Budding rocket scientists from the University of Cape Town build and launch their own mock satellites.
Space, to borrow a phrase from Star Trek, is the final frontier. However, without highly skilled individuals capable of pushing technological boundaries, humanity is never going to move off this rock we call home. Fully aware of this, two student teams from the University of Cape Town build and launch their own mock satellites. While adopting completely different approaches, they endeavour to achieve the same result.
A half-century since its inception, space travel may still seem like the stuff of fantasy; but designing a satellite and launching it into space has to deal with real-world issues such as size, weight, time and budgetary constraints. Juggling these often conflicting variables – not to mention co-ordinating a multi-disciplinary team of specialists –to produce a successful outcome can get complicated. No wonder it’s called rocket science.
Without satellites, we wouldn’t have a global positioning system. Weather prediction would be even more imprecise than it already is, and we’d have none of those breath-taking images taken from space that highlight the beauty of our planet. To say we’re on the cusp of something truly phenomenal, with many yet-undiscovered benefits from space programmes still to come, would be an understatement. So it stands to reason that we need to invest in bright young minds, providing them with solid foundations to push our understanding of space technology into realms as yet unfathomed.
Fully aware of our dearth of local expertise, the University of Cape Town (UCT) has initiated an Introduction to Space Technology course, specifically geared towards senior undergraduate and first year postgraduate students keen on pursuing careers in this exciting field.
Astrophysicist Dr Peter Martinez, chairperson for the South African Council for Space Affairs (SACSA), presents the course. “This is such a specialised area that there are very few space technology experts in any of our academic institutions, so my aim is to foster good communications between universities, industry and government research institutions,” he says. “This course is a perfect example of this type of synergy.”
Now in its fourth year, the course incorporates both theory and practice, giving students an holistic taste of what to expect once jettisoned into the workplace.
Guest speakers, all international experts in their fields, lecture on a diverse range of topics from space weather to liquid and solid rocket propulsion systems, space law, orbits and astrodynamics, space debris, telemetry and much more. To round things off nicely, American space tourist Dr Gregory Olsen, who spent time on the International Space Station in 2005, delivered a lecture this year titled From Entrepreneurship to Spaceship.
On the practical front, the students, drawn from UCT’s various departments, are separated into teams. The plan is twofold: first, it groups different skills and interests together, giving them the collective ability to problem-solve. Secondly, and just as importantly, it gives them exposure to the kind of group dynamics typically found in multi-disciplinary technical teams.
The teams have to design and build their own CanSat, a small, pseudo-satellite that incorporates many of the components found in the real thing. Then they get it to perform various autonomous tasks.
Each CanSat must have the maximum weight (350 grams) and dimensions of a standard cool drink can, hence the name. Inside, the teams have to pack in enough hardware and electronics to collect live telemetry (acceleration, velocity and altitude) data and transmit it to the ground in real time, as well as develop their own ground systems to receive the transmitted data. The data also had to be recorded onboard for downloading at a later stage.
Once ejected from the launch rocket’s payload bays at an altitude of approximately 1,1 km, the CanSats have to navigate themselves autonomously to a predetermined location on the ground. The examiners, in this case Martinez and the head of telemetry at Denel’s Overberg Test Range, Japie Venter, pick the location just a few hours before launch. That forces the teams to program the necessary co-ordinates on the fly.
How the CanSats arrive at their destination depends purely on the ingenuity of the teams. They can deploy either glider or parafoil configurations, even fall from the sky, hit the ground and drive; whatever they do, they just have to get there.
“This course highlights the difference between engineering something on paper and making it work,” explains Martinez. “The students end up learning lessons they didn’t even realise they’d learnt from an experience like this.”
This year, two three-member teams took up the challenge:
Team Excelsior, made up of electrical engineering’s Dayne Kemp, maths honours student Lior Neu-Ner and mechatronics specialist Geoffrey Kilpin.
Team NitiKnight, comprising electrical engineering students Andrew Nicol (who, incidentally, already has a computer science degree under his belt) and Karthik Rajeswaran, along with William Tipping- Woods who’s busy with his Master’s in mechanical engineering.
Both teams were given a month and a measly budget of R3 000 to complete their projects. Apart from the seemingly impossible timeframe, their funds – expected to cover everything – had to be stretched to the limit. “The short timeframe and small budget given to the project are by design,” says Martinez. “We want them to become aware of the pressures involved when preparing for a specific launch window, as well as the very real financial constraints all projects face in industry.”
Adopting the KISS (keep it simple, stupid) approach, Team Excelsior opted for a parafoil design to steer their CanSat to the landing site. Kemp explains: “We eliminated the glider option because of our lack of experience and time constraints. Instead, we chose a small parafoil because they’re so simple to control. All you need do is pull one line in while releasing the other side, something we could accomplish using one actuator. And, by buying one from a hobby shop, we gave ourselves more time to concentrate on the electronics. All we did was go for the most logical, simple method available, really.”
However, things are rarely as simple as they initially seem. To steer the parafoil, the team ended up having to write some pretty intense algorithms. “We soon discovered that, when we pulled the parafoil’s right chord to turn right, for example, we had to stop the momentum early by applying a little left on the actuator, otherwise it would quickly oscillate out of control.”
For the CanSat itself, they chose a laser-cut Perspex spine and circular ends (giving them plenty of space to mount the electronics and battery), and a cylindrical carbon fibre enclosure to protect and strengthen it. The main printed circuit board, mounted to one side of the Perspex spine, contained the actuator, 16 G accelerometer, gyroscope, magnetometer, pressure sensor and a 90 dB buzzer so they’d be able to find it if it landed in dense vegetation. An IMU (inertial measurement unit) analyses inputs from all the sensors to provide spatial awareness and orientation should anything go wrong with the GPS.
As for the GPS module, its circular circuit board was made by hand, and then positioned on top of the CanSat to ensure clear satellite reception. Says Kemp: “Carbon fibre isn’t very permeable to electromagnetic waves, if we’d positioned the GPS inside the casing we wouldn’t have been able to get a fix.”
Why do simple when you can do complex? Nicol explains: “There’s no established way of doing this, so we wanted to do something that was completely novel and innovative. We wanted to push our own boundaries.”
Inspired by the Mars rover’s landing system, they wanted to create a deployable payload. Their first plan was to construct a quad-copter out of Nitinol, a memory alloy made from nickel and titanium. “We came across Nitinol wire and had no idea how to use it,” says Nicol. “This project was a great opportunity for us to try something interesting with the stuff.”
Unfortunately, after much research, they realised their “ideal” system was becoming way too complicated. So instead, they decided to build a glider, with Nicol concentrating on the electronics and Tipping-Woods working on the mechanical design.
Next complication: they soon discovered that it was nigh on impossible to fit the wing inside the rocket’s payload bay, so it ended up another scrapped design. Fortunately, though, they had achieved one breakthrough: because they needed to keep weight to a minimum for the glider to fly, they were obliged to pack the same amount of electronics as Team Excelsior into a package weighing a mere 60 grams. “Technically, the electronics could have fitted into a matchbox,” explains Nicol.
With only one week left before launch and time starting to run out, they decided to do the tried-and-tested thing – that is, build an aircraft. However, controlling an aircraft autonomously is a complex affair, so Tipping-Woods fell back on the tried and tested: keep it as simple as possible. Using balsawood, he made its centre of gravity as low as possible and gave it highmounted dihedral wings for extra stability.
Not wanting to give up their dream of a deployable system, they used Nitinol wire to connect the wings to the fuselage. Once their payload exited the rocket, the plan was for its folded wings (held in position at their tips by a band of paper connected to a small parachute) to spring into position to allow for stable flight.
Because they didn’t want the flying controls to activate while the CanSat was still inside the rocket, they fitted a photo LDR (light dependent resistor) to tell it when it was outside. Only then would their system take over the autonomous tail and rear elevator controls. They also included an ultrasonic ranger so it could determine the distance to the ground and pitch up its nose just before landing – preventing it from ploughing into the ground.
Come launch day, the weather forecast wasn’t promising. The first big cold front of winter had just hit the country, bringing with it heavy snow and flooding that forced the N1 and N2 national roads to be closed.
But no one was going to allow the lousy weather to detract from the occasion. Bundled up to keep out the cold and driving rain, both teams, along with a gaggle of enthusiastic supporters, waited for a break in the clouds over Denel’s Overberg Test Range, situated near Bredasdorp in the Western Cape, for an opportunity to put their projects to the test.
While waiting, Martinez commented, rather prophetically as it turned out, “Even if it blows up on the launch pad, both teams will have gained a tremendous amount of experience just reaching this point.”
Finally, a break in the weather brought on a flurry of activity. After an impressive whooooosh, the launch vehicle, accompanied by a billowing tail of exhaust gases, rocketed flawlessly into the sky.
Unfortunately, instead of reaching an apogee just over a kilometre high as planned, it levelled out at about 800 m.
Because of its shallow trajectory, the rocket’s three parachutes didn’t stand a chance. The momentum ripped all three off their attachment points, followed by the rocket and its two-CanSat payload plummeting groundwards in a trajectory that would have made an Olympian javelin thrower proud.
Later, while the two teams rummaged through the wreckage to see if anything was salvageable, Nicol deadpanned: “Oh well, and that’s our final result!” Martinez was just as philosophical. “That’s engineering for you. You invariably learn more from your mistakes than from your successes.”
Neither team was put off by the experience one bit. In fact, both are keen to rebuild their projects and launch them again in their own time – just because they can.
Kemp, who is the chairman of UCT’s South African Space Association (SASA) student chapter, would also like to throw down the gauntlet to other universities. “What we want to do is host an annual CanSat event. Apart from being loads of fun, it’ll also be a great way for us all to compete against each other and create more awareness about space studies here at home.”
To find out more, visit SASA UCT’s Web site http://sasa.soc.uct.ac.za/