This is a HAB project financed by the University of oxford AOPP department, to sample carbon particles from the troposphere and stratosphere, hopefully allowing some vertical resolution to be achieved. The aim is to design an build a modular high altitude balloon system that can be modified for future missions, and avoids proprietary hardware at all possible stages.A more detailed description of the project
This was sourced from RFsolutions. The sensitivity isn't the best, but the current consumption is low and it apparently will work fine at altitude. It can get a lock after a few minutes on the top floor.
This was a very tricky issue, the plan is to use an electrostatic precipitator to collect particles directly onto SEM targets. The targets are grounded, and mounted at the bottom of a perspex tube. Electrodes on the opposite side of the tube are charged to high -ive voltage, causing corona discharge. There is a pump at one end of the tube, with conventional filters either side to prevent contamination. A differential pressure transducer is used to calculate flow rate. A piston on the end of the sampling tube is closed using a resistor cutter at balloon burst or cutdown, to protect the sample during descent. The sampling tube is removed from the gondola upon recovery, and opened in a cleanroom for analysis.
HV multipliers ( two 2KV units for upper troposphere and stratosphere, and a 5KV unit for lower troposhere)
This is based around an NGW100 linux board, powered by an AVR32 processor. There is a mobile on the rs232 port ( /dev/ttyS0 )for back up communications, and a gps and radio module on /dev/ttyS2
/dev/ttyS1 is used to communicate with an atmega168 on the daughterboard.
Flight computer close up
This was being developed for general UKHAS projects, but as several boards were available, and an “in house” solution is required, will be used for primary communications. This design has never been flown before, so testing is essential.
The radio sending the character “y”. This is the output voltage, so the RF trasnmit frequency should follow this shape.
scope screenshot (literally, as the floppy drive is toast on my scope). The sinusoidal pulse shaping can be seen.
another shot, here the timing is a little off