LA Heli Rotorshape No. 2

Eventually my first Rotorshape broke it's rotor mast again.  This time it wasn't possible to get the mast to relocate correctly in the mount so it was time time to build up my spare kit.
I decided to make a few modifications based on my experience with the first one.

Wider undercarriage to make ROG and landings easier on rougher surfaces.

firewallThe original undercarriage was 8" wide which is OK for use on smooth surfaces as long as take-offs and landings can be performed directly into the wind.  

The wider undercarriage is made from some 14 s.w.g. (2mm) wire bent to shape.  The new wire is still mounted in the original firewall/motor mount so the "apex bend" still had to be 90 degrees.  I wanted to maintain the same nose height on the ground so an extra bend was added in the wire just below the point where it exited the motor mount.  The final shape resulted in an 11" wheel spacing.

The picture also shows the extra balsa quadrant I glued to many of the joints to give then some extra strength,

front viewI also decided to use a different motor and prop for this version.
The motor is a Hacker A20-26M and the prop is an APC-E 8"x3.8". 

The wheels are the larger 55mm ones I had fitted to the previous model to improve the ground handling.

Motor
The motor has four tapped holes on its back plate, and two of them lined up with the exisiting holes in the motor mount.  The instructions  show to fix a couple  of captive  nuts into the motor mount, but because I had decided to use this motor before I glued the mount together I didn't glue the nuts in place.  The motor is held in place with a pair of bolts through the motor mount.  Their heads are just reachable with a long thin screw driver.



Strengthening the bracket that holds the rotormast.

Back PlateThis is the plastic part that forms the back of the fuzelarge.  The tail boom passes through the hole in the front and the rotor mast is held in the collar at the top.  This part eventually snapped on my first rotorshape.

Plate with doublerI made a "doubler" from some of the 0.5mm gf sheet I had purchased to make the hinges on my Minimum (not yet started).  I used epoxy resin to attach this to the front of the part and then cut out the hole for the boom.

I also trimmed 0.5mm from the back edge of the gf sheet that makes the bottom of the battery  box  to take account of the extra thickness .



Bracing the rotormast and tail boom against each other.

On my first rotorshape both the tail boom and the rotormast had been broken close to the points where they are held in the bracket.   To share any "shocks" between the two carbon tubes I decided to add a brace between the two tubes.
braceThe brace is made from a piece of the same carbon tube as the mast and boom.  The ends of the brace was shaped using a round file so that they fit around the boom and mast.  Then brace was initiall held in place with a few drops of thin CA, then the joint areas were covered with epoxy and wrapped with "extra strong polyester/cotton thread".  Extra epoxy was smeered over the outside of the thread to finish off the joint.

Maiden Flight.

I maidened the No. 2 Rotorshape today.  It flew really well!  I think it's a bit lighter than the old one as it seemed to "float" much more on landing.  I flew it for over 15 mins in total.  It did need lots of right trim so I will be making some adjustments to the servos.  I also need to re-program the throttle to aileron mix which makes "dead stick" descents easier.
And yes, I know the stickers are on upside down!  I wanted to keep the colours on the correct sides, and I used the other two (right way up) stickers on the first one.

Ready to fly

Trouble with Takeoffs.

Takeoffs with this model always proved more difficult than they were with my first Rotorshape.  It needed very careful throttle control during the ground run because too much power caused the tail to lift which reduced the tilt back on the rotors.  So rather than reducing the take off run, too much power increased the run because the rotor spun up more slowly.  Often the model would be at the limit of visibility before it lifted off, which made control very hard for the first few seconds until it was visually clear of the ground.

Yesterday I tried adding a small piece of lead just infront of the tail plane in an attempt to keep the tail down during the ground run.  I was amazed at the improvement ! I can only think that the epoxy and other glues that had been used to repair the tail of my first model had added a similar weight before I started to try to ROG with it.