UK High Altitude Society

User Tools

Site Tools


PicoAtlas II


To actually see if it is possible to super-pressure mylar/foil balloons - this requires more balloons (of the same size) and a lighter payload. Launching would need a quicker ascent rate and an earlier launch time to make use of the solar heating.

Improvements for MkII

  • Lighter payload
    • Smaller battery - 1000mAh
    • Better power saving in the code
    • Smaller GPS antenna
    • Lighter wire UHF antenna
    • Use seeeduino rather than arduino pro
  • More balloons - order 4 91cm balloons and see how many we need to use
  • Preamble before tx'ing when in modes that have gaps
  • Battery voltage monitoring
  • Solar panel to charge battery during the day
  • Temperature sensor


  • Seeeduino film
  • GPSbee
  • Mini GPS antenna
  • Radiometrix NTX2
  • Wire 1/4 wave + radials
  • 1000mAh LiPo battery - ordered
  • Some insulation
  • Aiming for 75g total mass.
  • 36“ (91cm) Balloons ebay - £6.65


  • PicoAtlas will be on 434.073Mhz RTTY ASCII-8 50 baud, 425 shift, no parity and 2 stop bits - select pico in dl-fldigi.
  • Sample telem:
$$Callsign, count, time, latitude, longitude, altitude, lock, no of sats, nav status; tx mode; batt voltage; ext temp *CRC16
  • There will be three transmission modes for the flight.
    • Mode 0 - for the first 500 loops will be txing continously - this is to allow preparation before launch and for people to easily find the signal.
    • Mode 1 - 'Day Mode' - between 0700 and 1900 - the payload will transmit every 20 seconds turning the radio off between to save power - there will be a preamble of UUUU to help dl-fldigi catch the transmission. The GPS will remained power through out this.
    • Mode 2 - 'Night Mode' - between 1900 and 0700 - the payload will transmit 10 strings of data with gaps of 10 seconds and then will shutdown the GPS and radio for 10 minutes. If when the GPS is restarted it has failed to achieve lock it'll continue to tx until it does before shutting down again.
  • Battery Voltage
    • 1024 = 4.0v
    • 1000 = 3.9v
    • 975 = 3.8v
    • 950 = 3.7v
    • 925 = 3.6v
    • 900 = 3.5v
    • 875 = 3.4v
    • 850 = 3.3v


  • rebuild 1/4 wave with lighter radials
  • code to reset GPS settings if fails
  • finish insulation
  • Tests
    • Run battery test - 12hrs
    • Test if solar panel can charge battery (sat)
    • calculate best time for launch - launch before 10am
    • Test narcsleep
    • check radio EN


Overall the launch went well, filling the 4 balloons was easy, a pleasant change from worrying about damaging the balloon we were able to put the balloons in the boot of the car while we filled the rest. The payload also worked as planned and so once all was rigged up we released the balloons and payload. The ascent rate was slow but steady and we had good reception from multiple stations - at times up to 12 stations were decoding and uploading the data to the server.

Just after midday the balloon began to descend at approximately 1m/s and was tracked all the way down to 155m near to Rugby. As this was only an hour away by car I went to recover - arriving in the last known position I could pick up the radio strongly and after decoding the landing spot proceeded only to find it stuck in a high tree. On inspection it was possible to see that one of the balloons had burst and this had resulted in the slow drift down. The payload was too high to recover so I left my number at the local pub and went home, I also emailed some of the local hams. M0NWY kindly went to look the next morning and reported that it wasn't in the tree any longer and I then received a phone call saying that someone had found the balloon as it had come lose and drifted west. I was able to collect the payload in good condition the following Friday.


  • The current 36” balloons are not strong enough to super-pressure with the current setup.
  • The solar panel worked well at maintaining the LiPo charge during daylight hours. It was a shame that we weren't able to get any data the next day as it had blown out of the tree.
  • The Seeeduino film is just too fragile to use - a Arduino Mini Pro would be tougher and have the same mass.
  • Gaps between telemetry are fine as long as the pre-amble was present - this increased the success in decoding.
  • Ublox GPS modules when working are great!
projects/picoatlas/picoatlasii.txt · Last modified: 2011/02/09 18:49 by jcoxon

Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki