3D Robotics

NASA simulates Phantom airflow

From NASA:

For decades, NASA has used computer models to simulate the flow of air around aircraft in order to test designs and improve the performance of next-generation vehicles.

At NASA’s Ames Research Center in California’s Silicon Valley, researchers recently used this technique to explore the aerodynamics of a popular example of a small, battery-powered drone, a modified DJI Phantom 3 quadcopter.

The Phantom relies on four whirring rotors to generate enough thrust to lift it and any payload it’s carrying off the ground. Simulations revealed the complex motions of air due to interactions between the vehicle’s rotors and X-shaped frame during flight.

As an experiment, researchers added four more rotors to the vehicle to study the effect on the quadcopter’s performance. This configuration produced a nearly twofold increase in the amount of thrust.

The findings offer new insights into the design of autonomous, heavy-lift, multirotor vehicles for uses such as cargo transportation.

This research was presented at the 2017 American Institute of Aeronautics and Astronautics SciTech Forum in Grapevine, Texas, by Seokkwan Yoon of the NASA Advanced Supercomputing Division at Ames.

E-mail me when people leave their comments –

You need to be a member of diydrones to add comments!

Join diydrones


  • Rob, the incoming air above the rotor is being drawn in from a larger area at a lower velocity than the air leaving the propeller - the streamtube contracts below the rotor so that you have constant mass flow but increased velocity.

    The net result is that anything above the rotor will see a lower velocity than anything below the rotor.  See the image below courtesy of wikipedia - airflow is coming from the right, through the disc and exiting to the left.  


    I guess the interesting question is the comparison of larger wetted area for structure above, versus higher velocity for structure below.

    If you can solve the layout problems for landing gear and ground clearance, then rotors below the arms may give you better efficiency and also lower noise (since you don't have high energy vortices hitting the arms.  It all, however, may go to custard as you translate to forward flight and the fuselage starts shadowing the aft pair of rotors.

    Interestingly, I think Prouty also proposed that having your c.g. above your rotor disc is a more stable configuration.  It worked for the Hiller flying platform:


    Also, the Martin Jetpack fairly rapidly switched from their original "high duct" configuration to having the ducts at the pilot's feet.

  • Oh, I never bothered with that.  A small copter can flip over fast enough it didn't matter.

    But, a little state machine could be done to make it climb first.

  • Developer

    I would think the transition is the problematic part. Normally to make the flip you punch the throttle, then go into neutral pitch while flipping over to avoid driving the copter sideways and into the ground.

  • No, inverted mode never made it to Master.  But it's trivial to do so I could test it again.  All you need is a mode where the Roll target is moved 180 degrees, and the code does the rest.  It's like a 2 line change.

  • Developer

    Does APM still have the inverted flight mode? In that case it would be an interesting test to hover a heli both ways and see if there was any significant difference.

    Also, looking at the airflow disturbance in the video there is a lot going on around the motor and propeller hub. I remember one of the sigh.. Local Motors entries had a motor housing covered in golf ball dimples. Maybe he was on to something..

  • Gary, the incoming air isn't zero velocity however.  The propeller is forming a column of inflowing air.  It's essentially feeding itself.

    Yeah, the relative size of the fuselage compared to the disk area, is small.  Also, most of the body is in the low-velocity area of the disk (center).  Tail boom has full exposure, but it's pretty slender.

    And I wouldn't call myself a 3D pilot.  Not at all.  I prefer yank and bank high speed heli flight. ;)

  • Developer

    It is a fascinating visualisation. I would love to see a plane version!

    @Gary MCray, so according to that a pusher prop is more efficient than a tractor prop on a plane?

  • Hi Rob,

    Because it is unstructured zero velocity air and not a high velocity down wash, disturbance of the inflow produces a lot less loss than an equal disturbance the out flow.

    Among other things any impingement of the out flow directly results in whatever thrust is absorbed by the obstruction being a net total loss.

    It is pushing itself back down with the force it is supposed to be using to push itself up.

    In a ducted fan system disturbance of inflow and outflow are more similar, but for an unshrouded propeller a minor interference with inflow is much less significant than an equal interference with outflow (thrust in this case).

    This is especially true at zero vehicle velocity - hover for instance.

    EG why Paul Pounds selected supports over the propeller in the first place.

    Harder to build an inverted heli though.

    Although it is certainly possible to fly one upside down which seemingly could form a good evaluation test bed for this phenomenon.

    Be interesting to look at power consumption in a stable hover right side up and upside down, so long as over-control issues didn't become dominant.

    Horizontal velocity would confuse everything.

    And to be a true test it would probably require rotor blades with a symmetrical airfoil.

    I'd really like to know the results of that. - Hint - Oh great and accomplished 3D heli pilot!

    I'll go out on a limb and predict a 5 to 15% improvement upside down if the control issue isn't significant.

    Obviously the improvement would be smaller for helis with bigger rotor diameter versus heli fuselage size.

    So aside from the fact that I doubt you want to do this with one of your Procyons with it's giant rotor, a smaller relative rotor size would demonstrate it better in any case. 



  • It's interesting that the complicated model predicts the streamtube contraction, and the less-than-double-thrust-increase from doubling the rotors, as the very simple Momentum Theory model predicts.

  • And an octocopter is even worse.  Was really surprised when I did some speed testing last week with my Tarot X8.  It hovers on 35A unloaded.  And was 45A at 10 m/s.  But pushing it full forward, it was achieving only 16 m/s on a whopping 80A.

    Compared to a helicopter which uses less power to fly at 15 m/s than it does in a hover.

    Now Gary, the thing about the inverted propellers, is that while there may be less resistance from the high velocity downwash blowing on the arms, you also have restricted the inlet to the propeller.  The thrust generated is all about mass airflow.  And by having the arm above the propeller, you are placing a restriction on the inflow.

    There is no free lunch.

This reply was deleted.