TBBT: Missile Defense

The second challenge of The Big Brain Theory was a missile defense problem – given 5 days to build, and $20,000, how do you shoot a 20 pound foam projectile out of the air before it explodes, spraying goo all over you and your team in a steel bunker? Check out two ideas here:

 

Interestingly, they introduce this challenge by mentioning the Star Wars program, an initiative to create a national missile defense system that would protect against incoming ICBMs. In its time, the program was considered completely unrealistic and a near-total failure. Foreshadowing, much?

Before I get into the engineering, I want to take a moment to address what this episode didn’t show, and by and large, didn’t even allude to. At face value, this episode paints me as the person who did all of the design for the entire structure, who ham-handedly forced everyone into fabrication and dull layout roles, and who is entirely to blame for the thing not working. What they don’t show is the fact that everyone laughed and exclaimed when they saw the design I proposed to the judges, and the group was consulted and agreed that we should move forward with the propeller design together – even given a relatively low probability that it would work. What’s also not shown is the division of engineering and design that did occur; Amy and Alison took on design of the release mechanism, pan joint, and base frame, Joel prototyped the triangular head section and determined the optimal size of the propellers given our construction, and Tom took on the full design of the ammunition (including the ultimate selection of propeller blades). My role ended up being designing the powertrain and drive, spec-ing out and installing the pan/tilt actuators, integrating everyone’s design into one functional system, producing drawings that we could use for fabrication, and keeping everyone on-task.

The show also didn’t show the voting process (where we used decision matrices to tally everyone’s reactions, and all voted as a team) that we used as a group to select Tom’s proposal of windmill propellers over airboat, plane, or fan propellers. It didn’t show me reacting to Amy’s request for more work by, in fact, giving her more design work to do (something she talks about here). It didn’t show us reviewing Amy and Alison’s release mechanism design and approving it as a group. It didn’t show just about everyone contributing to the fabrication (including yours truly, though I certainly did less than others). One final, really sad thing it didn’t show is the team being excited to work on the project in the first place! We were trying something really hard, we were doing something creative and off the beaten path, we were generally getting along and being productive, and we were having fun. None of that came through, sadly – I guess it doesn’t make for a good story. Do those capes, hats, goggles and smiles we were wearing on competition day look like they came from a team that was full of drama, or that didn’t like the process?

Blue Team Outfits

What it did show was me coming up with a wacky, off-the-wall idea in the first place, that was significantly different from any other proposal that was made in the blueprint challenge, and being chosen to lead a team because of proposing that wacky idea. Given that the goal of the competition, repeated again and again by Kal Penn, Mark Fuller and Christine Gulbranson, was to find the most innovative engineering in America, I didn’t feel bad about proposing something out of left field. The episode also showed me insisting on a structure for decision-making and design, and insisting on a process to replace the chaos that had been the previous challenge. This, by necessity, meant that everyone had less input, but it also meant that we got things done under budget and ahead of schedule. I’ll definitely admit I was pushy in the introduction to the challenge, unreasonably condescending in a couple of moments, and unequally distributing some of the work, and for that I apologize. I don’t think it deserved the one-sided portrayal of this particular episode, though.

But now, on to the engineering. We ended up building something that looked like this:

Whirligig Launcher

The fundamental idea was to spin a propeller on a shaft until it produced enough lift to fire itself, and let the propeller go once the missile was in range – essentially, creating a giant hand helicopter toy. Since we knew exactly where the cannon would be, and more or less what the ballistic trajectory of the missile would be, all we needed to do was sweep that area of space thoroughly and make sure nothing could get through (which is exactly what ended up happening with both of Red Team’s ‘successful’ shots).

Hand Helicopter

Of course, it’s never that easy. The device never fired, as shown on the show. The judges believed that the propeller ammunition was simply too heavy; while that certainly could have contributed, the Blue Team thought it was much more likely to be due to the release mechanism not letting go, and to a lesser extent too much friction building up in the drive mechanism. We based this assumption on test fires (that weren’t shown) where we demonstrated enough lift to stretch an 1/8″ steel safety cable that was anchoring the ammo to the launcher by half an inch or more – enough to separate the propeller round from the hex that drove it at its base. I want to talk through some of the engineering decisions that we made that led us to this belief, and why we made them the way that we did.

One of the first major decisions was to decide on a type of propeller blade to power the system. Tom gladly took the design of the ammunition on, and came up with a number of options. The first option was to use a set of airboat propellers; these would certainly generate all the force we would need, and were incredibly robust. Unfortunately, they cost over $1,500 for a set, and all the suppliers we could find wouldn’t guarantee that they could get us a single one within the time frame we needed – much less the 5 that we were asking for. The same story was true for small plane propellers – they were extremely expensive, and likely not available. At one point we even considered buying Big Ass Fans (not even kidding about the name) and stealing their blade sets, but got extremely concerned about their weight, resulting kinetic energy at top speed (we calculated that one metal blade flying off at full speed would go straight through one side of the Red Team’s bunker and out the other) and similar cost to regular propellers. The cheapest, safest, most readily available propellers we could find were windmill turbine blades; they were made of fiberglass, we could have dozens by the next day, and they only cost $300 a set. Unfortunately, they’re not particularly designed to produce thrust; in the end, Tom had to modify them by changing their camber angle with angled shims. We’re sure this reduced our efficiency, but as I said, we did prove to have a significant amount of lift available.

The next big hurdle we faced was designing a release mechanism, also known as an escapement. How do you hold on to this propeller when you’re spinning it up, but release it at just the perfect time? Amy and Alison took the lead in designing this portion of the project, and came up with a design the whole team thought was reasonable in the brief design review we had time for; they proposed a set of jaws, geared together, that would clamp down on a ball bearing on the ammunition shaft and spring open when a pin was pulled.

Release Mechanism Unfortunately, this design is one of the least effective release mechanisms we could have implemented – the springs attempting to open the jaws are actively fighting the friction from the side of the bearing pushing against them, due to the force of lift from the propeller, and the more force we put on the propeller, the worse the problem gets. Worse than that, though, none of us really caught on to this problem until it was too late; we had all looked at the design, we all gave it the thumbs up, and it got made. I don’t blame Amy and Alison for it in any way, since we all looked at the design and gave it a thumbs up without much hesitation.

All that said, the design certainly had other problems. The shaft was extremely long, and ran the risk of torquing inside the barrel before it fully released. The rounds were, as mentioned by the judges, relatively heavy. The propeller shaft was driven from behind by a hex feature inside a socket, and it’s possible that a significant amount of friction was building up in that interface and in the sleeve itself. We managed to trip the generator by ramping the motor speed up too quickly, a problem which had never happened before in our testing. If I had to do this particular design again, I’d make a number of changes to address all these little problems, including:

  • Decreasing the shaft length and proportionally decreasing the propeller size, to cut down on balancing issues, reduce weight, and potential torquing inside the sleeve.
  • Driving the rotation of the propeller with spur gears located near the front, to allow for a much lower resistance to linear motion.
  • Attaching a magnet to the back of the shaft on a bearing, and holding the shaft in place with a strong electromagnet on the chassis, so that we would have an electric release mechanism with much less potential for binding.
  • Using proper lift-generating blades such as airboat or propeller blades.
  • Properly programming a slower ramp into the motor acceleration to reduce current draw.
  • Potentially fire the propeller by means other than lift, so that the spinning blades could still effectively sweep the target area without being relied upon for force generation.

Of course, if I had to propose a design from scratch all over again (and if I had known the budget was $20,000 in the first place), I might just propose buying an army of baseball pitching machines and building pan/tilt mechanisms and control systems for them.

I do want to make one final note, though this has gotten long – I do want to talk briefly about Joel. It’s true that the ultimate, top-level design of the propeller-firing device was mine, and I accept responsibility for that. It was an ungainly, unlikely design, but if it worked it would’ve been one of the first devices of its kind to ever be on television. It was different, it was a difficult engineering challenge, and it was fun to conceptualize and work on, but in the end, it wasn’t a great answer. That said, I think the big, un-aired reason the team didn’t universally suggest throwing me out was because the build ran so smoothly – we got things done as a group, we voted on details of design, we finished 3 hours ahead of schedule, and people had fun while doing it. It certainly sucked to lose, but the process we went through was sane and relatively free of drama. The reason I mentioned Joel explicitly at the end was because, of all the team members on Blue Team, he alone carried a contagious bad attitude much of the time, and a disregard for the quality of the work he wasn’t excited by, which actively hampered the team’s progress. That’s not to say that Joel slacked off, or didn’t do a great job in certain key ways – the welding he did on the main shaft sleeve was amazing, for instance, and I’m not sure any of us could have pulled it off without him. It was just that, when it came time to pick someone, he was the person that I (and other members of the team) felt was the weakest for occasional, very specific anti-social and anti-productive behavior. Honestly, I like Joel as a person – I think he’s hilarious, and he has very solid design insight to offer. In this case, though, he just wasn’t acting as the team player we needed him to be.

I will say that one of the only regrets I have from this whole competition show is not seeing this thing fly. I’ve often considered redesigning and rebuilding it, if only to watch it launch.