Thursday 23 January 2014

Le Fish aerobatics progress

A few more tweaks.

The new solid 3mm CF elevator axle / spar is working perfectly.

I removed 10g from the nose to move the CG back to 80mm giving much better inverted flight control. Decreased the "normal flight" elevator throw and increased expo for smoother flying. Full throw, about 60º, is still available via the momentary switch.

I tried moving the pushrod down one hole on the elevator control horn to get 90º throw for proper mad flips but the elevator would return fully inverted. Might need to move the control horn back from the pivot axis or try a pulley system.

The vertical stab and rudder CA hinges failed in a nasty upside down crash so I re-glued the join and used clear gaffer tape for the hinges.

Anyway I'm steadily ascending the aerobatics learning curve and having great fun in spectacular locations. Here's some video from 13th beach in a lovely 10kn breeze then some thrilling 20 to 30kn later in the day.

Sunday 12 January 2014

Le Fish build follow up

After 3 hard sessions of flying (and crashing) it's time to assess how my build methods held up, and share some setup tips.

The wind has been 15kn gusting up to to 20kn for every flight. At that wind strength it's easy to lose control of the glider and see it cartwheeling backwards across the top of the slope, but I guess it teaches quick reactions and accelerated learning.

Elevator spar / axle
The 3mm CF tube elevator spar / axle snapped on the first day, a clean break across the fibres, which is very unusual for CF tube. That leads me to think it was a pre-existing weak point. So I replaced it with another 3mm tube. Dremel and knife freed the spar from the elevator planes.

The new spar broke on the second session, but cracked along the fibres this time, from torsion stresses maybe. Conclusion - 3mm CF tube with 0.5mm wall is not strong enough for the job. So for the third spar I glued a 2mm rod inside 3mm tube (until I can source solid 3mm rod) and that seems to be holding up OK, although two elevator planes don't feel as solidly locked to each other as I would like.

Fuselage laminate covering
The light 45um laminate did not bond well to the fuselage and came unstuck right down each side. Might have something to do with the fuselage flexing on impact.

So I tore it all off and recovered the fuse with 75um laminate. This was a bit fiddly with the plane fully assembled. I used 300mm x 50mm strips along the entire top and bottom with lots of tabs cut for the curves, 2 layers built up forward of the wing, single layers above and below the wing and two layers running down to the tail. The result is much stiffer, smoother and better bonded to the foam.

Fuselage cracks
While the fuse was undressed I checked all over for cracks. The nose was split open a little, there was a vertical crack just behind the wing leading edge, and the wing / fuselage join had parted in places. Gorilla glue to the rescue. I also added clear gaffer tape to protect the nose. This stuff is great, very strong and more stretchy and flexible than the coloured cloth tapes.

Loose servo
The elevator servo had broken loose in its hole. Just needed a few drops of CA to fix it back in place.

Transmitter setup
Full throws on the control surfaces are way to much to start with so I dialled in 50% dual rates for everything, 50% expo for rudder and ailerons and 70% expo for elevator.

I still wanted full throws available for attempting mad stunts so programmed the momentary switch to turn off dual rates and expo.

Snap flaps can be switched on, giving about 8% camber (down ailerons) for up elevator and 8% reflex (up ailerons) for down elevator. Apparently that makes loops and turns sharper.

I have full throw flaperons / spoilerons on the throttle stick, called 4D by some folks, for on-demand camber changes. I have only used this for landing so far. The curve for this mix has a 50% dead zone in the centre so I'm less likely to make camber changes accidentally.

I also have 15% and 30% flaperons on the 3 position switch for low lift times.

Centre of Gravity
Doing a loop

Initially I set the CG at about 70mm which is OK for starters but noticed heaps of up elevator was required to fly inverted. This tells me the CG is too far forward, so by removing 10g of nose weight it is back at about 80mm now. That will make the elevator more sensitive so I may need to add more expo to keep it flyable. The next flight will sort that out.

Sunday 5 January 2014

Le Fish maiden

The forecast was for 15kn NW increasing to 30kn mid morning, so I raced out early to the closest NW slope - Pt Henry.

The actual wind direction wasn't great, 45º off the the right, but OK for a first throw.
On the first crash the elevator had inverted and I clumsily cracked the spar / axle twisting it back the right way. The off side of the elevator could almost rotate independently of the controlled side but I decided to press on regardless to a better slope at Clifton Springs. .

In the meantime the wind had increased to 15 to 20kn but from a perfect direction.

Here's the maiden video warts and all. Le Fish flew spectacularly and I even managed a few tentative loops and inversions. Gliding back and forth along the slope was so smooth and controlled. I can only improve from here.

I learned not to take my hand off the elevator stick before trimming for level flight. I reached up to turn on the GoPro and the plane went into a dive and disappeared down the slope. It landed in nasty head height scrub but no damage apart from a few scratches to the pilot.

Dual rates and expo of 50% for everything helped calm the flight down for this first test.

A new stronger elevator spar / axle was fitted afterwards. 3mm CF tube with 2mm rod glued inside this time. Might still need a solid 3mm rod for durability.