Harnessing RP2040 for microgravity research in zero-g parabolic flight
Last year, engineer Christian Wenzel-Benner and researcher Dr. Michaela Dümmer from Prof. Dr. Christoph Forreiter‘s research group exploring plant behaviour in microgravity got in touch. The team’s experiments demand a high degree of automation, actuation, and sensing, and they had found RP2040 to be a low-cost solution ideally suited to the task. They wondered if we’d be interested in learning more; naturally, we were! Below, Christian and Michaela share the challenges of their work and how Raspberry Pi hardware helped to tackle them.
On a sunny Sunday morning in 2024 a group of scientists headed from Marburg, Germany, to Bordeaux, France, to carry out experiments in weightlessness on a parabolic flight campaign supported by DLR (German Aerospace Center)/BMWi funded project 50BW2134. The van was packed to the rafters with luggage and equipment, and we had thirteen hours of driving ahead, almost without any break. The roads seemed endless, but finally we reached the military area of the Bordeaux–Mérignac Airport, home base of the Airbus A310 Zero-G waiting for its take-off.
Roughly a year previously it had been very much in doubt that this day would actually come. The heavy and expensive equipment we’d inherited from other scientists was becoming unreliable, and development licences had lapsed. Licence renewal was out of budget, and some vital replacement equipment seemed to be made from pure unobtainium. This would be the flight that had to go right, but massive g-forces, stringent safety requirements, and the constant danger of a scientist-operator becoming incapacitated were among the long list of challenges facing this very unique experiment. Reliability was mandatory; money and development time were limited.
Brooding over the options, the group’s engineer had a crazy idea: what if we could replace all the problematic, unobtainable, heavy and costly components with something like a Raspberry Pi Pico and a custom printed circuit board? A fully integrated measurement and detection device – MD-Device for short. Crazy! But if it worked, it would get us to France, where, after a hard week of preparing and testing, the real fun would begin.
The real fun: that is to say, week two – the flying week. The first parabola is always the most exciting: strapped to the ground next to your experiment you eagerly await the “pull-up” announcement, when the engines boost the aircraft straight into the sky, almost stalling, at an angle of about 50° for about 20 seconds. Subjected to 1.8 g, twice the normal g-force, you are pressed against the floor. You can barely move until the loudspeaker says: “Injection!”. After a short transition you encounter 22 seconds of weightlessness. All weight drops away, and you start to float from the floor. What a feeling! Of course, you are not allowed to float around uncontrolled or freely and, of course, you are busy with supervising your experiment, while the plane goes over the top, nosedives, and heads for the ocean below. After 22 seconds of weightlessness the plane pulled out of the dive, exposing us to 1.8 g again. We experienced this 31 times on each flight, on three consecutive flight days. Exhausting, but an incredible, exciting experience!
Our experimental “pets” were young, green plant seedlings, and our goal was to analyse the physiological response of plants to altered gravity. We were therefore investigating the distribution of calcium ions in plants that had been genetically modified to make them capable of emitting light (luminescence) after calcium binding. Light emission was detected by a photon multiplier tube and the output pulses were counted. To achieve this, we had conducted earlier ground-based experiments using an inherited industry-standard test, measurement and control system. Digital I/O, motor control, and automation of the experiments’ control flow were all handled within this digital ecosystem. But the licences had lapsed and replacement parts were hard to get. The switch to an RP2040-based MD-Device, originally a course of action in which the group had little choice, turned out to be the gift that kept on giving.
Experiments under weightlessness require a much more sophisticated approach than ground experiments: every experimental process needs to be automated and synchronised with the detection unit, since experimenters have only limited access to the samples during zero-g. The MD-Devices handled all that automation. To avoid a single point of failure we built four complete setups (called Lumiboxes), each controlled by an MD-Device. Two Lumiboxes went into a heavy-duty metal box (called a “Rack”), each with an independent power supply, control laptop, and scientist-operator. Since each RP2040 sends the data to the control laptop as well as a local SD card, this provides 2×2×2 redundancy. It worked like a charm. Every Lumibox, every MD-Device performed flawlessly over the three days. We obtained both redundant and diverse experimental data at a volume and quality that our inherited equipment would never have produced.
RP2040 was instrumental to this success in more than one way. The MD-Device monitors g-forces, controls a stepper motor, and logs light pulses – redundantly. Those core functions are tied together by an experiment control flow sequence written on a PC (for science) but executed independently on the MD-Device (for redundancy) in case the USB cable to the PC should become unplugged at 1.8x normal gravity. We needed USB: RP2040 has it. We needed to count pulses up to 55MHz: RP2040’s PWM is good for 62.5Mhz at stock speed. We needed a gravity sensor: RP2040 boards by Adafruit and a matching helper board provided that, as well as a way to back up data to SD cards. All of that is quite a bit of software, hard to develop and to run fast enough. The Raspberry Pi Pico C/C++ SDK sped up development, and the dual-core RP2040 provided the execution speed. But before we could go on board, everything needed to be tested under realistic conditions.
Our task was to generate a repeatable photon emission test that triggers reliably on a specific gravity vector change and fits into a petri dish (diameter 10cm) suspended vertically in a sealed dark chamber. That feat takes either divine intervention or an RP2040 microcontroller board hooked up to an accelerometer IC and a small battery. The result was “LumiBerry”, a CircuitPython-based synthetic plant. It served us quite well during the developmental phase, allowing us to test automation approaches without real plants and ridiculously expensive luminescent chemicals.
But where on earth could we try out our experiment setup in weightlessness? The answer was at the Center of Applied Space Technology and Microgravity (ZARM) at the University of Bremen. They can provide 2–9 seconds of weightlessness in the Bremen Drop Tower and the GraviTower Bremen Pro. This allowed us to successfully test the MD-Devices under real (but brief) conditions of weightlessness at the end of 2023, to prepare for boarding the aircraft the following year.
The people at ZARM were supremely helpful, and we consider their slogan “test before flight” to be spot-on advice for zero-g experiments. Anyone interested in the ZARM facility should check out Tom Scott’s excellent video about the Bremen Drop Tower.
RP2040 transformed a large, heavy, expensive setup into efficient equipment for our experiments in weightlessness, enabling us to take our work on a parabolic zero-g flight and then return to Bremen to carry out further experiments in November 2024.
The post Harnessing RP2040 for microgravity research in zero-g parabolic flight appeared first on Raspberry Pi.
from News - Raspberry Pi https://ift.tt/LJMHTC1
Comments
Post a Comment