I recently came up with a conceptual idea for creating aerodynamic lift which would use an unconventional centrifugal impeller. I would like to know if this conceptual design of mine is something that has already been built and tested in the past.

I have emailed this design over to some aerospace historical societies and aerospace institutes inquiring if they know if this particular design, or something very similar to it, has already been built and tested by an aerospace company in the past. They replied back saying they had not seen this particular design. So, I thought it would be worth asking the Aviation community on Stack Exchange if anyone has already seen this particular design.

I have created some CAD drawings (shown below) to illustrate what this mechanical apparatus would probably look like. Its primary components would be a centrifugal impeller with two rows of blades, a motor to turn the impeller, and a curved panel which would be fastened to the fuselage or to a wing of an aircraft. I want to point out that the curved panel would not rotate along with the impeller, rather the centrifugal impeller would rotate around the curved panel.

This mechanical apparatus would have a simple working principle. The centrifugal impeller would have an inner row of blades and an outer row of blades. The inner row of blades should create high air pressure along the bottom surface of the curved panel and the outer row of blades should create low air pressure over the top surface of the curved panel. This should result in aerodynamic lift.


I am thinking now that it may be better that the outer row of blades be backward-curved in order to reduce turbulence and to maximize the static air pressure drop along the top surface of the curved panel.

enter image description here

enter image description here

enter image description here

Would this mechanical apparatus be a new means of creating aerodynamic lift?

  • 5
    $\begingroup$ And what would be a supposed advantage compared to, say, propeller oriented vertically, turbofan or any other existing means of creating aerodynamic force taking energy from rotating shaft? Usually one is talking about lift in context of (static) wings converting motion of air into perpendicular aerodynamic force. If you need an engine to generate "lift" the resulting force is usually called "thrust". $\endgroup$
    – Martin
    May 15, 2021 at 19:05
  • 7
    $\begingroup$ What you have there is a birdcage fan that can only blow out the bottom half due to the obstruction of the top half. So to the extent you are getting a thrust reaction from making the exhaust go down, you'd get some lift, but it wouldn't be very efficient. $\endgroup$
    – John K
    May 15, 2021 at 19:24
  • 5
    $\begingroup$ This moves a little bit of air a lot. You'd get much more lift by moving a lot of air a little, i.e., by attaching the same motor to an appropriately sized propeller. $\endgroup$ May 15, 2021 at 19:32
  • 3
    $\begingroup$ Not to burst your bubble, but that is just a fan. I replaced one just like it on my furnace a few months ago, and just last week I removed another similar one from my shop vac motor in the process of replacing a fusible link. Not suitable for making an airplane fly, but many practical uses! $\endgroup$ May 15, 2021 at 19:47
  • 3
    $\begingroup$ You've gone through a lot of effort to reinvent the fan used to cool most modern GPU cards. This is not a useful source of lift for heavier-than-air flight. The figure of merit there is useful lift per unit energy spent; simply producing a trivial amount of lift is not enough. $\endgroup$ May 15, 2021 at 23:30

4 Answers 4


Yes, this device would produce some thrust. If oriented properly, this thrust could be used to lift a weight. But it would be extremely inefficient when compared to a standard airfoil, such as a wing or helicopter blade.

What you've basically created is a centrifugal pump without a case. Air would enter the "front" of the device (assuming the motor is the "back") just like a regular centrifugal pump, but without a case, it would be sprayed in all directions, including at an angle out of the "front". The curved panel would partially block the radial flow of air upward, creating a net upward force. But the fact that air exits at all angles, not just straight down, means that the various flows would partially cancel each other. What's worse, a lot of the air mass that emerges from the "front" would be drawn back into the incoming air stream, which would consume some of the motor's power but wouldn't contribute to the net thrust at all.

In short, I would be very surprised if this device could match even 1% of the performance of a simple ducted fan of the same power rating.

It's kind of hard to explain what I mean in words, so, here, have some bad, hastily-created images instead:

Side view of fan, with arrows showing the flow of air, and snarky captions.

Front of the fan, with arrows leading away from the center, and captions detailing exactly how much power is wasted for each angle.

  • $\begingroup$ You could add that the pressure differential pointed out in the original drawings has no effect on lift at all, as it seems to me that this is the "invention", actually. That's the only reason one would place the half-circle-thing in the middle of the fan blades instead of on the outside. $\endgroup$
    – Haukinger
    May 16, 2021 at 10:02
  • $\begingroup$ @Hidden Windshield, the way I look at this is from a static pressure perspective. I am thinking that if the top surface of the curved panel experiences say a pressure drop of -3.0 psia and the bottom surface of the panel experiences a pressure increase of +3.0 psia, then would mean a +6 lbs per square inch of lift force. So, for example, if there is a surface area of say 20 square inches, then this would result in 120 lbs of lift force. $\endgroup$
    – user57467
    May 16, 2021 at 14:18
  • $\begingroup$ @Haukinger (1/2) Actually, there would be a (slight) pressure differential across the curved plate. That's how accelerated air is able to move a solid object. The process of accelerating a mass of air generates pressure, which pushes on the object that's doing the accelerating, thus balancing Newton's third law. If you analyze the downwash from a wing, you'll find that the amount of force needed to push the air down is exactly the same as needed to push the wing up. So, one can analyze a wing in terms of "high pressure on bottom, low pressure on top, pushes the wing up". $\endgroup$ May 16, 2021 at 14:26
  • $\begingroup$ (2/2) Or, one could say "accelerating air downward results in an equal and opposite force pushing the wing up." They're two different ways of solving the same problem. The same is true here, I just skipped the intermediate role of pressure because I thought working in terms of thrust would be easier to understand. $\endgroup$ May 16, 2021 at 14:26
  • 1
    $\begingroup$ @user57467 (1/2) Since you don't have a case around the device, turbulence and recirculation are going to cancel out a large portion of the pressure difference you would otherwise get. So, you'd need a very powerful (read: large and heavy) motor spinning very fast (as in, I'd be worried about the disk flying apart under the strain) to generate such a huge pressure difference. And even then, it's not going to generate 120 lbs of pressure up, but 120 lbs of pressure out. The horizontal pressure differential would basically be 100% wasted, and the differential that's across an angle would be $\endgroup$ May 16, 2021 at 14:36

What you appear at first sight to have is a hybrid between the cross-flow fan and the paddle wheel, with a dash of the centrifugal fan thrown in. I have never seen it described, for example it is not discussed by Foshag and Boehler. I suppose that technically I can only give my opinion, since if nobody has written about it then one cannot declare a factual judgement. The device appears at best highly inefficient, I'll try to explain why.

First of all, recall the fundamental principle of lift; any heavier-than-air craft must accelerate air downwards, to create a reaction against that acceleration. That reaction is the lift.

The centrifugal fan throws air outward, typically collecting it in a duct and pointing it downwards. Accelerating that airflow down creates an upward reaction against the duct, i.e. lift. But I have never seen it used on an aircraft (in an engine it acts as a compressor).

The cross-flow fan uses subtle shaping to entrain air from one side and blow it out the other. Again, it is reaction against the acceleration of the air that provides lift. It has been known since Victorian times. It is a common form of fan in low-profile devices, such as small hot-air room heaters. It has been adapted as an aircraft wing relatively recently; checkout the FanWing.

The paddle wheel is the least efficient. It cannot accelerate the airflow beyond its own tangential velocity, which limits its ability to generate lift. Pressure differentials obtained by mechanical means are far less than those obtained aerodynamically for the same amount of work. This is why the centrifugal fan is used to accelerate the air along a duct or similar surface, not just to exert a pressure.

So your device will have low efficiency as a lifting device and once it starts to move even slowly, crossflow effects will quickly dominate. Remember too the fundamental principle of lift; if there is no device directing the air downwards, there will be no downward acceleration for the fan to react against and produce lift. You have identified no such device. What I would suggest you have come up with is an inefficient crossflow fan, in which your curved panel acts as a poor man's half-duct (and compared to the flyable FanWing is upside down, which is worrying).

You could put a shroud around it to eliminate the crossflow effects, with a downward exit duct and a central intake duct, and then it would just be a crude and inefficient centrifugal fan.

These issues might explain why nobody else has bothered to describe it before now.

  • $\begingroup$ @Guy Inchbald, I am familiar with the FanWing and I am very impressed with it. Its design is similar to mine, yet the FanWing is a STOL aircraft while my mechanical apparatus is intended to be used on a VTOL platform such as a UAV/drone. $\endgroup$
    – user57467
    May 19, 2021 at 14:28
  • $\begingroup$ Ah, okay, if the distinction is important to you I'll amend my answer. Just need to look up a couple of NASA references first. $\endgroup$ May 19, 2021 at 15:32
  • 2
    $\begingroup$ Much detail now added, hope that's an improvement. $\endgroup$ May 19, 2021 at 16:26
  • $\begingroup$ @ Guy Inchbald, thank you for providing that very in-depth analysis of my mechanical apparatus and also for providing the link to the document made by Foshag and Boehler. I briefly looked over it and saw some designs that I never saw before. I am going to read through that document over the next few days. $\endgroup$
    – user57467
    May 20, 2021 at 0:33

No, not new, as pressure differential is the key to heavier than air flight.

In air, your device will also pull itself forward, so you now have the 2 necessary components of flight, thrust and lift. Theoretically, it works, but ...

Things that roll on the ground such as trucks and locomotives can be a bit heavy, but with aircraft, weight savings is essential.

Thrust and lift per kilogram vs a prop/wing or even a helicopter...

Sorry, not even close. For successful applications, one must either find a new niche, or be better than the competition at an existing one.

The closest the "mechanical apparatus" comes to being aviation related can be found as a centrifugal compressor in early jets. One could conceivably put in a fuel combustion chamber in the part labeled "motor" and with vectored thrust we have the "jump jet". +1 for thinking (and the diagrams).


It seems that your proposed device would have something in common with the following alternative proposed device--

Create a helicopter with coaxial counter-rotating rotors, widely vertically separated.

Angle the rotor blades so that they force air upwards, not downwards.

Mount a stationary solid disk halfway between the two rotors.

The intention is that the lower rotor will create high pressure on the bottom surface of the stationary disk, and the upper rotor will create low pressure on the top surface of the stationary solid disk.

Possibly enclose the entire contraption in a vertical cylindrical shroud.

Maybe the shroud should have some ports around the circumference to prevent the air from stagnating. Just above and just below the disk would appear to be the optimal location for these ports.

Would such a thing create some lift? Probably-- but in the downward direction.

Could a more optimal device be created from the same raw materials? Certainly.

The basic point here is that you have to consider the entire system-- you can't just home in on how a pressure difference may be created between two sides of one particular element.

  • $\begingroup$ Hmm, the analogy seemed to fall apart as I was typing the answer. I'll leave it up for a while though. $\endgroup$ May 19, 2021 at 19:11
  • $\begingroup$ I see what you're saying. I decided to post this question just to see if someone had come up with this same idea perhaps a long time ago and to find out how much lift force that device had created. $\endgroup$
    – user57467
    May 20, 2021 at 13:46

Not the answer you're looking for? Browse other questions tagged .