back to article It's raining drones, but just one specimen: DJI's Matrice 200 quadcopter

The UK's Civil Aviation Authority has temporarily grounded DJI's Matrice 200 following reports of the commercially used drone suffering a complete loss of power mid-flight and crashing to Earth. Chinese firm DJI has admitted its engineers are "urgently" probing customer claims related to the waterproof quadcopter series, which …

  1. Anonymous Coward
    Anonymous Coward

    Drone crashes shock. Next story Bear in wood scat scandal.

    I thought intermittent loss of power and falling to the ground without operator control was a given. Though that maybe because the 'Anonymous Coward Drone Mk 3' I am flying was built by the 'Incompetent Idiot' Mk 1', who is also at the controls.

    1. IDoNotThinkSo

      Re: Drone crashes shock. Next story Bear in wood scat scandal.


      My Mk 1 build made quite a dent losing power at 50m, as it was hauling a Canon camera at the time. As per the regulations, there was nobody anywhere near.

      Fly a drone? Gonna crash. Some operators seem to forget that. Don't forget your hard hat.

  2. Anonymous Coward Silver badge

    Rise of the machines

    or rather, "Fall of the machines"

  3. Anonymous Coward
    Anonymous Coward

    Expensive brick

    "When prompted on the DJI Pilot App, we recommend all customers to connect to the internet on the app or DJI Assistant 2 and update the firmware for their aircraft and all batteries to ensure a safe flight with their drone."

    Which suggests when you update the firmware, they'll effectively brick the multi-£000 drones until they're safe to fly again. Wonder what the refund status is for that situation?

    On a side note, there is firmware for batteries now. Good grief!

    1. I ain't Spartacus Gold badge

      Re: Expensive brick

      There has to be firmware for Li-ion batteries, as the charging needs to be controlled. Plus you need to have a battery meter - which are quite unreliable on consumer tech. In the case of a drone that presumably has to auto-land on low battery status - I guess you need a method to test the battery meter and re-calibrate it every so often.

      1. Andy 73

        Re: Expensive brick

        None of which has to be in the battery itself. However, certain manufacturers are keen to lock their customers into their own 'special' batteries rather than buying equivalent items at half the price. Customers lap up the 'smartness' of the batteries they pay through the nose for :)

        1. Sampler

          Re: Expensive brick

          Given the model was released in 2017 that means the batteries have had ample time to be charged and discharged a fair few times, given they're commercial drones and not playthings, which probably means the batteries can't maintain peak output, so, although not depleted, output isn't high enough and that trips something (a resistor or capacitor is suddenly in the wrong state) and the whole system gets confused and goes down.

          Kinda like my Note4 when I tried to open the camera, I'd have a 1 in 4 chance of it just shutting down, ah, those were the days, when you could easily replace a battery in a phone...

    2. Simon Ritchie

      Re: Expensive brick

      DJI describe their batteries as "intelligent". Putting intelligence into a battery is an expensive process, apparently. They cost about £200 each, and some DJI drones need five of them to carry their full payload.

  4. Anonymous Coward
    Anonymous Coward

    "there is firmware for batteries now"

    There has been for ages.

  5. Anonymous Coward
    Anonymous Coward

    But they're supposed to do that ...

    My introduction to R/C helicopters was my friend who would fly his 'copters until they broke. If they were still flying after a few minutes, he would say, Hey, watch THIS! and a few minutes later, the unit was ready for the repair shelf, waiting for parts.

    So when I got my first R/C helicopters and drones, I thought I could do a lot better than my friend. I was going to fly the 'copter/drone, land it, put it away unbroken.

    Of course, you know what happened next ....

    (Now I know why the drones I buy come with spare propellers.)

  6. Anonymous Coward
    Anonymous Coward

    Raises an interesting question

    For someone without an aeronautical engineering background anyway.

    What are the trade-offs of implementing autorotation or some sort of controlled glide in the event of total power loss in a drone?

    1. Chrissy

      Re: Raises an interesting question

      Gliding requires wings: On a drone, those are dead weight and drag inducers for all the flight envelope for the entire life of a vehicle unless its actually used in an accident.

      Autorotation requires a large rotor surface to convert its kinetic rotating energy to a large amount of lift in a very short space of time by changing the pitch of the blades using cyclic. No matter the size of drone, you don't have the cyclic system on any rotor as that would introduce a massive weight and complexity penalty, so you cannot autorotate at all.

      Parachutes cannot deploy quickly enough ti be useful under certain heights, so the Dead Man's Curve for a manned drone is huge compared to a heli.

      1. Khaptain Silver badge

        Re: Raises an interesting question

        What about a small canister full of compressed helium, which could inflate a balloon, thinking along the lines of airbag tech. Might not stop the impact but at least if it would slow down the descent.

        A couple of jagged edge kilos dropping suddenly from 50m or so must be quite deadly.... so almost any idea would be better than none.

        1. Chrissy

          Re: Raises an interesting question

          Helium: Maybe for a small drone, but the cubic metre of helium:kg of lift ratio puts that out of reach for anything above <10kg.

          Think of the average weather balloon's size: maybe 2metre + wide to slowly lift maybe 1kg of payload? Then think of the huge balloons used by Kittinger and Baumgartner and imagine trying to compress even some of that into a cylinder small and light enough to not shrink your payload capacity enough to make the whole endeavour pointless.

          Lifting balloons' lifting effect is also too "slow" for this "need lots of lift NOW" use, else Baumgartner wouldn't have taken 2.5 hrs to reach 38km - 253metres/minute or 4m/sec with a realllyyyy slow initial acceleration.

          A basic airbag solution could maybe work for non-human payloads, but deceleration trauma of 120mph to zero in around 4 feet would really mess a human up.

          Maybe rocket assisted braking like Curiosity, but I wouldn't want to be the poor sod on the ground set alight under the rockets!!

        2. Annihilator

          Re: Raises an interesting question

          How about a hydrogen canister instead?..

          1. Anonymous Coward Silver badge

            Re: Raises an interesting question

            How about explosive disassembly on power fail? That way none of the bits that fall would be big enough to cause damage.

            Don't need big explosives, just match-head style would suffice in the right places.

      2. Anonymous Coward
        Anonymous Coward

        Re: Raises an interesting question

        > Gliding requires wings: On a drone, those are dead weight and drag inducers for all the flight envelope for the entire life of a vehicle unless its actually used in an accident.

        Errm... lots of fixed wing drones out there. In fact, anything that requires range over manoeu... (fuck it) doing lots of turns.

    2. notowenwilson

      Re: Raises an interesting question

      Auto-rotation requires variable pitch propellers which have a major weight penalty compared to the fixed pitch units currently installed on multi-copters. It also requires higher inertia (longer and heavier) rotors than currently used. If you were going to do that you'd be better off with a traditional 1 or 2 rotor helicopter.

      Better option would be to put a ballistic parachute on it.

  7. Chrissy

    Manned multicopters? No thanks

    This is why I ain't ever getting in a flying machine that can't either glide or auto-rotate to at least a survivable landing in the event of any - partial or all - power failure... although events in Leicester make me now doubt auto-rotation as a solution as, no matter where and what heli you fly in, they all have to transition through the Dead's Man Curve

    1. }{amis}{ Silver badge

      Re: Manned multicopters? No thanks

      it is a bit of a 6 and 2 threes though helicopter crashes caused by hardware failure are usually catastrophically caused by either a rotor head or gearbox fail your really lucky if its something that leaves enough of the system working to make a safe autorotation landing.

      manned multi-rotors do have the possibility to be safer as each rotor can be an almost totally independent system and unless you are a total looney you won't be flying a manned quadcopter with no redindancy if you look at most of the test manned multi rotors they are using 8 or more props which should allow at least a 2 prop fail more if you are lucky with the location of the breakages.

  8. J J Carter Silver badge

    It has to be said

    Thankfully it's CPU pilot always heroically steers the doomed drone away from primary schools or puppies.

  9. Stevie Silver badge


    Fixed wing for the power-loss, no-crater win.

    1. Griffo

      Re: Bah!


      Fixed wing for the power-loss, no-crater win.

      Tell that to the people on JT610

      1. Stevie Silver badge

        Re: Bah!

        OK, we’re playing “Yes But” are we?

        Hmm. My turn: Sully!

  10. I3N

    Damn batteries ...

    SCMP not my favorite source -

    But to blame 4G???

    1. Khaptain Silver badge

      Re: Damn batteries ...

      Could be someone having some "fun" with a drone jammer such as one of these :

      Drones would normally use 2.4 or 5.8 Ghz which is not 4g but it is easy to understand how they are often confused... simply because they are wireless is enough for some...

      1. Tom Chiverton 1

        Re: Damn batteries ...

        Surely sensible LOS procedures would be to hover until contact is restored, or critical battery, in which case a slow descent to last-known-good zero height ?

      2. notowenwilson

        Re: Damn batteries ...

        Unlikely to be a jammer issue. DJI drones will either stay in place, land, or go home if they lose signal. They don't just drop out of the sky due to lost comms/gps

  11. Anonymous Coward
    Anonymous Coward

    "It's Raining Drones!" Meh

    Meh, It could be worse....

  12. Mayday Silver badge

    This is how drones go down in Aussie

    Wedge Tailed Eagle vs Drone:

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2019