Thursday, 7 September 2017

The TT7F4 Flight

Date: 2017-09-08
Time: 6:00 GMT (8:00 CEST)
Place: Roznov pod Radhostem, CZ
Live Tracking: and

Callsign: OK7DMT-9
Frequency: Geofenced local APRS frequencies
Packet: Base-91 position and telemetry, Backlogged position and telemetry in Comment section
Example: !/5K{RS);uOJ3W 1F94!5MI,S(LWHC!+=7+;0c!'|!"!&:c+>0i!)!$|

Initial Prediction: 2017-09-07
Power Scheme
If the voltage on the solar panel is higher than 1.6V and the battery voltage is higher than 3.0V, the tracker transmits an APRS packet once per minute. If either the solar panel voltage or the battery voltage falls below its limit, the tracker transmits an APRS packet once per two minutes. In case the tracker registers altitude decreasing at a rate of 0.5m/s or higher during last two transmission cycles, it reduces the sleep period and transmits an APRS packet every 30 seconds.
This tracker uses Biwin GM10 GPS module instead of the usual U-blox MAX-M8Q. Its lower price is bought out by higher current consumption which is somewhere in range of 45mA during acquisition and about 6mA in backup mode. Another distinction of this tracker is a 20dBm Square Wave match for the Si4463 transmitter. This matching is intended for lower frequencies hence the APRS only transmission. The antenna used was once again a guitar string dipole with 492mm for each element. On this tracker, I wanted to test adding two 4.7F supercapacitors to a 400mAh LiPo battery. The two supercaps are wired in series to increase their voltage range to 5.4V while the battery is in parallel to the caps. The LTC3105's output is at 4.08V and the MPPC resistor is 180kΩ. The solar panel is composed of two parallel branches back to back at 90° angle to ground. Each branch consists of four 52x39mm solar cells in series. The tracker was encased with anti-static foam and black electric tape. The more complex power setup showed itself in the tracker's final mass which rolled by the 30g mark.
Mass: 31.4g
The main function running the tracker can be found in ARM_TT7F4.c on GitHub. The program flow is somewhat simplified by the absence of RTTY and SSDV routines. I used that opportunity to add a new transmission condition. The code keeps track of previous two altitude readings and the durations of the following sleep cycles, then calculates the average ascent/descent rate in the last two cycles. If it detects average descent over this period higher than 0.5m/s, the next APRS transmission will occur in only 30 seconds. A couple of modifications had to be done because of the Biwin GM10 module. It probably uses some older U-blox firmware so it required UBX-RXM-PMREQ message to enter backup mode. The navigation solution is still polled by the UBX-NAV-PVT message, however the returned data string is only 92 bytes long as opposed to 100 bytes as in case of MAX-M8Q firmware. The bitfield in an APRS packet's telemetry now signals whether the GPS module acquired fix (FIX), whether it is set in the Airborne mode (NAV), whether the balloon is descending and the extra short sleep period is active (DES), and whether the longer power saving cycle is active (PSM). The chart above is an example behaviour of the tracker as extracted from received APRS packets.
The balloon was made from 50 micron PE/PA/EVOH/PA/PE material and it is 1.84m in diameter. The estimated stretch is 1.74 Gamma (old inaccurate method). Witnessed wild swings in TT7F3's altitude at sunset, believed to be due to the gas cooling down, led me to go for a little more free lift this time. The chart above shows the stretching profile of this specific envelope. The increase in volume went quite quickly and I ended the stretching at indication of meridional weld opening.
Expected Float Altitude: 15100m
Expected Ascent Rate: 1.1m/s
Expected Superpressure: 1732Pa (at 19°C supertemperature)
Gas: Hydrogen
Free Lift: 17g
The final free lift was 18.2g, 2.7g more than in case of TT7F3 with similarly heavy tracker and same sized envelope. At this time in the morning the surface winds are usually calm here and provide for an easy launch even in a tree rich area. The average ascent rate was a little higher (1.14m/s vs. TT7F3's 0.95m/s). Everything worked as intended until the balloon reached 11864m, failed and started to descend at about 3m/s. Continuing at this rate, it eventually landed on a hillside at the Polish-Slovak border some 203km away from the launch site. The last packet came from 1035m with the nearing mountain range at 900m.

APRSfi_TT7F4.txt - collected raw packets.
Python parser of raw packet format.
The positional and telemetry data extracted from received APRS packets and decoded backlogs:

1 comment:

  1. Excited to follow your progress!

    Greetings from Brazil