Dragon

A peek in my scrap-box, and I concluded I'd probably be able to lay this one together entirely out of 'discarded' wood. I selected three sheets of 'premium' 1/8th balsa for the ribs, but the bulk of this aeroplane will be built from scrap.

Selecting scraps of 1/8th and 1/4 balsa, as well as scraps of 1/16th, 1/8th and 1/4 ply, I drew out the formers, and cut them out on the bandsaw. Then, with the good 1/8th sheet, I repeated the process for the wing-ribs, obviously doubling the sheet so as to produce two of each rib to facilitate a left/right pair of wings.

The scroll-saw will be brought into play to remove interior wood from the formers, to reduce weight. Also, the ribs need to be drilled. I want to try buildng on a jig made of two 5/32nd wires, so those holes will have to be drilled. Also, since I don't have any 1/8x3/8 spruce strips for the main spars, I will use hard balsa in their place, and slots are needed for this. I'll also run two 3/16 hardwood dowels as embedded spars. Holes will be needed for these, as well as for servo-lead tubes, and maybe some extras for lightening on general principles. This will be needed to finish the ribs, and allow wing construction to commence.

The wing is built up conventionally. I made a few changes, though. The 3/16 hardwood dowels were added as spars, and the 5/32nd wire jigs were also used (removed before this photo was taken). The jigs were a good idea, but not properly implemented by myself; there were a few problems. Next time, I'll do a better job with them.

Here, the trailing edge is being finished.

Winglets are removable. They were framed up out of strips of 1/16th sheet balsa (ply at the anchor points) and cross-grain sheeted with 1/16th balsa. They are located on the wing with two 3/16th wooden pins (you can just see the ends of these flush-sanded with the winglet on the upper wing) and retained with one nylon wing-bolt per winglet. Look at the inside of the tip-rib of the lower wing. You can see the inside end of one of the locating pins glued into a small balsa block just behind the cylindrical spar. The black object glued into another balsa block just behind the main-spar shear-webbing is the plastic captive-nut for the attachment bolt.

The wingtips can be swapped, or can be inverted for a 'drooped' winglet effect, similar to the 'Canada Goose' canard aircraft.

Here we can see the wing dummied up. Two clamps hold the panels together for this photograph. Other possible configurations under consideration at this time can be seen here, here, and here.

The fuselage is a simple, slab-sided box design, with large triangle-stock in the corners to allow considerable rounding out. The Dragon is unusual in that the entire top of the fuselage is a hatch, from the nose-cone (not yet attached) to the front of the wing. And with the wing off, the remainder of the fuse right back to the rearmost bulkhead/firewall is accessable.

The hatch itself is constructed in almost the same way, but not as deep. The rear end of the hatch incorporates a cockpit, which also fairs in the wing.

Seen here inverted, with the triangle-stock and bottom sheeting yet to be applied. Two longerons extend aft (left) past the firewall, and must be cut off. Some of the remaining formers (for the hatch) and the bottom sheeting for the fuse itself can be seen in the foreground.

Here is another shot of the fuse. The bottom-sheeting has been added, and the gear test-fitted. The steerable nosewheel is attached to an internal ply bulkhead with a balsa doubler, and the main leaf gear is bolted to a ply plate fitted to the base of the model. I was suspicious of the position shown on the plan, so I have incorporated the facility to move the mains easily if needed.

The round-topped bulkhead seen protruding from the top of the fuse is key. The wing dowels into it from behind, and the hatch dowels into it from in front.

The ends of the longerons and lower triangle stock have not yet been cut away from the rear (right hand) end of the fuselage. They must be removed and an engine mount fitted to accommodate the Thunder Tiger Pro .46 and pusher prop.

First mockup looks pretty good.

There is no noseblock, and the top hatch has not yet been fabricated. The stab is shown with no elevators (so it looks unusually narrow of chord) and is simply resting on top of the fuse instead of slotted into it. But it still looks good, and if you squint a bit, you can see how it might finish up!

The noseblock is fabricated from balsa laminations, and for once, I can use the heavy balsa from my scrap box! The block remains in two pieces with the top (right) piece attached to the hatch and the bottom (left) piece glued permanently to the fuse. There is a hardwood block rebated into the lower piece, and tapped for a 1/4-20 bolt. You can see the bolt in place. Another ply plate is rebated into the upper noseblock to act as a doubler under the head of that bolt when it is screwed in from above. A short length of dowel acts as a locating pin, and it (and the hole it fits into) are clearly visible.

While the laminations were separate, I cut a tombstone shaped hole in three of them, and made up a matching mold for lead counterweights. These weights were cast in various thicknesses. This allows for the CoG to be fine tuned by swapping weights of different thicknesses. The weights can also be drilled to remove small incriments of weight. To remove a bit more, drill a bigger hole! When the hatch and upper noseblock are bolted into place, the weights are retained in a snug little home designed just for them. They won't shift or rotate, and if there is a slight rattle, a small piece of foam can be sandwiched in with them!

This photograph shows the construction of the stab and elevators, which is similar to the winglets. 1/8th balsa strips are used to frame out the stab and elevator and sheeted on each side with 1/16th balsa. This gives the same 1/4 thickness as called for in the plan, but the hollow center makes for lighter parts and the lamination theoretically make it stronger than the plain 1/4 sheet slab recommended.

Here we see the stab and left elevator without the top sheeting, but the right elevator, the end of which can just be seen, is already fully sheeted on both sides.

Two torque rods were made up from 4-40 pushrods, and fit into paper bearings in the stab and internal slots in the elevators. The two aileron links can be seen on the board along with the two 2-56 pushrods and clevises that will attach to the twin elevator servos.

To my surprise, the first attempt resulted in an elevator that was as stiff and strong as well masticated chewing gum! All attempts to reinforce it failed. Or looked too complex to be worth trying. I rebuilt the entire stab a second time, but instead of using only 1/8th balsa strips for the inner structure, I also added 1/8th square spruce spars running lengthwise, and glued three ways: Onto the top skin, the bottom skin, and to an adjacent balsa strip. The result was a very rigid stab that contained the spruce, but was otherwise identical to the one pictured above.

Here we see the two parts, supported at one end, and loaded centrally with identical lead bricks. The first one (in the foreground) has bent downwards well over an inch! The new one doesn't deflect noticeably, which is what I need, thanks. I am telling myself that it probably isn't any heavier than a plain, 1/4 slab, and probably is stronger. It certainly was easier to get the torque rods into it, and given the hi-jinx that the torque rods will have to go through to get installed in the fuse...

The elevators will be actuated by two torque rods built into the structure of the stab. The actuating arms for such a mechanism are usually at a nominal right-angle to the structure that carries it. This presents a problem for this aircraft.

This photo shows the two torque rods in the stab. The right-hand rod is in it's normal operating position. But the left-hand torque rod has been folded flat, to allow it to fit through the 1/4" slot in the front of the fuse. When the stab is finally fitted, both torque rods will be folded flat. After the stab has been glued into place, they will be folded up into their normal operating position, and a supporting plate and doubler (seen here) glued into the rebate visible in the area of the torque rods. This will leave two 1/8" square holes in the trailing edge of the center section. These are the slots into which the rods fold for insertion. Two 1/8" square balsa plugs will be inserted to close these.

The lengthy hatch was constructed in similar fashion to the fuse. It is simply three sides of a balsa box, with triangle stock in the corners, allowing it to be rounded later with the plane. You can clearly see six little ply tabs I added to locate the hatch sideways when it sits on top of the fuse.

The left hand (rear) end of the hatch will include a cockpit. The bottom of this cockpit is still open. I must make up a block on which I can mould a canopy. I've also got to create a pilot, given that Williams have gone out of business!

At the front (right) end of the fuse, you can see the slot into which the stab will be fitted. It can be seen more clearly in the next photograph...

This rather phallic looking object is the very front of the fuse, with the noseblock temporarily attached so as to check the fit.

The razor plane has been applied here to rough out the shape of the nose. As the fuse itself is rounded and shaped, the noseblock will also be worked on so the result will be a single, flowing shape. The laminations in the block are clearly seen. As is the dividing line between the upper and lower parts of the block, and the nylon fastening bolt. With the bottom block epoxied to the fuse and the top block to the hatch, that screw will mate the two parts of the noseblock into one, and thus hold the hatch onto the fuse in normal operation.

Behind the noseblock you can clearly make out the slot for the stab. It sits at a slight angle. Canard aircraft generally have several degrees of positive incidence in the foreplane/stab, and this aircraft is no different.

Using some old balsa scrap, I roughly fashioned this balsa box to sit in the cockpit. After some razor-plane and sanding-block loving, it will become (I hope) a plug for the canopy for this aircraft.

I hope to be able to use this plug to shrink up a neat canopy out of a plastic bottle. I have never tried this before, so disaster lurks...

The wing was joined with 30-minute epoxy on the face of the root ribs, and a single 1/4" dowel as a joiner-brace. To furthur strengthen the wing, I laid in a 2" wide strip of glass cloth, and laminated this in place with 30-minute epoxy as well.

Shown here after the epoxy has cured but before the rough edges have been sanded, the two servo-lead holes have had overlaying glass cut away. You can just see a piece of the rolled-paper wiring tube on the right, under the glue-brush.

At this late stage in the game, I realised my plans for strip ailerons (as shown on the original plan) would not work. The trailing edge of the wing was much too sharp to accommodate strip ailerons of any reasonable thickness.

I cut sub-trailing edge spar slots and aileron leading edge spar slots in the wing ribs, and let in apropriate spar members. The top spars are balsa, while the two bottom spars are 1/8" spruce for strength.

This picture shows the starboard wing with the sub-trailing edge spar at the top in place, as well as the two spruce spars in the bottom. The slot for the aileron top leading edge spar is empty but visible. I didn't want to glue that into place in the aileron yet due to it's being so close to it's companion in the wing. I thought I might accidentally glue them together, and certainly the close fit would hinder sawing.

Also in this photo is the port aileron removed from the wing and awaiting the attachment of the top leading edge spar.