High-Altitude Balloon Payload

SSD Balloon taking flight
SSD Balloon payload was a cooler
Solid State Depot members Brandon Skari and Tim Pegg wanted to fly a high altitude balloon and chose the 2017 eclipse as a good goal because totality would be close-by in Wyoming. They roped a bunch of other members into the project and I contributed one of the payloads. I implemented a version of Dave Ackerman's Pi In the Sky camera with a LoRa radio, two ground stations feeding the internet-based tracker at habhub and a hand held receiver to aid in finding the package after it parachuted to ground. The mission was a success and we retrieved our payload (although we had some parachute issues) and I got good data from my transmitter throughout the flight including a final transmission only a few hundred feet off the ground that helped us find it.

The flight is more fully documented on hackaday.io.

The code and design for a Teensy based PiTS tracker is on my github.


The project video including images from the payload taken during the flight.

Unprocessed video from Brandon's 360° camera during totality.

Payload hardware

The Pi In The Sky system records images on the Pi's SD Card as well as transmits a telemetry stream over the LoRa radio. Data includes temperature and GPS location as well as very highly compressed jpeg images. The entire reason behind this payload was to have some image data even if the payload was lost or completely destroyed.
LoRa payload guts

My payload consisted of a Pi Zero, Camera, high-altitude GPS, DS18B20 temperature sensor and RFM95W breakout with quarter-wave antenna. We got packets at distance of over 70 miles. The system was powered by 3 lithium AA cells to handle the cold.

Pi in the sky Raspberry-pi based receivers

Two Pi receivers were built, one for the launch crew and one for my use in Casper, near where we had estimated touchdown to occur. They, and the payload, ran the PiTs code. I added a LoRa configuration for North America that got up to 1200 bps.

Andrew looking at telemetry near the launch site

Launch location ground station near Shoshoni, WY. Data was displayed on a monitor attached directly to the receiving Pi. Both ground stations relayed data to habhub when possible and other people could track our flight and view the low-resolution images in real-time.

My base station was on the roof of a house in Casper WY

My ground station in Casper on top of a roof. The Pi connected wirelessly to computers on the dining room table below but I kept having to climb on the roof to aim the antenna.

Tracking LoRa receiver antenna mounted on a vice

I designed a third receiver using a Teensy LC, RFM95W module and old Sparkfun bit-map LCD display. It parsed the PiTs packets and was designed for the recovery crew to track the payload. Unfortunately we didn't connect the day of the flight and I only got to test it on the roof.

Tracking LoRa receiver readout

Displaying information about packets received from the payload.

Balloon temperatures all the say down to -55C

Temperature as recorded by the payload during the flight.

Balloon flight path in 3D thanks to google earth

3D GPS data from the flight plotted on google earth.

Test image as stored on the SD Card

Unfortunately I lost all the images from the payload after a disk crash. This is an image stored on the payload Pi's SD card during range testing. I hung the payload in a tree on a hill near my house that is on the front range of the Rocky Mountains. Then I drove into eastern Colorado stopping every once in a while to see if I could still receive data. I was able to receive data almost 50 miles away.

Test image as received in low resolution

This is the same image as received somewhere east of the payload. It shows the effect of lost packets. Each low resolution image requires multiple packets.

Final image durint ascent

Just before the balloon burst at 94,000 feet (28,600 m).