Designing a VTOL Tiltrotor UAV


A big hello to all the intelligent lifeforms everywhere and to everyone else out there, the secret is to bang the rocks together.

I just want to update you all on my current project work. The design started by deciding the performance objectives for the UAV which are as follows:

  1. 30 min cruise endurance + 10 min VTOL + 5 min take off landing
  2. Around 15 m/s cruise speed and a flying altitude of 150 m
  3. An operational radius of 30 km
  4. A payload capacity of >2 kg which will include thermal/IR cameras etc.

The UAV configuration will comprise of 2 tilt-rotors which will be active during all of the flight phases and also one main rotor which will be just active during the VTOL and hovering phase. In the cruise mode, tiltrotors will be turned to the flight direction using simple servo motors and the power of the coaxial central motor will be boosted to provide necessary lift. Once the transition is completed, the central motor will be slowly turned off so as to maintain constant altitude and the duct will be closed to reduce drag.


The above image is the result of trade studies that were performed to estimate thrust and weight loading for the UAV.


After running initial weight estimation algorithm and performing the trade studies, T/W and W/S were decided along with the thrust distributions for the rotors. The image shows the weight estimation algorithm that was used along with trade studies for determining thrust and wing loading.


After preparing concept sketches for the UAV, blended wing configuration was selected by keeping in mind the portability and efficiency. Flying wing configuration has a minimal wetted surface area which reduces profile drag significantly. The absence of stabilizing surfaces induces higher maneuverability, which is essential to the UAV. Use of commercially available autopilots will greatly ameliorate the dearth of stability in flight dynamics. Once this was completed we moved to selecting airfoils for the flying wing VTOL tilt-rotor. Different reflex airfoils were analyzed in PROFILI so as to enhance the maximum lift coefficient while limiting the maximum drag. In addition to the flying wing design, blended winglets will be used to enhance the aerodynamic efficiency. Winglets will house the rudder controls for the UAV. 


This image shows the UAV in VTOL mode with the central open hub which we plan to close in cruise flight. Eppler 334 was selected for the central plank section, NACA34112 for the wing plan-form and NACA0012 for the winglet. The next step was to perform preliminary iterations in a basic software like XFLR5 so as to get some results for number crunching.


The above image is the result of aforementioned XFLR5 analysis showing the air stream flow around the UAV at trimmed flight of 4 degrees along with the pressure contours. A CL/CD of around 20.5 is achieved which proves the initial guess of superior flying wing efficiency as compared to conventional configurations.


This image shows the side view of the mesh that was generated for CFD analysis. 


This is flow field around the UAV in cruise flight with the flap open, the drag penalty can be seen clearly in the form of vortices.  


This brings me to the end of my blog post. I'll update you guys on more once I get some substantial results. 


Looking forward to your expert opinions in the comments below. 

So long and thanks for all the fish.


E-mail me when people leave their comments –

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

Join diydrones


  • A very interesting job, but there are some points that I dont get... First one is why did you use a flying wing concept for such a slow plane. You say its cruise speed is 15.9 m/s and you lose lots of lift for sweeping the wing backward. Considering the manufacturing the plane you can extend the booms rearward and use a inverted v configuration and make your le sweep zero that must reduce your wing area and decrease your skin friction drag and also ease your manufacturing unless you're going to build a full composite which will be very heavy plus you will reduce the weight of the plane. The second thing is that for the naked eye your fusulage or letsay cargo Bay is too small. I was thinking how would you carry the batteries your 2 kg cargo. Of course it also depends on its length too considering 12 % thickness.
    All in all the job that you've done is amazing. Keep up the good work.
    K. Aria G.
  • Nice Job. I'm very curious how the winglets will operate the rudder controls. For the winglets are there for a very specific purpose aerodynamically and given the high stability bandwidth needed by the autopilot to control this airframe it might add undesirable effects especially during a maneuver (bank and turn). 

  • Looking good Karan! I will be watching for your substantial results!!

    D M Earl ll

  • @Richard Burd Thank you for the elaborate reply. That was really helpful. That's true, these softwares just churn up loads of data without giving the designed a feel of how the plane is going to fly and behave. I'll start working on the drawings as you mentioned and get X-Plane.Presently, all the CAD modelling is done in solidworks itself with which the CG calculations were carried out.

    I have a washout of -4 at the moment which was calculated using Panknin twist formula. This gives me cruise flight at around 4-5 degrees but as you said, it's not very accurate. I'm thinking of taking up the problem of control systems later in the design process as I want to first finish the aerodynamics part. I'm not really experienced with it but I guess that can be figured out. The airfoils that I'm using are Eppler 334 and NACA 34112, both of which have reflex. My static margin at the moment is 5% 

    I'll get in touch with you as soon as I'm done with the drawings. Thanks for all the help.


  • Take a look at the FIreFLY6 if you haven't seen it.

  • Moderator

    Very interesting

  • @Karan

    I would be happy to help you sort out some issues with that. I am quite on Advanced FCS. We can speak later on.

    I would first of all suggest: 

    Put the center of thrust (in VTOL) on the CG location

    Put the Aerodynamic center behind of the CG about >=10% of MAC

    The reasons I say this is because you may end up with no control power on VTOL config:

    * with Center of Thrust on CG, try to maximize the distance perpendicular to the thrust between the CG and the thrust  vectors of each motor

    Good luck

  • @Gary True that but I still feel VTOL are the future. I took this up because I wanted to work on something challenging. Sure I'll trying adding a tip underneath and post the results soon.

    @Hyperguy It's barely anything. The Cd value changes by around 0.00158 with and without booms. 

  • @David Torres Thank you for your reply. The current CG position is what it would have been, had it just been a flying wing and I plan to place the motors,payload/electronics in such a way that it comes to the required cg position. As for the change in CG position with tilt of motors, I'm yet to perform the calculations. I think the elevons would be enough to provide the counter balancing moments while tilting. Yaw control can be achieved by the rudders housed in the winglets. 

    As for the VTOL control I'm clueless. I'd be glad if you could help me out with that part. I'm not adept with Control systems so currently I'm trying to learn that and develop a consys model in MATLAB simulink. 

  • What is the interference drag like with the booms?

This reply was deleted.