We’ve all seen the Red Bull events where they get people to build ridiculous vehicles and race each other, the Flugtag (German for '“flight day”) is really cool and also the Soapbox race. I was super excited to learn that Red Bull was hosting a Flugtag in Toronto in 2022, and my friends and I couldn’t pass on the amazing opportunity to participate :)
I’ll start from the end, we did participate, we built the wing and the cart, we put together the costumes, we did the dance in front of the whole crowd, but things didn’t go as expected (in a way, they went better than expected, maybe). See it for yourself:
On the final run down the launch pad, we accidentally dropped our wing, somehow we never thought to rehearse making sure our pilot was well strapped before pushing the cart with all of our might. It was “as bad as it can get”, we lost “the one thing we needed to remember”, but even though we lost our instruments we kept playing music, “the show must go on” (10/10 live commentary from Red Bull). At the end we scored an 8/10 from Commander Chris Hadfield. Funny story, I had previously met Chris to pitch him Watfly and he didn’t like it one bit, so this felt like a win to me.
The team (shown above) was my friend Manit aka ‘Pine’ (met at Tesla, went to JPL, now running his own startup, has built a bunch of crazy things), Ignacio (insane guy, runs Tiny Mile, has jumped out of an airplane more times than everyone I know combined), Bryan (met on freshman year at Waterloo, was there for the construction of all of the submarines), and my gf Marisa (whom I had just started dating a couple of months prior yet bravely agreed to be part of all of this). In my mind, this group might as well be called the Guardians of the Galaxy, pretty sure they could get anything done while having fun doing it.
So that was cool, probably the highlight of my summer. I could write about a bunch of things I learnt throughout this build, I think there’s a lot to be said about gathering a group of people and convincing them to work towards a particular goal, specially when there’s no reward other than “it’ll be cool” (and you have to spend all your weekends + pay for the materials). But keeping on trend with this blog, I’m going to talk a bit about how we actually built the thing and if anyone cares about the other stuff, you can ask.
It all started with a sketch, this was part of the Red Bull application that got us accepted into the event (it isn’t Harvard but you do need to prove you are motivated and competent enough). We spent only a couple of nights planning and got some CAD going, mostly just to make sure we were rule compliant in the overall dimensions. A lot of the “smaller” details (like, how to attach the wing, or how will the pilot will hold onto the wing) were left for later to be figured out during the build.
This was entirely recreational and everyone in the team had jobs, so we only worked on weekends, the first day of building was August 20, 2022 and the race was on September 24, 2022 (5 weekends total). I also employed my Mazda to transport the materials, which had previously transported the wing molds for Watfly, and the fridges for Ninja (this car will continue to transport random things until it falls apart).
We got some early wins super quick by building the cart, and left the more complicated wing for later. In general I usually try securing the easy wins first, this is the opposite of the “Monkey first” approach preached by Google but creates momentum and gets the team motivated to keep going (smarter people than me say this is the wrong approach, so maybe listen to them). These simpler parts were also much more flushed out in CAD and requires much less thinking and problem solving when actually building them.
The idea for the wing was simple, 2D NACA airfoil cutouts from Home Depot’s finest insulation foam, and two hollow aluminum shafts running across acting as spar elements. Rules constrained the total wingspan and then we made some design choices to complicate our life, including a wing sweep and winglets at the tips. The swept wing meant at least one of the spars had to run diagonally, which complicated the joint of the spars at the wingtip, and also at the middle of the wing (we anticipated that the wing would be too big to fit thru a door, or in a truck, so it needed to be separated in the middle by design). The swept also meant that the overall surface area of the wing could be slightly greater, drag (theoretically) slightly lower, and it’d push the center of pressure backwards (the wing profile we were using had an upward aft tip which needed to be counteracted). Same idea with the winglets. Plus a healthy amount of engineers flexing their aerodynamic knowledge.
Pretty soon the wing was too big to fit in the room and in my iPhone camera frame. There is something special about building things that are larger than yourself, bigger is just better. In the pics above you can also see that we started building the winglets, these were attached at an angle to the main wing so couldn’t use the aluminum spars. We also started adding wood elements in the leading and trailing edges, plus top and bottom of the wing, these would help secure and shape the outer skin of the wing later.
Soon it was time to start tackling the details that we didn’t flush out in CAD. The main concern was how would the pilot hold onto the wing, this was the most structural component of our vehicle: wing needed to be able to support the whole weight of the pilot without snapping (assuming we generated enough lift to achieve flight in the first place). This involved mainly looking around for scraps, bolting it together, and agreeing that it was “good enough” (don’t worry, we did do full load tests later, no chance we were going to fool ourselves in the event, keep reading).
More pictures of wing details. Attaching the winglets was a feat, every time you have a joint at an odd angle which you need to attach with some degree of precision and not a lot of tooling, things get interested. Results were good, I think we did pretty good all things considered (accuracy was measured by the eye). At this point we also started testing the full cart + pilot + wing stack, as there was a maximum height limit rule. We also did some trials of pushing the cart with the pilot holding the wing on top, and it became obvious that it was really hard for the pilot to both: 1. hold onto the cart + 2. hold onto the wing at the same time, so we added some extra wing supports.
Here are those extra supports we were talking about, the wing could now rest on its own without the pilot needing to hold it. In hindsight, maybe this wasn’t the best idea? Maybe it was part of the reason why we started running without realizing the pilot wasn’t holding the wing, and we dropped it in the main event. Just maybe, at the moment it felt right. The wing was huge, and although it looked light, holding such a big thing while you are getting pushed on a cart that’s not the smoothest ride ever, is hard.
Here we tested the wing structurally with full load, first attempt had a bit of a snap, second attempt was solid. Jokes aside, this was a full aluminum tube structure, it was not hard to make it strong enough to hold a person. The main concern was around the joint between the two sections of the wing, particularly the front one where the spars met at an angle and were joint by flanges screwed into a block of wood (totally fine in ideal compression, but needed to ensure that under full load, compliance wouldn’t cause a weird load case).
Here’s me admiring the wing as we approach the finish line. The next step was a first for the team, we were going to wrap the wing in 2.97oz Dacron, which is basically a special polyester fiber by DuPont. This material is commonly used in para-gliders as it’s airtight and durable, but also shrinks when you apply heat to it so it makes it easy to achieve an ideal, tight fit around your wing (for optimal aerodynamics).
Few things we struggled with when applying the Dacron and compressing it: First of all, when it starts to compress it puts stress on the wing structure in ways that we did not anticipate, you can see in the middle picture above how the wing cross section is curling, and the trailing edge appears to curve, this was not intended and in certain areas we feared it’d break the foam. Another complication was around attaching the Dacron to the wing itself. In the middle sections where the fabric could rest on cross sections and orthogonal members, this was fine and the aerodynamic loads would probably transfer adequately. But near the open edges (winglet and middle section) we didn’t have anything to attach to and end up improvising with hot glue (in general, we steered clear of things like staples and nails, since we had to design for a scenario where the wing crashes violently on top of the pilot). The last complication was around poking parts of the structure through the Dacron, I suspect with careful planning this would not be too problematic but when improvising with scissors, the results are not super clean.
Also, if you are thinking DuPont is just selling you polyester under the Dacron brand to charge you more, that’s not exactly true. At some point we ran out of Dacron and we had to run down the street to a tailor to buy polyester fabric from them, it didn’t behave anywhere close to Dacron (although it was much cheaper).
After we finished wrapping the wings (not a perfect job but learnt a lot from it), the next step was decoration, costume building, and coming up with a dance. After all, a big part of the Flugtag is showmanship and winning the crowd. Not going to include the dance rehearsals in this post, this isn’t that type of blog.
Once everything was done and the team was prepared, the last step was to get it out of the door and drive it into the event side. Unfortunately, none of it fit thru the door so it had to be disassembled and reassembled in site. This wasn’t the fun part of the project, just the tedious last push to get it done (every project has this part). We got everything disassembled and into a UHaul, and then drove it to site the next morning, and reassembled it. Eventually the entire team + backup came and we put the final finishing touches. We were stationed in a parking lot along with all of the other teams, pretty much everyone was working on assembling their vehicle and putting up the final touches, the energy was great, Red Bull really knows how to throw a good party.
And of course, the rest is history, you know what happened next :) The last thing I’ll add is that jumping from the launchpad, which is like 30 feet, into cold water, felt like being reborn. First the water slaps you hard, it’s cold, it’s dark, you don’t know what’s going on. You come out of the water and everyone’s looking at you. That’s gotta be what birth feels like, right? I don’t remember.
