Hexacopter

It all started when Doug Radford forwarded an email with this link http://mikrokopter.de/ucwiki/VideoAbspielen?id=188.

The link points to a German web site with a video demonstrating a Hexacopter. Doug's interest was aroused and, as the website is mostly in German, he asked for some German - English translation from yours truly.

The more I started reading, the more I got interested and the link to the shop became increasingly tempting. The only problem was that the price level in Germany is somewhat different from the Chinese benchmark we are used to...but after a few days of hesitation I had a weak moment and started clicking. A week later an almost empty box arrived....

imag0017 

I was lucky that my flatmate Andy was fascinated by the Hexacopter as well and offered to build the frame. So instead of spending $200 on some aluminium pieces, I ordered the electronics from Germany and the frame parts came from our local hi-tech supplier - Bunnings.

Many hours later (I think Andy underestimated the work required!) a new creature was born....

imag0027

As you can see, building a Hexacopter has not much in common with building a Hurricane. A Hexacopter consists of 6 aluminium profiles with a motor on the end, a baseplate in the middle which holds everything together and a lot of electronics.

There are 6 speed controllers (each motor needs its own controller) and in the middle is the main flight control board.

imag0028

Work hasn't quite finished in this picture but clearly visible is the amount of soldering required. In fact, all the boards require the soldering on of some additional components and then comes all the connections to the speed controllers and motors. The 6 smaller boards on the picture are the speed controllers and the one in the middle the flight controller.

The Hexacopter has no moving parts apart from the motors but all motors are mounted fixed. So how on earth does it fly?

It all works with some smart software which varies the speed of each motor as required.

  • opposite props are counter-rotating to eliminate the rotational forces
  • for moving forward the rear motor speed is increased and the Hexacopter is, thus, tilted slightly forward
  • for moving backwards the front motor speed is increased and this tilts the Hexacopter backwards
  • same principle applies for moving to the right and left
  • rotating (on the spot) is achieved by creating a speed difference between opposite motors (rotating clockwise -> increase speed of left and right props and lower speed of front and rear props)

The speed controllers are not quite the usual cheapo version either. A normal ESC, as we all know, is controlled by PWM signals from the receiver. Updates from the receiver to the ESC are done 50 times per second and while this seems rather fast, it is not fast enough for the Hexacopter. The 6 special ESCs are connected via I2C bus and are updated 500 times/sec (each)!

The flight control board contains an AVR Atmel microprocessor, 3 gyros, a height (pressure) sensor, a 3-axis acceleration sensor and a few more bits and pieces.

And as we already have a powerful microprocessor on board we don't need a normal receiver either. Instead we take a Spektrum remote receiver and solder the wires directly onto the flight controller board. The flight controller talks directly to the remote receiver and decodes the signals recieved, no normal main receiver required. (On the picture above the twisted wires from the top go to the receiver.)

As you would expect, instead of setting up aileron and elevator thows the Hexacopter is setup via computer and the special Microkopter tool which is available from the developer's web site.

Setup itself is reasonably easy if you go for the default settings but there are many, many, parameters to play with if you want to optimise and customise the behaviour.

Power is supplied by a normal LiPo. Either 3 cell or 4 cell can be used. I used a 4s 4000mAh as this was what I had available. Alll 6 motors are supplied by the single battery but luckily the motors are quite small only drawing 10-12A max. But still, this adds up to >60A for full power.

 

Here is the completed Hexacopter. The green "canopy" comes from Coles (it looks suspiciously like Pete has been going to Tupperware parties - Ed) and some bright red and blue LEDs have been added to help orientation. Without some visual aid it is difficult to see the orientation with a symetrical design like this.

dsc00509

 

And the all important question -- how does it fly???

In one word - excellent!

It flies similar to a heli but the self-stabilisation functions level it out automatically and make it a fair bit easier to fly than a normal heli. It is capable of rolls and loops however sustained inverted flight is not possible as there is no pitch control.

All the little electronic helpers can be adjusted and disabled if desired but at least in the "beginner" mode it is very easy to handle.

Here the video from the maiden flight.

{flv img="Hexa.jpg" showstop="true" width="960" height="540"}Hexacopter{/flv}

 

What next?

There is certainly more to come.

First will be the navigation controller (already on it's way from Germany!) which adds GPS navigation. This allows autonomous flights, position hold and "coming home" functionality.

The Hexacopter is obviously an ideal platform for photo and video work and the camera mount is already in the planning stage. Soon we should get some great video from high up....

And then there is still the possibility of MKII. The software can handle different configurations. So from a light-weight 4 motor racer to a 12 motor monster everything is possible Cool

 

Link: http://www.mikrokopter.de

 

Peter Wyss

Part 13 is ready for download - [HERE]

Austar’s Fieseler Storch

I needed another back-up for Scale events as my well-worn 84” Spacewalker was getting a bit shy in lifting her skirts in performing set maneuvers.

Going over to Austar’s premises at Prestons near Liverpool, I came away with the Fieseler Storch ARF for engines from 20-26cc. This STOL aircraft became quite famous in WW11 as a reconnaissance aircraft as well as rescuing Mussolini from his castle when the Allies were closing in. At first glance it’s an ugly looking machine with its forward pointing undercarriage, large tail and long nose. However, after a time, she becomes quite beautiful as the build progresses.

 

 

The Storch is a high wing tail dragger, single engine, 96” span aircraft and I elected to power her with the 26cc Turnigy petrol two stroke.

Opening the sturdy cardboard box from Austar, I was disappointed to find scant, meager build directions, flimsy formers and some quite shoddy hardware. One of the wing tubes was missing and the covering, although drab olive, was bright and shiny. With a wipe of that wonder primer ESP followed by a spray of matt Estapol (courtesy of Doug Radford), our Storch now looked like a military aircraft.

I was warned this was one tail heavy machine and NOT to put the dual elevator servos down near the tail (as per directions) as this only exacerbated the problem. Little did I know it would eventually need nearly 2kg of lead up the sharp end!!

A non-standard leaf-spring tailwheel assembly aided steering and would take the bumps out of our grass field. 5 x standard Futaba 3001’s powered the throttle, ailerons and flaps. 3 x Hi-Tec 645MG servos worked the pull-pull rudder and split elevator. With the addition of an extra servo tray, the elevator servos were secured in the cockpit area with dowels and 440 wires transferring the drive to the horizontal flying surfaces. Other than that the build was fairly straight forward.

Inside the cockpit, the bare white wood stood out like the proverbial “dogs” so a coat of drab was given to dull down the interior. To encourage forward CG, both ignition and flight batteries were installed hard up against the firewall as was the CDI unit. Fuel tank was cable tied inside the cockpit - forward of the CG. A choke extension rod was mounted on a bell crank that exited under the aircraft. There was nowhere to install switches so, with the addition of ply and balsa (with the kind help of Barry Campbell); a patch was fashioned on the port side to accommodate the dual switches.

Ignoring the supplied piddley screws, I used large wood screws to anchor the undercarriage as well as the support struts. Even this was not enough. A hard landing broke the U/C so it was glued back with support timber and the whole lot fibreglassed from the inside (Thanks Grant!!). The kits stiff, no-give U/C needs oleos which I will fit in due course.

To finish off (to avoid bounce), 4” Cub inflatable tyres replaced the hard rubber kit wheels.

Instruction sheet stated the CG was between 85-95mm from the leading edge. To accomplish at 95cm, 1.1kg of lead had to be added to the firewall to get her to balance. She was now ready for the maiden.

It was a clear day with a steady wind from the sea. After hours of fiddling to try (unsuccessfully) to get the engine to fully rev out, Tom Sparkes gave her throttle and she did a three point take off.

Her bum hung noticeably and Tom had to fight to keep control. Then – shock – horror. Over the western end of the field she just kept turning, entering into a right-hand, downward pointing barrel roll. Onlookers scattered at breakneck speed as the big German monoplane came hurtling towards them. Skillfully our ex-CFI pulled her out with just meters to spare.

Shaking like a jelly, Tom landed the Storch, sat down, wiped his brow and commented this was one very sick aeroplane. Heaps more weight was needed in the nose!!!

I took her home and wondered why this baby was so tail-weighty. Another half a kilo of lead was attached to the engine stand-offs with cable ties and wood screws. This brought the CG back to around 82cm (never trust the instructions) from the leading edge. I was now worried about all the weight (1.6kg of lead plus 890gms of engine) hanging off the firewall.

A few days later at the field, Tom Sparkes checked engine timing to see if top end performance could be improved. It was factory pre-set at 45 degrees BTDC. No wonder it ran rough. Adjusted to 28 degrees BTDC, the engine revved out fine and there were plenty of herbs in reserve.

Tom again took the controls and the Storch, with all the extra nose weight now raised her tail, ran on the mains and lifted gracefully into the air. After a few circuits, she came in for a spotless landing. Tom commented she was a different aeroplane, behaving like a puppy dog. My main problem was how to integrate all that temporary lead into the aircraft.

After plenty of discussion, Tom Sparkes kindly machined up four heavy and longer solid brass stand-offs as well as a brass spinner nut to replace the lightweight aluminium hardware. With the aid of Barry Campbell’s artistic eye, a thicker 5/16” plywood firewall was cut and glued onto the original. I lavished a heap of chopped fiberglass cloth and resin into the front of the aircraft to add both strength and weight to the firewall.

With all this extra weight and the engine sitting around 30mm further forward, I STILL had to fiberglass in another 130gms of lead into the cowl. She now balanced at 82cm.

A week later all was ready and it was my turn to fly the lady. After a few deep breaths I slid forward the go lever and the Storch ceremoniously lifted her tail, ran for a few meters then majestically climbed out. I ran my tongue round a very dry mouth. After a few completed circuits my breathing returned to near normal. She floated in wonderfully on finals with the aid of a little flap and gently touched down. After taxiing back to the pits and hitting the kill switch, I wiped the moisture from the transmitter then my hands. At long last, I now owned a very gracious and polite scale model which will now (hopefully!?) continue to eat out of my hand.

I would like to sincerely thank Tom Sparkes, Grant Furzer, Doug Radford, Ron Clark and Barry Campbell for their invaluable help in this project.

Col

Part 12 is ready for download HERE
Part 11 of the Hurricane diaries is ready for download here.