is worth it and recommended in this case.
    though, I do like the design for possibly compressing other gases… perhaps hydrogen from an electrolysis cell? use CO2 and discharge it to make a simple source for dry ice? hmm

  4. Of all the dumb things I could do to increase my risk while diving I think filling my own tank (or altering my tank mix by myself) has to be the dumbest.

    I’ll hack together plenty of stupid potentially dangerous things, but with scuba if you did something wrong and don’t discover it until you’re at depth… well… fatality isn’t just a potential anymore, it’s pretty likely.

    1. What as opposed to your refiller screwing up, in the article he uses viton rings in a specially made compression cylinder An electric pump swapout for a petrol one. They have accidents in professional dive shops in the end you are as safe as what you plan for. Filling outside so if things do go wrong the sunsequent catostrophic fire is contained, tank cooling and safety blast shields increase your chances for survival. I dont see its any more inherently dangerous than using a haskell boost pump. There are better design decisions that can be made but in the end you should plan for what if. reduce air hose length and use non flamable metal hoses. Ideally design hoses so if you use hydraulic fluid leaks are contained. Compressed oxygen is a dangerous substance but I think if you are aware of the risks and plan for them this activity can be performed as safely as can be done

  5. “the first dangerous aspect due to the lubrication oils and water vapor present in the output line. Also, being a nitrox certified diver myself as well as a chemical engineer, the air mix is crucial down to the percent… 32% and 34% are drastically different”

    i am no expert, and certainly have almost no experience with this, but it *appears* that he has compensated for all of the aforementioned issues. he has a “water trap” on the input line (i assume this reduces or eliminates the presence of water vapor?), he isn’t using lubrication in the piston (if you read toward the bottom of the page this becomes clear — he has a bit of trouble as the piston o-ring keeps disintegrating), and he is at least attempting to make sure he gets the right percent mixture by using a more precise gauge to read the amount of air he is adding to the tank.

  6. There’s some good books out that go into detail on project like this — “The Oxygen Hacker’s Companion” from Airspeed Press seems to be the canonical guide to doing boosters.

    Yes, the ultimate %O2 does matter, but that’s not a function of the booster — partial pressure blending is pretty common, where you put X PSI of oxygen into the tank, then top off to Y PSI with Air. The pupose of the booster is to get more O2 out of a bulk tank than you can by simple transfill/equalization.

  7. No offense…but if I going to be diving at 30+ feet of water I’m NOT going to trust my only breathing source to this hack. Underwater EVERYTHING can go wrong, and the margin of error is most likely fatal.

  8. by all means, nitrox certified diving is the way to go… less surface time between dives and I have noticed I don’t feel as drowsy after a day of dives… but I am going to reiterate the consensus: this isn’t the way to do it.
    Does he use an over-pressure relief-valve? I don’t see one… also, it looks like he’s filling his tanks without a water tub, so what’s his plan if a tank fails compression..?

    Has anyone heard any stories about what happens when a scuba tank fails? A quick quote:

    “The explosion was roughly equivalent to several sticks of dynamite. According to one scuba tank inspection expert, ‘The explosive potential in a fully charged 80cf aluminum SCUBA cylinder is approximately 1,300,000 foot pounds — enough to lift a typical fire department hook-and-ladder truck over 60 feet in the air!'”

  9. Mythbusters should do the scuba tank… they already did a hot water heater – at 350PSI… If I’m not mistaken it’s probably an exponential increase (i.e. richter scale) rather than a straight order of magnitude…

  10. Hack on, I’m tired of seeing people screaming about the risk, just because it’s homemade doesn’t mean it’s inherently unsafe, if you don’t have the skills, then by all means don’t attempt it, but looks like the guy has chosen a fairly accurate process here, and lets say that he doesn’t have access to the 12bucks per fill scuba shop that other do, or lets say that for some unknown reason, the 12 bucks per fill scuba shop no longer exist, this guy won’t be left out in the cold.
    If people were scared to do things for themselves, then progress in all aspects of the world would slow to and excruciatingly slow pace
    HACK ON!

  11. “If someone could make one that outputs HPA at 45000PSI, I would love them forever!

    (it’s for paintballing)”

    45,000 PSI, sure would hate to get hit with that paintball. Not that it would be intact at that pressure.

  12. to add to that, it really make ABSOLUTELY no difference how dirty your working air is, or how much water vapor there is as it is completely isolated from the booster. If anyone disagrees please explain to me how contaminated air can escape one piston, travel through a solid steel bar, and permeate solid brass to contaminate your NOX tank?
    And if you know the volume of you tank it’s simple mathematics to determine the EXACT amount or percentage you have of each gas only by measuring the presure change as the gas is being added. So, again, hack on, and kudos to the dude for coming up with a way to do something on his own.

  13. pedro the paintballer means 4500psi. an example of potentially lethal typos.

    @greg- that pressure is only for storage. it goes through a series of regulators before the air hits the ball. the final pressure is on the order of 100 psi, dependent on the marker and settings.

    if i ever find air at 45000 psi, i’ll strap a barrel on it and let you know…. hehehehe….

    I like this hack. never woulda thought 3000+ psi could be generated with a shop vac. the force multiplying lever idea is pretty slick. one more machine off my list of “how the hell does that work?” devices.

  14. I think anytime the words DIY and Scuba are beside each other you are bound to get a negative reaction. Simalar alarm bells go off when I hear DIY base jumping, DIY heart surgery, DIY fusion reacation and the like

  15. If you don’t understand how a booster works then by all means don’t even consider messing with any boosters (let alone a homemade design like mine). And if you don’t understand Partial Pressure blending and the risks of high pressure gases then it goes double…

    fentanyl3 has it right (thanks)

    “just because it’s homemade doesn’t mean it’s inherently unsafe, if you don’t have the skills, then by all means don’t attempt it”

    “to add to that, it really make ABSOLUTELY no difference how dirty your working air is, or how much water vapor there is as it is completely isolated from the booster. If anyone disagrees please explain to me how contaminated air can escape one piston, travel through a solid steel bar, and permeate solid brass to contaminate your NOX tank?”

    For the person who commented about Vance Harlows “Oxygen Hackers Cookbook” you might want to check out his “Improvised and Low Cost HP Boosters” Guess what you will see in there? Yep, my booster…

    As far as cost savings mixing recreational Nitrox there are none. This was a costly project which required a year of continual development. The reason to do it was my local dive shop went under and the closest place to get mix and deco bottles filled was 100-miles away. It has saved a lot of money topping off partially used Trimix fills. With a set of doubles filled with 16/45 (16%O2, 45%He, 39%N) running almost $100.00 plus two deco bottles 50/50 and 100%o2 running $30-$40each the booster has paid for itself. Please don’t send me hate mail about 100%O2. Contrary to what most recreational agencies teach there is a place in Technical diving for it and other Voodoo Gas mixes we use.

    In regards to the comment on the lack of an over pressure valve. If you do the math the shop air pressure multiplied by the ratios of the pistons and lever arm will only allow the supply gas to be boosted to a maximum of 3200psi (well within the range of all my tanks).

    This booster has served me well for the past six years. Boosting O2 and He into hundreds of tanks then topping with air from my compressor (with homemade hyper filter stacks). I’ve dove these mixes to over 265′ with deco obligations up to 100 minutes.

    Would I build this again? No! It was much harder than I ever thought and there is still room for improvement (especially in the spool valve arrangement). If I had to start over I’d bite the bullet and buy a Haskell but then again I’d have missed out on a lot of learning and a great feeling of accomplishment.

    There is nothing wrong with DIY scuba except you can’t sue the manufacture :-) Heck I know people building DIY Rebreathers.

    Play safe but always ask questions such as:

    How does that work?

    How could I build one of those?

    Brian B.

    PS – I don’t respond to flames but will answer questions as time permits.

  16. This is a great little mechanism for increasing force – I never would have thought a simple lever would be so effective.

    To all those thinking “This is dangerous!” – that’s fine, don’t do it.

    And to Pedro and anyone talking about Paintballing, this would be right on for filling an HPA or Nitrous tank @ about 3000 – 4000 PSI. I may have to try it out.

  17. Maybe DIY scuba isn’t the best example, but still, if getting the mixture right is that important then the only way to guarantee it is done right is to do what this guy is doing. It’s kind of the point of doing something yourself.

  18. As an enriched air diver, there are a couple of misconceptions on this board – first, every NITROX diver is trained to analyze the mix after filling with a gauge – so by the time this guy straps the tank on, he knows what he’s breathing. Additionally, filling an empty tank with a certain pressure of 100% O2 and topping it up with compressed air (from a certified, clean SCUBA compresser) is a pretty accepted method for mixing enriched air – I know professional divers who roll their own helium mixes this way. The advantage to the method shown here is that your tank does not need to be ‘oxygen clean’ – you can still use rubber vs viton o-rings, etc.

  19. @srilyk: Mythbusters Special 8: JAWS Special – They did several JAWS movie related myths, including shooting a scuba tank to make it explode. They didn’t make a tank “explode” persay (eg. shear open), but they did find great ways to make a tank decompress in uncontrollable manners.

    See also Episode 63: Air Cylinder Rocket where they made bulk gas cylinders fly through concrete block walls.

  20. joelanders, yes, thanks for correcting that. I added an extra 0 on the end, and did in fact mean 4.5k PSI.

    for anyone who’s interested, I use a 1.1 litre carbon fibre wrapped steel tank to store high pressure air at 4.5k PSI, which gets regulated down by my tank reg to 850 PSI. that then gets lowered to about 130 PSI by my inline regulator before going through a solenoid and bolt assembly.

    it costs me like £1 to get my tank filled or free for when I’m at a field playing, but it would be nice to have a compressor to have infinite fills while at home training

  21. “greg- that pressure is only for storage. it goes through a series of regulators before the air hits the ball. the final pressure is on the order of 100 psi, dependent on the marker and settings.”

    I know, I used to play, the tanks have regulators on them and a lot of guys would add a second one especially if they had a low pressure marker.

  22. All these Nancy woofters, Pooing desist. Due to various medical problems, I can’t get dive certified, so I have to free drive, buy rebreatehrs on ebay, and get ex military kit. So long as you are sensible, it is less risky than stopping in Miami for directions.

  23. Oh sweet merciful jesus this is dangerous. DO NOT BE DOING THIS UNLESS YOU’RE A CERTIFIED NITROX GAS BLENDER.

    This is a great way to kill yourself — even from something as simple as opening a gas relief valve too fast (Avoid Quarter Turn Knobs — read into adiabatic compression.) Even using the wrong material for the gas lines can cause sparking or an explosion.

    Editors should seriously think of pulling this down… if you’re not ticketed or licensed to do this, you can really screw up badly — blowing up your house, or at worst — getting yourself killed. And gibbed in the process.

  24. This project should not be used unless you REALLY know what you are doing. I not uncomfortable going to places that use something like this to boost o2. Professionally manufactured equipment with proper maintenance and with yearly gas testing is a different story. I would be willing to use this DIV project like this for non-volatile gases like helium for tri-mix or for argon for dry suit inflation, but o2+oil+heat is bad mojo.

  25. Chris… are you insane?!? Any tank or reg that is going to touch enriched air should be o2 cleaned per the manufacturer’s specifications at every service date (visualizes and hydros for tanks and rebuilds for regs.) Paying the extra $0.15 for the viton o-rings and $15 for the tank cleaning maybe over kill, but I’d rather do over kill then be killed. Take it from someone who has certs in full o2 deco diving and tri-mix blender, don’t do it. Home made fin straps, wrist slates, etc, are cool and not life support. Do you think the astronaut make their own gas pumps?

  26. Hello Owner.
    Am MrJerry Jones. and i will like to make some special inquiries on some Scuba Tank or any Scuba Tankyou have . Below are some of the types Scuba Tank i will need.

    Type Below
    XS SCUBA Neutral Aluminum 80cf Tanks
    Luxfer 40 cu. ft. scuba tanks
    Al 40 ft Tank.
    Al 80 ft Tank

    Above listed are some of the Size of the Scuba Tank i will like to order. I will like you to get back to me with the pick up price of those Sizes Tanks which will include tax only. and also if you don’t have this types of the Scuba Tank and the sizes Then i will Advice you to email me the types of the Scuba Tank you have along with the sizes and the Price Or you can email me the Website of your Company .Also i will like to know the forms of payment you accept for your orders. Kindly advise and get back to me. I will be awaiting your feedback on this email.

    Mr.MrJerry Jones


  27. Hi Brian Hows your pump holding up I was looking at a design of a waterjet pump the other day and noted they designed the compression cylinder so the rod had no sideways pressure on the bore. Linear rod shafts are cheap these days so some type of design using rods to reduce sideways foce on the piston may help reduce wear on the piston and the o rings. Of couurse you have built one im just talking about it. Anyway its fun to see your results. Im glad there are achievers like you around and not just naysayers and Hector the safety cat types who think anything technical is beyond anyone but experts.

  28. When I started Tech diving 15 years ago, Trimix, 50%, and 100% fills were difficult to come by (3.5 hr round trip to drop off, same to pick up), and so through research found vance harlow’s books Oxygen Hacker and Improvised Boosters. After some extensive study of these materials and from other sources, built a hydraulically powered oxygen clean booster. Worked OK, but wasn’t optimal. Took some parts and learned knowledge from first attempt and assembled an improved version, which has been pumping primarily 100%oxygen for 8 dive seasons, with no problems. Was a lot of work doing the custom machining of piston, figuring out how to safely manage all mechanical stresses and HP oxygen considerations, but has been an essential tool for my home mixing. A few years ago found a Haskel Booster at a reasonable price, but rarely use it, my homemade one is quieter and more energy efficient , although slower output. (The haskel is better for filling larger cylinders). The coolest part was figuring out how to get it to cycle automatically….just two relays needed in a “hold circuit”). Fair warning, anyone messing with high pressure gases, especially high % of oxygen, should read Harlows books and do some additional research before proceeding. And some engineering savvy/mechanical aptitude wouldn’t hurt……

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.