Repaired Manned Multicopter Flies without Horrifying Crash

[amazingdiyprojects] has been making lots of test flights in his crazy eight propeller gasoline powered danger bucket.

We last covered the project when he had, unfortunately, wrecked the thing in a remote-controlled test flight.  He later discovered that the motor’s crankshaft bearings had, well, exploded. The resulting shrapnel destroyed the motor and crashed the drone. He described this failure mode as “concerning”.

Also concerning is the act of stepping into the seat once all the propellers are started up. He tags this as “watch your step or die”. Regardless, he also describes flying in the thing as so incredibly fun that it’s hard to stay out of it; like a mechanical drug. It explains why his channel has been lately dominated by videos of him testing the multicopter. Those videos are found after the break.

The device drinks 0.65-0.7 liters per minute of gasoline, and he’s been going through reserves working out all the bugs. This means everything from just figuring out how to fly it to discovering that the dust from the ground effect tends to clog up the air filters; which causes them to run lean, subsequently burning up sparkplugs. Dangerous, but cool.

81 thoughts on “Repaired Manned Multicopter Flies without Horrifying Crash

    1. Strap a human dummy into the pilots seat -> fly it to the limits -> test it to the limits and learn where the safe boundaries are. Proceed to strap self into seat and enjoy flying :)

  1. Are the bars of the metal frame duck taped together? Also, the cockpit reminds me of Herman Monster’s hot rod. I can;t help thinking it but the cockpit looks like it could easily double as a coffin.

    It needs a flip sensor that fires an ejector seat on tilt greater than 30 degrees.

      1. Yeah, makes sense. I think it would be safer and more stable with the cockpit slung underneath. If it were my build, I’d call it” project screaming death chair!” — On a serious not I hope the builder stays safe and all its flights are successful.

          1. Yes, but no. It’s more about surface area above or below the axis of lift in crosswinds. Area under it, will make lifting surface dip down toward wind, which results in lifting force acting against wind. Area above it, the lifting surface goes up, lift vector swings to amplify effect of wind, craft flips over.

            This is considering lift to be the vector force generated from pressure differential, not the “lift is all forces acting in 180* opposition to gravity” approach that aircraft people tend to like.

        1. Actually, if I was gerry rigging this for myself, I think I’d go with first, some landing gear/legs, with a bit of spring, and second, a couple of automotive airbags right under the seat, with simplest mechanical triggers possible, basically feelers that hit the ground when the landing legs reach elastic limit…. though if you’re coming down unholy fast they might fire you back in the air again LOL.

  2. I dunno…it adds weight, but I think, for me at least, a full face helmet would be a must.
    And a pair of skis on springs might make those hard landings a little softer.

    1. I think this is capable of much greater altitude than Furze’s flymo. If I were to be thrown into a bunch of hot, fast spinning stuff, I think I’d rather do it nearer the ground.

      1. Furze did a great deal of experimentation before producing the hoverbike that he filmed. He found that it was just marginally capable of flight. If he had used four motors instead of two I suppose the payload would have been greater, but it would also have been more dangerous.

    1. Of course he’s using a self leveling device you dolt. According to one video description, a KK 2.0 flight controller. Do you realize just how much fiddling self leveling actually requires? It’s a finely tuned PID loop dependent on prop size, engine power, and weight of the multicopter.

  3. Man that thing is scary! Anyway, I have some ideas he could improve it with but they all add weight:

    1. Use a “tail rotor” (maybe also a “nose rotor?”) with a flat-bladed prop for yaw control, driven by an RC plane battery pack. This way you don’t have to sacrifice lifting power for yaw.

    2. If the air filters are clogging, run hoses from each of the motor intakes to a central airbox fed through a large automotive air filter. A washable filter could save money on constantly replacing individual motor filters, and eliminating the restriction of having the individual filters full of dust most of the time will improve performance.

    3. Set up a system to monitor engine RPM (just optical sensors by the props could work) with a big bright warning light that will come on if there’s a large discrepancy. This can warn you of a failing engine and you won’t have to peek around the sides to check that all the motors are running on takeoff.

      1. Probably not more efficient. It’s surprising how difficult it is to make a ducted fan more efficient than an open propeller, especially if you consider the weight penalty of the ducts. The problem is complicated, but it basically boils down to the drag induced by the air flowing through the ducts.

        One further difficulty is that in order to make an efficient ducted fan the prop to duct clearance must be very small. Maintaining that at thousands of RPM and various flight conditions requires extremely stiff ducts and props, which in turn requires real (prepreg, autoclaved) carbon fiber parts in order to be lightweight.

        I’ve read about one helicopter-backpack project that used carbon fiber ducted fans, but that was about safety, not efficiency, and it was a highly funded project with the engineering ability to create custom carbon-fiber turbomachinery.

        Definitely not a good idea for this project.

    1. Long air hoses increase resistance – not so good.

      But I don’t understand this from the beginning: Clogged air filters will make the engine run too fat not too lean.

    2. Typical RC engine spark ignition modules already use a magnet to detect the crankshaft position and to determine when to spark. The pulse that they detect is already broken out through a 3 wire servo cable. He shouldnt need to add any optical speed sensors to this setup.

    1. Agreed, not to mention the improvement in safety. As it is, if he comes down at an angle a motor could break off its outrigger and be twisted toward his “safety” chair. Plus a forward or longitudinal rollbar in case he flips over in the air.

  4. After some thought I do not approve of the approach this person has taken to this… He has done a lot of things in a really foolish way… It is incredibly reckless an approach. A slacker-ass minimum effort approach to death-trap suicide aeronautics.

    I really hope his tests go well.

    1. I particularly like the way it’s designed so that should he climb to a decent height, dive, and panic full power zoom, that the cross spars will fold in the middle and the props chop him to pieces to spare him the agony of waiting to die from hitting the ground.

  5. Awesome to have achieved flight in this way- all you lot who are suggesting add-ons (ducted rotors, central airboxes, nose/tail thrusters) may not realise that he’s got very very little room left in the power:weight ratio to play with.

  6. How does clogging up air filters (thus reducing air supply) cause an engine to run lean? Shouldn’t it be the other way around, ie. running rich? (I know nothing of mechanical carburators)

  7. Wow, Kudos to you Gerrit. You have more balls than me, and I’ve done some pretty stupid things.
    Heres my two cents. It seems any failure with your craft could turn catastrophic pretty quickly. And the proximity of those high speed blades so close to you has the real potential to cut you to shreds. What if you rotated the engines 180 degrees on their support beams so they point down, adding a layer of distance. And should you crash, the engines will hit first and the ground will turn them off for you. And to accommodate the new height add four feet in the form of landing gear, you could shape them into inverted “j”‘s of sorts and benefit from a spring action.
    Well, keep up the good work it was fun seeing your exuberance and delight.
    Bob

    1. Gerrit just wrote the article, he works for HAD, he isn’t the guy with the amazing deathcopter. Dangerous though it is, I still want one.

      People have mentioned, in the previous article on this, he could do better using fewer, larger engines, but that would be a lot more complicated. Still, I suppose there’s nothing stopping him adding another engine / prop or two. So he could get a bit of extra lift to carry things like guards around the blades.

      There’s lots of improvements could be made to this, but still, bloody hell, bloke’s got a flying chair!

    1. Not to mention it’s complete lack of an asteroid deflection shield, if he attracts one of those 2+ mile diameter suckers, then poof, biggest extinction event since the Permian.

      1. Nah he’s talking about a single point of failure, important to consider when you’re sat in a flying obituary. 8 independet engines and props offer some backup in case one fails, but there’s only one control. Gods forbid the battery runs out.

    2. Big red hardwired “land gently” button? And a smaller, covered, “kill engines” button.

      Maybe the “gentle” button could cut some of the engines, enough to drop down gently. Or maybe have 8 buttons so you can kill a malfunctioning one.

  8. I think in this construction better way is use variable pitch propeller. It’s give a faster thrust control and if we use large blades maybe we can use autorotation. Its a “stingray” variable propeller quadcopter on YT and they have better flight properties.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s