H8 Heavy Lifter

[teslahed]

Hi Guys,

Now it';s time for something different (from me at least!).

I need to start designing a new multirotor for work. The mission requirements are;

<7kg total flying weight (as close as possible).

Need motors / ESC redundancy as will be flying over sensitive areas (I could tell you but then I';d have to kill you!).

Needs to be able to carry a gimbal and / or heavy sensor load pointing downwards the whole time. It does not require a forward or sideways looking gimbal as it wont be doing aerial photography.

Needs an open source flight controller that will do waypoints and automated stuff.

My thoughts so far are to go with an H8 frame not dissimilar to Stef';s latest creation;

http://www.multi-rotor.co.uk/index.php?topic=9467.0

With some modifications;

Firstly i wouldn';t need the gimbal on the front.

Secondly I would open up the interior section by widening the two longitudinal supports away from each other so that wide loads could be mounted in the centre of the aircraft.

Thirdly I would go up a tube size so that it will cope with all the extra weight.

Fourthly I would redesign the motor mounts to allow it to fly in an H8 rather than H4 configuration.

And finally; Pixhawk!

Still at the early design stages but any thoughts / opinions / suggestions are most welcome

[Hands0n]

Awesome Chris.  This is going to be interesting.

Initially, for FC, I';d have suggested Naza M v2 for simplicity, ease of set up and use, and basic reliability. It has a good pedigree.

However, you then went on to say it had to be based on Open Source and in my opinion that brings you to something based on Arducopter.  Certainly Pixhawk or VR Brain put you in the Arducopter 32bit world. All future development of the flight code will be based on the capabilities of the 32bit hardware, although legacy APM 2.x support is likely to continue for some time yet.   I don';t think the code will fork, the developers have been quite clever in their code design to facilitate multi-platform support (the difference between software and hardware engineers ::) ). 

Are you wedded to an octo and not a hex?  Less complexity and weight but also slightly less redundancy in a hex of course.  But hex are used industrially all over the place so I don';t think there is any disadvantage there.

Have you had any thoughts on collision detection and avoidance?  Arducopter code supports a number of audio and optical sensors these days. They may come in handy, depending on the purpose of this aircraft.

Construction materials are a consideration I suppose.  Everything I see these days seems to be based on Carbon Fibre and is all gangly and bouncy, looking like some kind of flying black spider creature.  I don';t find them at all attractive.  Yet there are so many, I think better, materials available, both natural and man made. 

And while on the subject, looks. Why cannot a large utility UAV actually look good? 

Not sure what you';re slinging underneath as payload - but you mention gimbal and from the description it sounds like possibly something simpler might even be better suited all round. Maybe purpose designed and built for whatever it is, driven by lightweight servos or motors.

Some thoughts, seeing as you asked  ::)

[stickboy]

Nice, ever get the project completed or got any pics?

I';m working on plans for my own similar build using an H frame with quad-8 configuration. Wanting to using 15" props on an aluminum frame to get something with real lifting power but haven';t seen anything similar to what I';m shooting for.