In 2011, Jonathan Carroll, Bryan Chan, Andrew Harner, Tyler Reid, Paul Tarantino, and Tony Yu put together a stratospheric weather balloon mission launched from Gilroy, California. This mission tested the ability of a smartphone to compute GPS positions at high altitudes (>60,000 feet). US export control requires commercial GPS units to cease function if above 60,000 feet and traveling faster than 1000 knots. Since phone manufactures didn’t think people would be making phone calls above 60,000 feet, this was the only statement in place to cut off GPS. We worked with Dr. Frank van Diggelen, V.P. of GPS Technology at Broadcom as well as a Consulting Professor at Stanford University, who built a test unit smartphone to work at the high altitudes we achieved, in this case 102,300 feet (31.2 km).
This mission was equipped with a Canon A480 digital still shot camera and a Google Nexus S smartphone by Samsung. This equipment was housed in a styrofoam box for thermal insulation with added foam padding to absorb the impact upon landing. The camera was programed to take still photos at 5 second intervals using the Canon Hack Developer Kit (CHDK) firmware. This was done as the camera did not come with the ability to do this out of the box. The Google Nexus S smartphone was provided to us by Broadcom and came equipped with a special software version that allowed the GPS to legally operate above 60,000 feet. The phone was oriented in the payload enclosure such that the touchscreen was facing up and the battery was facing down, ensuring the best sky visibility of the GPS antenna. The camera was oriented with lens pointed toward the horizon. A circular port was cut in one side of the styrofoam shell to act as a viewport for the camera. A circular Borofloat® glass lens was cleaned and placed over the hole to add insulating properties while still providing an unobstructed view for the camera. Final results showed that this glass lense tended to fog up and ultimately was not needed during flight. However, we were lucky to have had it as we ended up landing in lake and this added crucial waterproofing. The payload box was painted a bright orange for visibility and was was hung from a parachute which in turn was hung from the weather balloon. The parachute is in place for the descent back to Earth. As the balloon ascends, the atmosphere gets thinner and the balloon expands. At some point the balloon can no longer expand due to limitations of the latex material and it simply bursts. The parachute is in place to slow the descent and ultimately give our equipment a gentle return to Earth.
This mission was an adventure from start to finish. We launched from a park in Gilroy, California (50 miles south of Stanford) that luckily had an outdoor racquetball court we could use for protection from the wind. We drew quite a crowd of mostly young kids who were playing in the park that day. They came over and were captivated by what we were doing and quietly watched and asked questions as we prepared. We were delighted to have such an audience for launch.
To our surprise, the balloon travelled much further than expected (90 miles) and landed in New Hogan Lake near Valley Springs, California. We travelled to the lake, bought an inflatable raft along the way, hiked in, and were surprised when we could not see it floating in the water. As none of us had cell reception, we did not know that the balloon had been picked up by a gentleman out fishing that day. We got on the phone with Laurel S. Coulter, a fellow student and lifesaver, who tracked the balloon on her computer and relayed where it was headed. This turned into a highway chase of a unknowing fisherman who had a head start. He finally stopped at his home and sure enough the GPS coordinates lead us right up his driveway to his boat in the garage. He was surprised (to say the least) when knocked on his door asking if he had found an orange box that day.
The GPS unit performed very well though had some trouble at the initial burst where the balloon was momentarily in freefall. The sudden 1g acceleration caused the GPS to momentarily lose lock and consequently was unable to compute the balloon’s position. Once the parachute slowed the descent, the GPS reacquired in a matter of seconds. These performance results were presented at the Institute of Navigation’s (ION) Global Navigation Satellite Systems (GNSS) conference in September of 2011 in Portland, Oregon and the paper can be found here. In fact, this project won the best presentation award for its session. This work led to Broadcom making it standard for their GPS chips to work above 60,000 ft (as long as you obey the GPS speed limit of 1000 knots), enabling smartphones the ability to track balloon flights out of the box.