Foamboard Wright Flyer
I like the classics, and have been wondering for a while how easy it might be to make the ultimate classic, the Wright Flyer. Everyone knows the Wright brothers, Orville and Wilbur flew the first sustained, powered, heavier than air flying machine with a pilot on board near Kitty Hawk North Carolina, Dec. 17 1903. They were bicycle makers from Ohio, actually from Dayton, not far from FliteTest headquarters, and were determined to make a flying machine that worked.
On Dec. 14 they flipped a coin to see who would fly and Wilbur won. But they set it up on a slope and Wilbur pulled up too much, stalled and damaged the Flyer. Three days later Orville had a go and managed to fly for 120 feet, which was the history-making first powered flight.
The foam-board version here is more or less to scale, but has slightly different controls. It has a canard elevator (a really awful idea) and a working rudder, but I gave it ailerons rather than wing-warping (though I can see how you might do that with paperboard wingtips). The finished product actually looked pretty good, and surprising to me it was dead easy to build, used only a single sheet of foam-board, and very tough when finished (you could almost sit on the wings the skewer system I made works so well). But does it fly?
The Challenge: Beating Orville’s Record.
Orville’s historic first flight was a distance of 120 feet. The Wright Flyer had a wingspan of 40’4”. My scale Wright Flyer has a wingspan of 30". Therefore, to beat Orville’s history-making 120 foot flight, my scale Flyer has to travel further than 90 inches. That should be easy, right? See the video below to find out...
As Josh would say, "Flies great!".
The wings are each one piece. Cut out two identical wings, but add ailerons to the one that will be the bottom wing (see plans). The wing plans have 8 marks where skewers will be inserted, make sure you mark these on the wing pieces before you assemble them (but don’t make the holes yet). Wing assembly is the same as other similar wings (e.g., the FT Spitfire). Install two servos (in the bottom wing only), bevel the leading edge, bend the airfoil, and glue down the spar, then glue and fold the wing.
Cut out two sets of struts, and reinforce them with popcicle sticks. Glue these to the but end of the bottom of the top wing where the under-camber begins as shown below. Then test fit with the bottom wing to make sure they are even (the two wings are parallel and the top wing should be slightly forward of the bottom wing). When you are happy with the fit, glue the struts to the lower wing.
Now it is time to add all the skewers, and the strength of the wing. I use big skewers (3 mm). Where each of the 8 holes are marked on the lower wing, push the pointed end of a skewer though roughly perpendicular to the wing. Once you punch through the bottom wing entirely, direct the skewer to the same hole-mark on the top wing, and push it through, but just the bottom layer of the top wing. Then pull out the skewer, flip it and insert the blunt end through the bottom wing and up to the hole in the top wing. Pull it back a bit, and put a bead of glue around the skewer just below each wing, then re-insert it, taking the glue with it into the holes. Rotate it to give a good tight bond and let the glue set. Do this for all 8 holes, and when they are all set, clip off the excess skewer flush with the bottom of the bottom wing.
The rudder frame is made from skewers and a bit of foam. To aid this part, I made a little jig from leftover foam to make a triangle that is 13 mm to the apex and 10 mm wide at the base. Place this flush with the trailing edge of the lower wing, centred, and mark where the triangle meets the wing and extend the line to the lower fold of the wing as well. This is where the rudder frame will join the wing. Now insert a skewer into the trailing edge following this line and flush with the bottom of the wing. It will go into the wing, then pop back out, before going into the foam sheet that makes up the bottom fold of the wing. When you are happy that it lined up right, pull it out, glue it, and re-inert. Do the same with a second skewer on the other side of the triangle. When they are both glued, line them up on the jig and chop off one at the end of the triangle, and then chop off the second at the inside angle formed by the fist so they make a nice clean triangle. Now trace the inside of the triangle they formed onto a scrap of foam and cut it out so it extends about 2.5cm from the end of the skewers, and glue this little triangle into place, and reinforce with some tape (see pictures).
Repeat all this with the top wing, making sure that the frame sits directly above the frame on the lower wing by sitting the wings on a flat surface and checking that the upper and lower sets of skewers are perpendicular to the surface. When this is complete, drill a small hole in the foam near the apex of the triangle that is slightly larger than the skewers you are using.
Now build the actual rudder, which is simply a box with two large sides and two small horizontal spacers. Make sure the spacers are just smaller than the space between the upper and lower ends of rudder frame. Glue the rudder hinge pieces onto the inside surfaces of the horizontal spacers (see picture) so the rounded part sticks out. When this is complete, insert a skewer through the four holes in the upper and lower rudder frames and hinge pieces. When it is in place, glue to ONLY into the two frame holes (NOT the hinge pieces). The rudder should move back and forth freely.
Elevator Canard Assembly.
I assumed the elevator was going to need to be able to sustain some damage because it sticks out front, and this was proven to be correct. It also looked to me like this was going to be a tail heavy design. So I built it to take a lot of force by using 2 mm wire. I used a length of stainless steel welding rod (easy to get and cheap, not as hard to bend as piano wire, but strong enough for this job). Bend 1100 mm (18”) of wire so it has two 150 mm extensions with a 100 mm (4”) section in the middle that is raised 35 mm (1”) (see picture).
This is attached to the lower wing simply by pre-poking holes 100mm apart in the leading edge of the wing with skewers, then inserting the two ends of the wire into these holes. They should insert so the raised section of the wire projects about 90mm (3.5”) in front of the wing and is parallel to the wing. Once you test fit, hot glue into place.
The canard is simply a box like the rudder with the two vertical spacers about 100 mm apart. To attach this to the wire, cut out two pieces of foam that are 3mm X 80mm. Set the canard on the wire, centred, and glue these on either side of the wire (as shown in the picture), making sure to keep the glue off the wire. Now cut out a piece of foam just big enough to cover these and the wire, and glue this onto place (again keeping glue out of the wire). The canard should move freely on the wire. If you have too much space and the canard flops up and down, then you can insert a short piece of foam or a small skewer into each end to tighten up the hinge.
To finish, I glued two chop sicks onto the bottom of the wires and onto the wing, as it looked better and gave the connection even more strength.
The ailerons hook up like normal. The rudder and elevator servos are glued onto the bottom surface of the upper wing (as shown). For the rudder I used a 5g, and 9g for the elevator, which I put into a cut out to make it look nicer. The rudder control horn is placed on the right side near the top hinge with the horn facing out and lined up with the hinge. For the elevator I put the horn inside the canard. This only works if you use a really high hole on the horn (too low had way too much throw and no holding force). Remember the elevator servo will be reverse what you normally use since it is up front. The servo wires were strung along the struts to keep it clean and the receiver mounted near the front. The receiver was an OrangeRX with the plastic case removed and replaced with shrink tube.
To finish I varnished the foam and gave it a super-quick wash of heavily diluted brown paint, which came out well. I added a pilot to balance the battery by zip-tying it onto one of the ESCs.
The Wright Flyer took off from a wheeled cart, so I also built one that you can see in the pictures and movies. I did not bother with instructions for this since you can do almost anything just as long as the props are off the ground. I tried touse a couple 1/2" sticks to keep this on track, which did its job.
Power set up - Thrust angle
The Wright Flyer had a single engine powering two large pusher props by a chain-drive. I considered doing this with inner-tube belts, but only for a few seconds. Instead, I used two small motors and this is where the problems start. Figuring out the thrust angle for a twin pusher biplane is not so simple, and I could find little information to help since it is a very unusual configuration. Added to this, the centre of gravity is a complex problem too. Here are my observations, based on a few hours of fiddling around, but this ultimately was not something I solved. I think it is solvable with the adjustable thrust-angle mount I describe below, but I lacked the enthusiasm to go through the process.
The power set up I tried is like this:
Motors - 1500 kv 24g bell type
Battery - 850mAh 3S, then later 500mAh 2S
Props - counter-rotating 8X4SF props that were cut down to 7” and mounted on prop-savers
To start, I mounted the motors perpendicular by adding another skewer and balsa post that was vertical when the plane was sitting on a flat surface (i.e. slightly different than the struts). With a 3S this was massively overpowered, but even on low power it had HUGE down-thrust. When cart-launched it just nosed over the front of the cart, and when hand-launched it did two fast outside loops before hitting the ground.
Adjustable Thrust-Angle Motor Mount.
The reason for this is that the thrust-angle was inappropriate for the centre of gravity, but of which wing? Counter-intuitively, if you change the motor angle to point down, it should reduce the down-thrust (because the thrust-angle was previously pointing way forward of the CG of the lower wing). I added some with washers, but it was not enough. I looked at the real Wright Flyer, but its props were so different it seemed irrelevant and I could not find specs for the working models of the Flyer you can buy, which may not have mattered anyway.
It became clear that the thrust-angle was something that needed experimentation, so I made mounts that were easily adjustable, which would be a generally useful thing to make anyway. For these I used metal, specifically a band of thin stainless steel that is used to attach chain-link fence to a post. It is easy to bend, but not too easy, and is about 3/4” wide. I cut 90 mm of this and bent it twice to divide it into three equal sections: one end has two holes for the motor mount, and the other end glues onto the bottom surface of the top wing. I glued a piece of foam over the attachment site for added strength (see pictures). Now you can try a thrust-angle and change it based on the flight just by holding the mount where it is attached to the wing and bending it one way or the other.
I initially tried a very steep angle to point the motors at the CG of the lower wing. This reversed the loops so it now had way too much ‘up’ and did inside loops even at 1/3 throttle and a 2S battery. By trial and error I moved it back and forth and it got better and better. Then I just got bored. Finally with an angle about half way between pointing at the CG and perpendicular I tried a flight to beat 90” just as a joke and it worked, so that was the end of my experiments. (and by the way, I am really kidding because I do realize Orville was not hand launched from a height of 150 feet.)
As a joke, I eventually made it into a quad too (see pictures). I figure this is what Wilbur and Orville should have done from the get go.
Bottom line & a challenge.
So this plane looks good, was fun to design, not too big (one sheet total), and easy to build. The controls look great and are interesting to set up, but the power system needs some attention. It is widely acknowledged that the design is marginal at best (the elevator in the front is a disaster since the airflow from the prop never hits it so you cannot use that to correct for the thrust-angle), but here we are constrained by history. It is possible it is too heavy to ever really work, but I doubt it (the wing area is close to a blender and it weighs about the same). So I thought the actual airframe might be adequate given the design limitations and I would describe it in detail here and hope that the Wright Flyer is sufficiently interesting that someone would build one and crack the thrust-angle/power problem. I think this will require a combination of:
- different motors (smaller, less power, lower kv, less weight)
- different props (smaller with less thrust)
- carefully figuring out where the put the CG
- reducing the weight can’t hurt, I used 15A ESCs, which could reduce a lot, or smaller battery (though the 500mAh 2S is pretty small already)
- getting the thrust angle right for the motor/prop combination