Pixhawk and DJI Spreading Wings S900 Frame Kit

Hello everyone,

I am planning to build a hexacopter using the DJI Spreading Wings S900 Frame Kit, Pixhawk, and  T-motors. I will definitely appreciate any suggestions about any other frame or motors to use or if someone has already done what I am planning on building.

I intent to have a payload between 5 to 6 lbs.


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

Join diydrones

Email me when people reply –


  • Virgil,

    A bit perplexed as to the T-motors if you are using a DJI S900.  We have built multiple S900's for mapping and only use the Pixhawk FC.  We have also retro'd some S800's with the WooKong for the  Zenmuse and use the Pixhawk for flight control due to DJI's inability to resolve vibration issues on the S800.

    Next will be to integrate the Navio+ from Alliant UAV here in the States, with the S900 for mapping use, and utilize 4G LTE for cloud access and telemetry via the internet.


  • Virgil,

    We started with DJI gear and had multiple fly aways including the famous DJI V1 GPS "Flip of Death". Touch wood, we have not had a single problem with APM or Pixhawk. Flawless operation. That may be based on experience.

    Given that you are not buying an Off The Shelf drone, you have already made the decision to "Integrate".  An S900 is a big drone with six rotating knives. This makes it expensive and difficult to Integrate, Test, Tune.  I can not see why you would want to do this when there are packages out there that will cost you exactly the same amount which have combinations of motors, propos, ESCs and controllers that are know to be compatible. That said, if you still want to go down that path, I would offer the following;

    1) Given that you are integrating, the Pixhawk and the DIY drones community means that there is much more information out there and flexibility to integrate, test, log, graph, monitor parameters than you have with DJI. You do not need to learn c++ to tune a pixhawk. All of the tools are readily available. I feel the problem people have with Pixhawk is that it offers you so much flexibility that there is a steep learning curve. Do the background research and all will go well. There are hundreds of YouTube videos out there that take you through the process from taking it out of the box to precise tuning.

    2) Failsafes! Failsafes! Failsafes! make sure that you understand the failsafes of whatever flight controller you go with and make sure they are enable appropriately.  Understand what to do once the failsafe has been activated (particularly the throttle position).

    3) The DJI community is flooded with people trying to work out how to add a gopro gimbal to a Phantom. It makes it a lot harder to find the information you need when you are trying to integrate your own design of drone. Most of the DJI community discussions are at a significantly lower technical level than that of the DIY drones community simply because the vast majority of DIY Drones users are building their own drones.

    4) We have an S800 with an A2, a tarot Ironman 1000 with a WKM and a couple of babies with NAZA.  The S800 has an IOSD, Zenmuse gimbal and 2.4ghz data link. We have never been able to get the full system to work. The dealer we bought it from has never been able to get the full system to work. Occassionally, we have to "Firm Up the Beast". This is a process by where you have to repeatedly update the firmware on all components until they talk to each other, like flashing the data link 13 times to get it to work.

    5) We have mapping drones running 320Kv motors with 18 inch props that are fitted with Pixhawks and gimbals and telemetry links and laser altimeters and OSDs that work perfectly without any C programming.

    6) One thing I have learned over many years is that the hardware is actually pretty irrelevant. The key to success is access to good documentation and a knowledge base.

    Good luck with the project.


  • Over 400 flights and 1 flyaway. That's my experience with Arducopter. The flyway was due to excessive EMI and because I am using AUAV-X2 (a derivation of Pixhawk) on that particular copter (hardware bug to be fixed in 3.3 rc9), the EKF failsafe did not warn or allow manual override.

     If you don't feel comfortable learning how to tune or don't want the myriad of options available with AC, maybe Pixhawk is not your cup of tea, but don't let anyone convince you DJI's are less prone to crashes and flyways.DJI is fixing stuff that AC did well over a year ago.


    dji s900 crash - YouTube
    Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.
    • While I can say I put a lot of time, effort and money into my Pixhawk Hexacopter with extended arms and DJI E600 propulsion system, I finally swapped the F/C over to a Naza M V2 due to general reliability issues and the lack of a really good 3-axis gimbal (the real desire for the larger platform).  Build info here:


       I had a lot of successful flights (more than 100) with my Pix - in Auto mode, stabilize, Drift (one of my favorites), and position hold, it just seemed that tuning and quirky settings seemed to bring this bird down A LOT with the Pixhawk.  And it was costly every time ($100 - $200).  My DJI F450 never had these issues with the Naza M V2 (over 300 flights), and the H3-3D gimbal generated superb video on my F450.

      I'm planning on using my Pix F/C on a new airframe, likely a TBS Discovery Pro top and bottom plates with maybe the DJI E800 propulsion system and AeroXcraft X55 arms.  This will be more of an experimental distance craft rather than a media platform.  My recommendation is to experiment with the Pix on a smaller, less costly airframe before installing it on a larger, more expensive air-frame - just my 2cents.

    • I have a IRIS+ that flies just like planned without problems one. I have over 200 flights on this little bird with out hiccup one. But my PixHawk right out of the box will not arm. I have tried three different GPS/compass  without any luck. I am a retired electrical engineer and my slant was satellite and land base communications. I just don't want to freshen up on my programming skills.. since I've retired in 2002. I may be behind the curve.  I wanted to build a larger Y6 that will lift a cannon 40D the IRIS cannot attempt to do this.  I really do not have the time to spend trying to just figure this out. Auto tune should take care of the PID's. But I have to get it armed to fly. So I am back on the NAZA V2 and yes it does have 16 way points. Same as the pixhawk however the pixhawk is unlimited on way points. Big deal I hardly fly more than 6 at any given time. So that is a wash as far as I am concerned. Still trying to make the PixHawk fly not giving up but not placing such a high priority in achieving flight. 

    • Have you went through the troubleshooting for arming?  http://copter.ardupilot.com/wiki/flying-arducopter/prearm_safety_ch...   When I can't get things to work it has always been something I missed.  

  • I've had a Pixhawk on a Tarot t960 for 18 months with many hours on it. It has flown flawlessly and not a single flyaway. I use it for flying waypoints and loiter while snapping pictures. The Tarot T960 is about the same size as the S900. I have never bothered to even tune as the base settings work just fine for a large craft. The main problem with any flight controller is to keep the GPS away from things that cause interference such as ESC, power wires and other RF noisy devices like go pros. I've had flyways with APMs in the past due to GPS signal being interfered with or electrical currents throwing off the internal compass. I know only use external compasses for both Pixhawk and APM.

    I believe Naza is a little easier out of the box but is limited on what you can do with it. To use 95% of the capabilities of the Pixhawk you don't need to be a C++ programmer. 

    What are your plans for the S900? If you are planning on just flying to a point and taking pictures then I would go with Naza. If it is to fly missions with waypoints or aerial mapping then I would go with Pixhawk. Think of the Pixhawk as assembling a PC from parts and Naza more of a Mac.

    I've enjoyed my Tarot frame and its made from really good carbon. However it comes assembly required.

    While I don't have Tiger Motors you can't go wrong with them.

    Just my 2cents.

  • With Pixhawk you will be able to use Arducopter capabilities so you do not need to go for an expensive DJI A2!

    Marco has done this before. watch his channel:


    • You do not need the A2 the NAZA V2 for 299.00 USD will power that just fine...

    • Naza v 2 is same as Naza lite only firmwear difference you can upgrade lite to V2!
      Pix hawk takes some tuning to work best. DJI not so much. Just as likely to crash ether of them .
This reply was deleted.