Bull Rider - Triple EDF Insanity!

Bull Rider

by DamoRC and DamoRC Jnr

Bull Rider is our entry for the Red Bull Air Race Plane DIY Competition 2018.  During the Edge 540 teaser video Alex asked a few questions including:

1. “What would a Red Bull Air Race plane look like in the golden age of aviation?” 
2. “What would a plane of the future look like?” 

Bull Rider is our attempt to answer both questions.

The Bull Rider Logo, the finished plane and ready for maiden pictures

Inspiration

Bull Rider was inspired by a broad mash up of contemporary planes, golden age racers, and futuristic craft.  We wanted the design to retain some recognizable element of the Edge 540 and incorporate features from some of our favorites like the De Havilland Comet, Caudron 460, and the One-man Light Jet from Tron Legacy.  

The plane would suggest extreme power combined with the elegance of the golden age. It would have steam-punk structural features and prompt thoughts of futuristic air-gladiators.  These ideas were distilled into a design that combined a vintage style fuselage and tail feathers with twin EDFs in oversized nacelles and a cantilevered Edge 540 wing.The plane would be piloted by a rider in an open cockpit and it didn’t take a huge leap of imagination to combine the rider concept with the Red Bull theme and come up with the Bull Rider moniker.

Clockwise from top left: The Edge 540, De Havilland Comet, Caudron 460 (Design by Rockyboy), and One-Man Light Jet - inspiration for the Bull Rider

Design

The plane was designed in Sketchup starting with a simple model of the Edge 540.  

Simple Sketchup Model of an Edge 540

We tried and failed to manipulate this model to produce a golden age style fuselage so we turned to pen and paper to rough out a fuselage shape that might work.  Eventually we found a shape that captured what we were looking for and this image was imported into Sketchup. The shape featured s-curves front and rear and worked well for the rider concept but the tail really looked like it should house an EDF thrust tube so one was added. This opened up the possibility of powering the plane with either twin 50mm EDFs in the wings and a fake thrust tube in the tail, or a single 70mm EDF in the tail and empty nacelles. In the end insanity ruled and we decided to make it a triple EDF!

Ink Drawing of the fuselage profile and the image modeled in Sketchup

The Edge 540 model served as the template for the plan view of the plane, having a narrow tail and a wide nose. The three dimensional fuselage was constructed using a vintage style former with a curved top and straight sides that tapered towards the bottom. These formers were distributed along the length of the fuse and sized to fit both the profile and plan of the plane. 

Once the formers were in place they were planked in-silico to create the 3D model of the fuselage.

Building out the three dimensional fuselage

The wing planform was basically identical to the Edge 540 so little additional design work was needed here. 

Edge 540 wing planform and Red Bull Livery

Next up was the cantilever structure for the wing.  10mm square carbon fiber tubing had been ordered for this build and the tubes were modeled in Sketchup. 

Two tubes would form the cantilever supporting the wing from the fuselage and two tubes would connect the wing to the cantilever and also serve as the spar for the wing.  Pieces of 1/8^th plywood were modeled to brace the carbon fiber tubes in a sandwich of plywood at the various joints. 

Modeling the cantilever frame in Sketchup in square section tubing and plywood bracing

The nacelles were modeled after the fuselage using a similar three dimensional shape and incorporating the same s-curves front and rear.  They were modeled so that their floor would be slightly above the floor of the fuselage in order that the wings and cantilever frame would not take the brunt of each landing. This failed to consider the possibility of landing in a tree or crashing wing low and assumed that the plane would fly and land multiple times!

Modelling the nacelles to match the style of the fuselage

The tailfeathers proved to be a little tricky to design.  We tried to maintain the ratios between the areas of the original Edge 540 wing and tailfeathers but incorporate the style of the fuselage and nacelles. The final versions look a little undersized in comparison to the wingspan and fuselage.

Tail feathers added to the model in Sketchup

Finally, using a picture of a motorcycle rider as a template, a helmeted rider shape was added and positioned in the cockpit. 

The Rider for the Bull

The final Sketchup model looked pretty cool but would we be able to build it and would it fly?

Final model in Sketchup from multiple angles

Preparing Plans

The wing panels and tailfeathers were modeled as simple flat pieces in Sketchup so little additional work was needed to print out plans for these pieces.  The fuselage and nacelles needed to be unfolded or flattened in Sketchup before they could be used to generate printable plans.

Unfolding one of the nacelles in Sketchup so that a flattened plan could be printed

Flattened pieces ready to be printed and transferred to foam board

The Build

The plane was constructed from Adams Readi Board foamboard aka Dollar Tree foamboard, 1/8^th inch plywood, 10mm square section carbon fiber tubes, hot and gorilla glues and assorted screws and fasteners.  Many of the pieces were made “FT Style” including the folded wings and tailfeathers. The estimated weight of the components and foamboard was approximately 4 lbs and we felt that we could generate sufficient thrust to power the plane. The wing loading was a little on the high side for our liking with a WCL of 19!

Cantilevered Wing Frame

This was the most difficult part of the build both in terms of its construction and later its fitment into the fuselage. The first version of the frame used the 10mm square section carbon fiber tubes sandwiched between plywood braces which were held in place using small screws and gorilla glue.  The square section carbon fiber was chosen as we thought it would be easier to construct these “sandwiches” between the flat ply and square tube.The initial construction relied almost exclusively on cutting the plywood braces accurately and lining the tubes up with the brace edges before gluing and screwing in place.  A paper printed plan of the frame was used for reference.  

Building the first version of the cantilever frame using plywood braces and 10mm square section carbon fiber tubing

Unfortunately this first frame proved to be a failure because it was mis-aligned and there was too much flex in the main cantilever tubes.  Specifically, the torsional weakness of the tubes joining the fuselage to the wing would likely lead to a lot of “flapping” of the wing.  

Flexing in the first version of the cantilever frame

The second frame build was improved in two ways.  First, the main cantilever tubes were replaced with 12mm round section carbon fiber tubes.  It was hoped that the round section and thicker tube would remove the flex. Second, the frame was assembled in a jig that comprised the bottom surface panels of the wings and a center piece that represented the fuselage width and gap between the wings and the fuselage.  These three pieces were taped together.  

We retained the 10mm square section carbon fiber tubes as the spars. Pockets to accommodate these spars were cut into the wing panels and the foam removed.    Nine inch lengths of the 12mm round section tube were cut and laid out on the jig so that their final positions could be marked.  Then these were also pocketed and installed in the jig.  Now the frame was aligned correctly with the wing panels we could start adding our plywood bracing. 

Before we started to fix the plywood bracing, servo extensions were passed through the center of the hollow tubes. 

Once the upper plywood pieces had been glued and screwed, the frame was removed from the jig and the underside plywood braces were added.  The second frame was significantly better than the first attempt.It was properly aligned with the wings and the torsional strength of the thicker round section tubes was much better with only a small amount of flex observed. Reliefs for the plywood braces were cut from the lower wing panels and the 10mm wing spars were hot glued in place.

Construction of the second version of the frame and the final result

Wings

Before adding the upper wing surface, the aileron servos, pre-centered, were installed in pockets cut from the lower surface and connected to the servo extensions in the cantilever frame.

The upper wing panels were score cut along the appropriate fold lines and were taped to the lower panels.The leading edge bevel was made by crushing with a blunt barbeque skewer, a technique that replaces beveling with a blade.  Because the lower panels were already glued onto the cantilever frame it was a little awkward positioning the lower and upper wing surfaces in such a way that the wing folding would keep the lower surface flat. The wings were folded in two stages.  In the first step, with the flat bottom surface on the bench, glue was added along the full length of the spar and the trailing edge section from the root to the start of the aileron.  The wing was folded and held in place until the glue set.  The wing was then flipped upside down so that the upper wing panel sat on the bench.  The remaining trailing edge from the aileron to the tip was glued and held down until set. The goal of this approach is to create a small amount of twist or washout in the wing. 

Wing construction, installing the spar into the lower surface panel, servos installed, and folding the wing

Tailfeathers

The tailfeathers were pretty straightforward FT style pieces.  Both the vertical and horizontal stabilizers were single layers of foamboard and each was reinforced with a strip of wood that was glued into a pocket relieved from the panel.  In the case of the vertical stabilizer, this wooden spar was extended beyond the bottom of the piece so that it could be passed through the fuselage providing additional strength.  An elevator was cut into the horizontal stabilizer FT style.  No rudder was included in this version of the plane.

Finished tailfeathers pieces dry fit together

Fuselage

The fuselage was made using a non-standard FT technique.  The fuselage pieces were cut and scored according to the plans generated from the unfolded parts in Sketchup.  The score cuts are widened slightly by running a barbecue skewer down their length. Then the score cuts are folded back to open them fully and a thin “shmear” of gorilla glue is applied to the exposed foam.  The part is then folded and the bottom edges are hot glued to hold the piece together.  The piece is fitted into frames that are cut from foamboard which accurately match the shape the part should have from the Sketchup design.These frames ensure that the part holds the correct shape while the gorilla glue dries over a period of two to three hours. This approach provides surprisingly strong sections that are totally hollow.

Building the nose section of the fuselage by adding gorilla glue to the score cuts and mounting in a foamboard frame

Main and tail fuselage sections folded and framed

When the individual fuselage sections were in frames they were dry fit to each other to ensure the interfaces between sections are true and aligned.  Once happy with the dry fit the sections were hot glued together.  When the gorilla glue had dried the frames were removed

Fuselage assembled in frames and a view from the nose showing the hollow core

Nacelles

The nacelles were constructed using the same techniques used for the fuselage.  In addition, cut outs were marked and removed from the upper surface to allow the nacelle to interface with the wing.

Both nacelles glued and framed

The Rider

The rider was a simple profile shape that was cut from two pieces of foam which were layered together with hot glue.  The width of the base was extended with additional foam pieces to allow a bigger surface to glue the rider into the fuselage and prevent him from being sucked into the tail mounted EDF!

The Rider prepped for painting

EDF Preparation and Installation

The thrust tube for the 70mm EDF was made from acetate sheet using a template produced in Sketchup.This acetate material works well for thrust tubes because it is relatively light and is easy to bend into tube shapes.  Then the unit was glued into a foamboard frame that fit into the tail of the plane. After making some reliefs for the motor wires and the elevator servo wire, the unit was hot glued in place.

70mm EDF thrust tube and installation into tail

The thrust tubes for the 50mm EDF units were also made using acetate sheet.  Motor wire extensions were added and then the units were glued into a foamboard frame that was hot glued in place inside the nacelle. When the 50mm EDFs were glued in place, the nacelles were glued to the wings.

50mm EDF thrust tubes and installation into nacelles

Wing Installation

In addition to constructing the frame, this was one of the more difficult parts of the build.  When the wing was installed in the planned position we realized that we would not be able to balance the plane at the CG of approximately 2 inches from the leading edge.  Adding two extra 2200 mAh 4S batteries (for a total of four) in the nose of the plane still failed to balance it on the CG. Solving this problem required ripping out the installed EDFs and re-positioning them as far forward as possible. In the case of the 50mm EDF units the new position was at the front of the nacelle.  For the 70mm EDF this meant pulling the EDF forward a couple of inches. This was as far forward as the 70mm EDF could come without starting to chew on the rear end of the rider!  Once the EDFs were re-positioned, the wing was moved rearward in the fuselage until the plane was easy to balance with just two 2200 mAh 4S batteries in the nose.  When this position was located, the wing cantilever frame was mounted to two pieces of wood that spanned the fuselage. The cantilever frame was fixed in place with a screw and copious amounts of hot glue.  Once installed, the wing was solid and very little flex was observed in the wing when the full weight of the plane and electronics was supported by the cantilevered frame.

Close up of cantilever frame installed in the fuselage

Tailfeathers

The tailefeather assembly was pretty straight forward FT style with a slot cut for the elevator servo and a channel for the servo wires cut in the vertical stabilizer. A servo extension was added and threaded through the 70mm EDF bulkhead then the tailfeathers were glued in place ensuring that they were square with the fuselage and wings

Tailfeathers finished and installed

Paint Job and Decals

The plane was spray painted with rattle cans in a traditional Red Bull Air Racer style with a yellow nose and blue air frame.  Yellow was also added to the font of the nacelles and to the open cockpit.The decals were based on images downloaded from the internet and scaled in MS Paint to match the sizes that were estimated from the Sketchup model.  Once printed and cut, these were attached to the plane using a glue stick.

Painted plane awaiting decals

The Finished Plane

The rider was spray painted black and aluminum tape was used to make the visor. He was then hot glued into his rider position.

Ready to ride the bull

Electronics

The electronics were situated at the nose of the plane in order to keep as much weight forward of the CG as possible.  Two 60 amp ESCs were attached via Velcro to the side walls of the fuselage for the twin 50mm EDFs.  An 80 amp ESC was mounted to the roof of the fuselage for the 70mm EDF.  14 gauge silicone wire was used throughout for motor wire extensions and to create a custom wiring loom to connect the two 2200 mAh 4S batteries to the three ESCs.  Velcro was added to the floor of the fuselage at the nose to hold the two batteries in place.

Electronics and batteries installed - not a pretty sight

Maiden Flight

The maiden flight was short, too short. Eighteen seconds to be exact. 

The plane launched well with wings level and climbed out gently on full throttle.  No trim was needed. 

Maiden Launch - pretty big chuck to get all 4 pounds into the air

However, after making the first turn toward the flight line, about 15 seconds after launch, all power was lost and the EDFs and control surfaces would not respond to the transmitter.  The plane then continued to glide beautifully into a tree at the side of the park.  

Damage looked a lot worse than it actually was

Short and sweet!

Luckily the plane had landed about ten feet up in the tree and a fearless teenager climbed the tree to recover the plane (thanks Drake!).  Being able to recover the plane this way meant that less damage was inflicted than if we had started to throw sticks and rocks at it!

The damage report was not too gruesome.  The tail had broken off pretty cleanly, there was some superficial damage to the nose and to the front of the nacelles.  The only worrying item was a dent in the leading edge of the right wing.  Although it looked small, the wing felt a little loose and we were worried that we might have cracked the cantilever frame. 

Tail came off cleanly but there was a nasty dent in the wing leading edge.

A more thorough examination of the wing made following some surgery revealed that the frame was intact and that it was the glue / foam joint holding in the spar that had given up.  Some fresh hot glue fixed this pretty quickly along with re-attaching the tail and repairing some of the other cosmetic damage. She was ready for her second maiden.

Second (and a half) Maiden

The second maiden was also short, but a whole two seconds longer than the first!  This time it was a different beast to fly.  As soon as the plane was launched it became apparent that it was very twitchy and had a severe roll to the left which I couldn’t trim out immediately as I fought to keep it in the air.  Clearly the initial tree-landing had done more damage than originally thought. On pulling up following a high banked turn the plane tip stalled to the left, dropped further, at which point I pulled up even more and it tip stalled again, this time hitting the ground.

A quick examination confirmed that there seemed to be only minor bumps and bruises so we tried again.This time we managed to keep the plane in the air for just under three minutes but it was tough going.  The landing was a little rough, it looked like the plane grabbed the ground with one nacelle or wing tip.  Ultimately the mounting for the cantilever frame gave up and was beyond repair.  The upside was that we got to see that wonderful silhouette in the air against the dull clouds!

In flight

Silhouette in the air

Conclusions

We set out to design and build a mash up between and Edge 540, golden era racers, and planes of the future.  Although this first build of the Bull Rider is no more many valuable lessons were learned and we are confident that we will re-build.  Future versions will be either twin 50mm EDF or 70mm EDF but not both.  The wing will again be separated from the fuse, but in the next version this will be achieved using a simple spar mounted straight through the fuselage. The revised plane will be significantly lighter and hopefully will fly better. 

This project was a lot of fun to design and build.  

Thanks Flite Test!

DamoRC and DamoRC Jnr

If you don't like to read, here's 15 minutes of your life that you will not be getting back...