Aquaponics - New air lift design

If you pump air down into a submerged tube, when the bubbles rise to the surface, by virtue of the fact that they take up some space, they change the average density of the water in the tube.

The result is that the new, lower density water floats, and you get water sticking up above the surrounding water.

In short, you get a pump.

This is nothing new, and even if all you do is stick some bubbles at the bottom of an aquarium and let them rise to the surface, you are seeing what has become to be known as an air lift.

Through the introduction of air into water, the bubbles cause a vertical current.

If you stick those bubbles in a tube, you create something of a pump.

An air lift in an aquarium is mainly used to create current and keep the top surface of the water moving. This top surface is exposed to the oxygen in the air, and the exposed water gains dissolved oxygen, or DO,  essential for the ongoing good times, and life, of the fish in your aquarium, or aquaponics system.

I don't have an air pump, because I run so much water through my pump, that it's not necessary. I do run a powerhead, which is a lot like a little outboard motor, that achieves the same thing, and in my case is totally unnecessary because, as mentioned, I pump so much water with my over sized pump.

An air lift, and a powerhead both move large amounts of water for a small amount of energy. My 5000L pump was 150 watts, my 5000L powerhead is only 13 watts.

The difference is the pump could lift water 5m, and the powerhead cant really lift water at all. All it can do is stir it.

An air lift can lift water, but it's a bit limited in how high it can lift it. It would be nice if an air lift could lift air to a greater height, because then we could create a really low energy aquaponics system in a conventional (grow bed above fish tank) arrangement, and still get the desired flow.

To this end, I drew this.

Normally a picture might paint a thousand words, but this is a picture of particular poor quality, so I'll need to add a few to make it paint up to around 150.

Picture a pipe with some chambers, all sealed and air tight, so whatever pressure is in there, stays in there.

If the air lift were to deliver it's water to a chamber, the water and air would separate.

If the air was then bled off from the chamber, and re-introduced under the collected, chambered water, it might be reused to create a second or third (etc) air lift.

This might mean we could lift air higher than with a normal air lift.

I have no idea if it would work, because I haven't been able to rope anyone I know who runs air in their system into trying it yet.

If I cant get anyone to test the idea, I'll have to buy an air pump.

Either way, I'll let you know if it turns out to be interesting.

120 things in 20 years, bringing you the vaguest suggestion of science in the hope that someone else will do the actual tests and see if my aquaponic, new air lift design actually works.


  1. I like the concept. I'd want to explode a pump so the electrical is on the outside of your lift-station chamber and the valving is on the inside.

    Thinking about it... you may not even need more pumps. The air would get high enough to start to bubble up the 2nd lift-station all on it's own.

    I may need to build one now...

  2. I'm not sure that I have any idea what you just said, but I have to say, I'm pretty excited!

    Can you just say all that again differently :)

    It's 7am here and I haven't slept yet, but it sounds like you plan on building something to test, so I'm all for it :)

  3. ive been thinking about this a lot (from BYAP) and i cant see it working.... tho im not exactly sure why.

    i think the thing you are not taking into consideration is you need several feet of depth in the fish tank to get a positive head height at the surface. you cant get a 1:1 ratio of "FT depth to positive head height" you would need for this to work.

    basically put, you need time for the water to catch up speed with the rising air.

    an easy way to prove this (and me wrong, and id love to be wrong and see this work) would be to get a peice of pvc and stick an air stone in from the top and raise and lower it in the pipe and see how much depth changes what head hight you can get.

  4. Mike, the plan was that would only be one air pump, and the air pressure would build up in each cavity and be reused, by being piped to the next stage.

    1. Then the air stones drawn up the upper segments confused me. The air stones in the upper segments would have no affect.

      Try this thought experiment, upside down bell siphon, for air.

      I am expecting a 3d printer by the end of the month. Mind if I make one of these to test?

    2. I'm not sure I understand what you are saying here.

      The air tones in the upper segments are powered by the air pressure created by the water sealing the exit (via the top) and supplied by the air trapped in the chamber. The same air from the pump gets trapped in the chamber, it's trapped at whatever pressure the pump puts out.

      Otherwise it wouldn't be trapped, and would wonder off.

      I think :)

    3. The water level will be the same in the air-stone line as it is in the upper pocket. Air won't try to go up the tube until it reaches the bottom of it. (that's a lousy sentence, isn't it?)

      How about a video? I like videos. Take a look at and see if I explained my thoughts clearly enough (probably not)

  5. Dan, I see the problem you have with it, and I cant honestly say I have a clue as to it working or not. For one thing I dont have an air pump :)

    But the pressure in each chamber might stop the water flowing back, so any gain might be kept until you do have the required head in each stage.

    Another option would be a return valve on the bottom, so you could prime the entire thing before you started, by filling it with water.

    In the event of a power outage, perhaps it could remain primed because of the valve.

    It might also be interesting to see what happens if you ... I'll post about it This needs a diagram.

  6. The more I am rolling this around in my head, the more I think Dan is right.
    However, I think your design basically takes account for this as when the lift station fills with air and water, the water will be displaced up the lift tube.

    I think the thing may actually JUST WORK.

  7. It would work but the gained lift per stage would be small and the whole process would be unnecessarily complicated. Air lift pumps work at ratio around 2 to one to 3 to one lift to submergence depth ratios. So to lift 1 foot you need between 2 to 3 feet of submergence below the water surface.

    Now how are we going to lit a height of above the depth of aquarium? Well I think we could make a well (or sump) of 3 to 4 inch PVC capped at one end. this would give the necessary submergence and the pump should work as needed. I think one long sump would be easier and more efficient than 3 smaller ones even if we had to bore a hole in the ground to put the sump in.

    Oh and air lift pumps handle suspended matter way better than any other type of pumps.

    Nice biog keep it up.

  8. I suspect after many sleeps, that the back pressure (against the air pump) would be the same as if it were just submerged deeper. eg. in an external PVC "sump".

    The hope was there might be some gain over a conventional airlift, but I I think the entire head of the device would be adding all of its pressure to counter the air pumps effort.

    But then I keep thinking that the air as it enters the first lift is under only the pressure from the head of the tank rather than the head of the entire apparatus.

    I like to tell myself I'm good at working this stuff out in my head, but for example, I have no idea if a submarine feels more pressure when a wave goes over it or not.

    That's been troubling me for nearly 20 years :)

  9. I can tell you that a wave flowing over a submarine has negligible (a cat wouldnʻt even notice: meaning no) effect on pressure. The only thing that does effect pressure is depth and concentrated water pressure around the vessel. SSNB-642 USS Kamehameha. North Atlantic. Based out of Holy Loch, Scotland. The boat is now long retired.

  10. Thanks for the input Anonymous. Are you a submariner? I really need a submariner in my E-address book, so I can siphon off your brain :)

  11. Neat concept, I've been trying to find numbers relating the depth of the pipe to possible pressure head in a way that Dan was suggesting.

    An airlift doesn't work by decreasing the density of the water. As the bubbles rise, the water needs to move around the bubbles, but cannot fall as fast as the bubbles rise, which means a net velocity upwards. The reason an airlift gains pressure head is by allowing the water to match the velocity of the bubbles, which takes a longer tube. You gain the pressure head by converting the kinetic energy into potential energy.

    If you wanted to derive some empirical formulas you could video tape the bubbles to measure their velocity and match it to the pressure head gained. Of course others have worked on this, maybe you could find some things looking at 2-phase flow. That being said, I tend to like just making things and observing how well they work.

    It seams to me that creating multiple chambers wastes the work the bubbles gained by accelerating the water upwards by stopping, then accelerating upwards, then stopping.

    1. This comment has been removed by the author.

    2. Thanks for the feedback p.squared.

      re the bit about density - It would be easy enough to test. If someone with an air pump and a tube could tun on the pump, and see if the water depth increases before the air reaches the top, or only after.

      Maybe :)

      I think the velocity of the water is too low to gain the extra head you get, but I'm always pleased to be wrong.

      People claim more head with deeper tubes. But this could be read as more momentum or as more expansion of a fluid in a semi-closed container, with a bias toward up, as far as current goes.

      I'm revising what I think to this...

      -Air travelling up the tube creates a current from bottom to top.

      -The air water mix decreases the average density as compared to water alone. That probably does something. I'm guessing that it makes the fluid in the tube float.

      -As the fluid in the tube expands by the addition of air, it overflows it's containing tube, but due to the current created by the rising bubbles, it overflows mainly over the top rim of the tube.

      -A small amount of additional head is gained by the waters upward momentum.

      -The lower density air/water mix floats as compared to the rest of the tank.


  12. This comment has been removed by the author.

  13. I made 2 types of airlift pumps for windowfarms. One is called T joint airlift pumps and another is called airlift in a bucket. I have got air to pump water to 6 ft high using 4 inches of pressure. (That is about 0.15 psi pumping to about 2.5 psi high. I also used 1 psi to pump 13 ft high. That is because airlift in a 1/4 inch diameter tube works by plug flow.

    1. Do you have a link to your work?

      I'd be interested to see what you're doing.

      -120thingsin20years typing from out of town

    2. Hi,sorry, I missed your reply. there are lots of links, you can google search for them if you like. You can find some in my playlists

    3. What kind of flow rates do you get?

      From the videos I saw, it looks like it's running drip irrigation.


Popular Posts