For the most part, the station has been tracking balloons this month as part of it’s duty in supporting university academic activities.
The balloons launched were student built projects for masters projects, and they all worked to varying degrees of success.
Initially we had Project Strathosphere, however this was launched on a day that I wasn’t able to track it and as such there’s no data downlinked, however it was launched and recovered successfully, admittedly with a little bit of luck. This used the base design for electronics was one that was developed with STAC providing the essential basics. This project had a wire dipole for an antenna.
Following this, we had Skypod, which was a very successful launch with all the relevant data downlinked comfortably. This balloon differed from the other two as the group developed their own circuit and code successfully, based solely on a basic specification given to them and used a wire j-pole antenna. Sensors included an accelerometer, pressure, multiple temperature readings and more. Launched from Newton Stewart and picked up in Northumbria, this went very well from both the group’s end and my end.
Finally, we had Fusen, another successful launch that had some technical hiccups once in the air, which proved to be the most interesting from a ground station point of view. Using a different GPS module bought at short notice, it hit a ceiling at 12.5Km, and due to the nature of the code in the flight computer no longer transmitted any data back to earth, however it did emit a constant tone. Using this tone and the readout of signal/noise on the high altitude ballooning specific software, dl-fldigi, a custom modification by members of UKHAS, the UK High Altitude Society, we were able to pinpoint the location angle of the balloon from the station and by collating the data with that from the HAB predictor we were able to work out that it actually landed more or less exactly where it was supposed to. See the pictures below for a high signal level example and a low(er) signal level example..
First off, we have the high signal level. Spot at the bottom right where it tells us there’s a signal/noise ratio of 23dB, or that the signal is 23dB above the noise level.
Here, we have a lower signal level - spot how the S/N value is now 5dB (still a strong signal) however the waterfall background is yellow, indicating a much higher level of noise which was also audible.
The peak S/N level before it landed was 28dB at 216 degrees AZ, which from the station is the Stewarton, Ayrshire area, where it was expected to land.
Future work to be taken from Fusen is that data should still be transmitted regularly so the other experiments (in this case, the thermal characteristics of the foam) can still be continued, and hopefully the GPS module may come back to life.
In the case of all the balloons, the Radiometrix NTX-2 transmitter was used on the unlicensed 434.650MHz frequency, with an erp of 10mW. Of all the antennae tried (dipole, j-pole, monopole) it was felt that the j-pole performed best, however there were some drift issues when it went into freefall/descent. If a more solid material had been used (maybe piano wire, or some sort of reinforcement of the copper wire) then falling performance would have been ideal. The main issue, however, is that tuning any of the antennae is difficult without access to a UHF capable antenna analyser and adapting them for connection to it.
For flight computers, Arduino microcontrollers were used because of their ease of availability, cost and simplicity to learn and use. The ideal configuration would have been using the chip (Atmel Atmega 328) on it’s own and only using the arduino board for programming, but due to the lack of experience in electronics the Fusen and Strathosphere groups had, it was felt appropriate to use the board as-is.
For GPS, the Falcom FSA-03 module was preferred ~ this ublox compatable module provided us with plenty useful data when polled, and worked well with a modified library for arduino.
All the flight computer software for Strathosphere and Fusen can be found at GitHub, with circuit diagrams to follow.
Now, any questions?