For my final year electronic engineering project I designed and launched a near-space balloon powered by the Parallax Propeller.  My 50 page report, a slideshow and firmware can be downloaded here,  but I will also give a short description below:

The project was called SUNSET (Stellenbosch University Near-Space Engineering Testbed). It had the following capabilities:

  • Powered by the Parallax Propeller clocked at 80MHz
  • Primary communication link was a 300mW VHF Radiometrix module which transmitted APRS packets created and modulated by the Propeller.
  • Backup coms link was a cheap ($22) GSM USB dongle which was hacked to be controllable over serial. It was supposed to send SMS messages upon landing.
  • The Prop used a a small composite video camera to take low-res, BW images. Images were transmitted to the ground line by line during the flight also using APRS packets. Custom software I wrote in VB reconstructed and displayed the images on the ground station during the flight.
  • Two thermistors measured inside and outside temperature.
  • An onboard micro-SD card logged images, temperatures and GPS coordinates throughout the flight.
  • Everything was powered by 6 AA Energizer Lithium batteries.

After a lot of testing the payload was deemed ready for launch. All up it weighed about 300g! The balloon was launched in cooperation with the Cape Town weather service from Cape Town International Airport. I remained at the airport with my ground station, while two other ground stations listened from nearby towns.  Two IGates forwarded the payload’s APRS packets to the internet. Everything worked perfectly for the first hour of the flight.  GPS coordinates and temperature measurements were returned once per minute.  Images were returned once every 10 minutes.  While the image transmission was proven to work, the quality of the images was too low to be useful, and they were mostly just white as it was a heavily overcast day.

Unfortunately, an hour into the flight, the payload stopped transmitting. It had traveled more than 50km and had reached 12km above ground level altitude.  The cause of the failure is still unknown, but I have a few theories.  I am still waiting for a farmer to find my payload in his field!  In the meantime I am working on SUNSET 2.  This time images will be transmitted using SSTV.  Using ROBOT8, images will be down-linked in 8 seconds, unlike the previous method, which required 2 minutes.

The report does not include information about the launch, but it is described, with some images, in the slideshow.