Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Aerospace Dyson Bladeless Fan, Inducted Airflow Propulsion, Personal Transportation

  1. May 1, 2010 #1
    I remember once observing a thread in a discussion forum by someone trying to build a backpack helicopter. Various people warned him about sprung-weight-vs-unsprung-weight, hazards of stability and flight control, etc.

    But ever since reading about James Dyson's bladeless fan and the well-known Coanda Effect, I've been wondering whether it could be adapted for propulsion purposes, into some kind of personal transportation device.



    I'd like to know - are there any inherent limits on using Coanda Effect and this bladeless fan idea for propulsion?

    Coaxial helicopters in particular have a smaller footprint because the counter-rotating blades eliminates the need for a tail-rotor, and of course the dual blades generate more thrust within that disc space.

    Would it be possible to make the bladeless fan ring work the same way as coaxial contra-rotating rotors, by linking 2 of them together somehow?

    The impeller inside the Dyson bladeless fan seems to present low sprung weight, and the airflow ultimately produced is smooth, as opposed to the choppier airflow of the helicopter's blades. Presumably this would add to flight stability. Crucially, the lack of external rotating blades would make it safer than any hypothetical helicopter backpack. Dyson's fan is also supposed to be fairly quiet.

    But what are the drawbacks to this impeller-with-coanda design? Does it trade off power-to-weight efficiency enough to make it infeasible for propulsion purposes?
    Last edited by a moderator: Sep 25, 2014
  2. jcsd
  3. May 1, 2010 #2
    The answer is no. If you look at the equations of momentum theory, you will find that the most efficient system is one in which you have large slowly rotating blades.

    As for the coaxial rotors, they also suffer from wake interaction effects, thereby reducing the system efficiency as being much worse than if you simply had one larger rotor of an equivalent disk area.

    Ultimately, Dysons 'Coanda' blades are garbage for any type of serious propulsion systems. There was tons and tons of research into VTOL in the 60s (almost anything under the sun). Quite simply, you won't find a better way to VTOL than a helicopter, and the equations back that up.
  4. May 1, 2010 #3


    User Avatar

    Staff: Mentor

    You can prove this to yourself by looking at/playing with the kinetic energy and momentum equations. The lifting force is a function of the momentum change in the air, but the size of the engine and the fuel used is a function of the velocity change of the air. So since kinetic energy is a square function of velocity, but momentum is a linear function of velocity, what you want is as low a velocity as possible and as high a mass flow rate: ie, large, slow spinning blades.
  5. May 1, 2010 #4
    So you're telling me that this is the most efficient flying machine possible?

    Last edited by a moderator: Sep 25, 2014
  6. May 1, 2010 #5
    Here are some UAVs that seem to use the Coanda Effect

    Last edited by a moderator: Sep 25, 2014
  7. May 1, 2010 #6
    No one said that. :confused: We simply said for hovering flight large slow rotor blades are the way to go.
    Last edited by a moderator: Sep 25, 2014
  8. May 1, 2010 #7


    User Avatar

    Staff: Mentor

    No. What does that have to do with anything?
    All aircraft use the Coanda effect. Those are a clever use of the Coanda effect for control (which is kinda cool)....but not using the Coanda effect to generate lift (except, of course, in the fan itself).
    Last edited by a moderator: Sep 25, 2014
  9. May 1, 2010 #8
    In the second video, it can bee seen that the skirt completely blocks the downwash from the fan. Should the airstream, hitting the skirt, not be redirected with a downward component due to the so-called Coanda effect, the airsteam would continue horizontally, generating no lift.

    I'm unsure whether or not the fan alone would generate the same lift for the same power input with the skirt removed.
    Last edited: May 1, 2010
  10. May 1, 2010 #9
    Yes, this is my point exactly. That fan is on top of that bell shape, and not below it. The thrust is moving around the bell and on downwards, which is Coanda Effect. This is what's causing the thing to lift off the ground. There is no hole in that bell-shaped fuselage. The thrust is going around it.

    So I'm wondering what are the limits imposed by the Coanda Effect on lifting force?
    How much thrust can be generated, before this thrust-redirection effect no longer works?
  11. May 1, 2010 #10


    User Avatar

    Staff: Mentor

    Thurst redirection is what it looks like to me. That implies that it could never generate as much lift as an unobstructed fan.
  12. May 1, 2010 #11
    Yes, I understand that, but at the same time there may be other benefits to a setup like this, in terms of safety, avoiding blade strikes, etc. This Coanda Effect thing seems to permit more flexibility in design/geometry, even if there is that tradeoff of reduced thrust efficiency.

    For a smaller personal transportation system, it's possible that such tradeoffs might be worth making, if you're not looking to fly thousands of feet in the air, or for very prolonged periods of time.
  13. May 1, 2010 #12
    No, there isnt. If you are really worried about operational safety, you might as well just use a shrouded rotor, as it is much better than a "Coanda" Dyson fan - whatever that means.

    I should add: the "Coanda Effect" is nothing more than a phenomenological description of the fact that a fluid follows a curved patch due to shear stresses (provided there is enough energy in the boundary layer that separation does not occur). There is no equation behind the Coanda effect that makes it of any practical value because its nothing more than an observation.

    I would add that these "Coanda" vehicles were researched as far back as the 60s and went no where - that should tell you something about their practicality as a system.

    None the less I think his house fan is very cool. (Ha, pun inadvertently intended!)
    Last edited: May 1, 2010
  14. May 1, 2010 #13
    But the reason Dyson's fan is liked, is because it offers more safety than a shrouded fan. It also apparently acts as an airflow amplifier through entrainment of air into the flow.

    Note that the Coanda Effect UAV shown in the videos above doesn't require any tail rotor to counteract torque. This might make it useful for a small personal air transportation device.
  15. May 2, 2010 #14
    This is not as big an issue as you are making it out to be. A shrouded rotor is more than sufficient for a micro-air vehicle in a cluttered urban environment.

    Neither does a shrouded rotor, they use deflecting vanes to provide anti-torque. Again, this "Dyson Fan" concept is not - at all - useful for small personal air transport. It makes for nice house fans, and thats about it. Why? Because you have a small high speed mixed flow impeller that provides the thrust. Remember, you want large slowly rotating blades on hovering vehicles - this smacks the face at that notion.
  16. May 2, 2010 #15
    If I've applied F=dp/dt correctly, the force from a fan disk generating uniform flow is

    F= ρAv2

    v is the exit velocity,
    ρ is air density and
    A is the effective disk area.
    F=W, the weight of the craft in a hover.

    This is what Russ and Cyrus are talking about in terms of disk area. Doubling the diameter of the disk means the exit fan velocity can be cut in half.

    The power of air in motion delivered by the fan, assuming a uniform crosssectional flow, is

    P = ½ρAv3.

    From the two equations given,

    P = ½√(W3/ρA)

    Decreasing the area, A of the fan, disk, or whatever element is generating the airflow, requires more power.

    Now, I'm interested in knowing if there is any power advantage using the thrust of the fan along compared to rerouting it over the surface of a "Condola effect" skirt.

    For simplicity, I initially assume the power available is constant, and the skirt contributes negligibly to the weight and added frictional and turbulent loss, air is incompressible, and other mild assumptions. Assume the same power and fan area.

    For the fan alone, F= ρAv2 directly out of the fan.

    For the fan + skirt arrangement, assuming the air steam can be redirected vertically downward, If A remains constant there is no change in force.

    I don't know if the flow passing over the skirt will entrain stagnant air, but the effect is detrimental. Say the area of flow is doubled due to entrained air. To conserve momentum with an equal volume of stagnant air the velocity is cut in half. Overall, the lifting force would be cut in half.
    Last edited: May 2, 2010
  17. May 2, 2010 #16


    User Avatar

    Staff: Mentor

    It's an induction effect and it takes high pressure air squirted through jets to entrain low pressure air around it. The net result is a larger volume of low pressure air.

    This induction isn't free: the high pressure of the primary air means that it requires a larger horsepower fan than more air of lower pressure. I suspect that the theoretical best you could do with induction vs just a larger lower pressure fan would be an exactly equal energy consumption.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook