A Grand Day Out: The Story

Part 1: Launch

It all started almost a year ago with two brothers and a dream: a dream to send a weather balloon with a
camera into the stratosphere, to take footage of the earth at 100,000 feet. We had thought about the possibility of such an endeavor for a long time, but not much came of it until one day, we decided that we would just go for it. In the months that followed, most of our free time was devoted to researching the different options we had, discussing costs, and looking at other people's successful projects. After hours upon hours of internet searches and reading, our project appeared to be set in stone. The only thing left to do was purchase items and put it together. We learned a lot as we went, sometimes developing our own methods, other times drawing from outside resources. We cut costs here and there, made a few changes to our plans, and slowly gathered all of the pieces for the mission, finding time to work together in the midst of Erik's busy work schedule. It wasn't until a week before the launch that we began to feel the full weight of our undertaking. The capsule was not yet complete and we had not lined up a helium supplier. These problems were solved during the week. However, we discovered another, more pressing, issue. We faced a possible reschedule due to the uncertainty in whether or not Erik would get the launch day off. Sunday, August 18th was the only day which worked for all of us, and rescheduling the launch would have been a nightmare. During the week, he talked with his coworkers to see if anyone would trade workdays, but to no avail. It all came down to the day before launch on Saturday, August 17th. It was 4:00 PM, I had yet to receive a message from Erik saying that he'd gotten the day off, and the helium place closed at 6:00 PM. He finally called me around five to say that he had worked it out and we were clear for launch.

I quickly called the helium place to ensure that they still had helium in stock. They kindly notified me that they were out of the 100 cubic foot tanks but still had 80 cubic foot tanks in stock. That was okay, I thought, because we had calculated that we only needed 70 cubic feet. Upon arrival at the store, I was pleasantly surprised to find out that there indeed were 100 cubic foot tanks in stock. With a tank of helium in the back of the truck, we headed to a home improvement store to pick up a swivel for the parachute and some tubing for the helium tank. Back at home, my brother and I began assembling and packing all of the gear. Laboring into the night, we finally went to bed at 1:00 AM, which would give us one hour of sleep before getting up at 2:00 and leaving by 3:00. The roads were clear as we drove to the launch site in Bickleton, Washington. It was a mere 162 miles and 3 hours away, situated near the Columbia river and in the middle of nowhere. We arrived a full hour before sunrise, at 6:00. Although the sun was still below the horizon, the sky was a bright orange-yellow, giving us ample lighting for setting up the balloon. We drove around the Bickleton area for a while, after discovering that our launch site, a supposed public park, was incorrectly marked on Google Earth, and was located on private land. There was a second issue as well: the supposed 3G cell connection displayed on the Sprint map was nowhere to be found. A recovery required cell connection, and we could only hope that there was better service in the landing zone to the East. We found a secluded road bordering farmland, pulled to the side, and began setting up our balloon. The wind at ground level was almost nonexistent, perfect for filling a very large, very lightweight balloon. Due to our delay in arriving and finding a launch location, we did not begin assembly until 7:00 AM, the time I'd hoped to be launching. Thankfully, the wind calculations gave us a large window of time and a launch by 8:00 AM would be okay. After connecting all of the components together and setting up the electronics in the capsule, we began to cautiously fill the balloon. 

Work was very slow, as both of us were nervous about popping the balloon, but we gradually sped up the filling process. Sometime during the filling process, two men from a nearby farm drove up in a truck to see what we were doing. We told them about our launch and they stayed to watch the whole ordeal. The balloon continued to get bigger and bigger, but it still wasn't lifting the counterweight. The counterweight was a simple milk jug filled with a given weight in water, a weight we had calculated to give us the right positive lift when equilibrium was reached. As we continued to fill the balloon with helium, our worries grew. The balloon should have only needed 70 cubic feet of helium to lift the counterweight, and it felt like we had already put that much in. Suddenly, the pressure from the helium tank fell. There was a full 100 cubic feet of helium in the balloon and the counterweight had not been lifted off the ground. We tried to think of an explanation, and assumed it was because the counterweight must have been inaccurate. I realize now that this was a correct assumption. The counterweight had leaked during transportation, and when we refilled it, we inadvertently added more water than necessary. With the full 100 cubic feet of helium in the balloon, we tied it off and began to cut the safety lines. We slowly let the balloon lines out, making a final check on the electronics before saying our goodbyes, counting to three, and releasing the balloon together. The time was 7:52 AM. It was a strange feeling, seeing all of our work, our hours and hours of research and design, rising into the sky and eventually out of sight. We breathed a sigh of relief, but the mission was only half-over. The more difficult, more uncertain half had just started, and it depended solely on cell reception and an Android app.

Part 2: Recovery

Due to the fact that we had no cell coverage in the launch zone, we had to go find a Wi-Fi hotspot and run a final flight prediction. The original intention was to run a prediction from the launch zone, tethering our phones to a laptop to give it internet connection. With this small change of plans, we packed our gear into the truck and drove north to Prosser, where there would surely be a McDonald's with Wi-Fi and some breakfast. As we drove, we prayed for a quick and easy recovery. I could not have imagined the ordeal we were about to experience. At the McDonald's in Prosser, we got something to eat and loaded up our computers. It was when we tried to run a flight prediction that I remembered one fatal point: flight predictions only work in the future. The time was already 9:00 AM, over an hour since launch, and the flight prediction would be totally inaccurate. We decided to base our landing zone off of a prediction I ran the night before, giving us a general area to search. The prediction I ran also gave us a flight time of 2 hours 17 minutes, so we had to get going in order to view the landing. We left the McDonald's and drove southeast down Highway 221. At this point we sent our first "ping" to the cell phone on board the capsule, making a GPS location request. As we went, I periodically checked the other cell phones to see if they had coverage. 

Things were not looking good as we drew closer to the landing zone, coverage was jumping between no service and 2 or 3 bars of 3G service. We arrived on location around 9:50 AM, driving up and down 221 to find a location with decent coverage. After finding a spot, we sat on top of the truck and scanned the skies, sending "pings" to the capsule cell phone every so often, and waited for a response. The predicted landing time came and went, and no response was received. We waited another ten minutes or so, but neither sight nor signal came. We assumed that the capsule had landed in a location without service, and we would need to find it manually. Back on 221, we selected side roads almost at random and drove down them for a while, before turning back and selecting another. After about twenty minutes, the impossibility of our task began to dawn on us. We were surrounded by hundreds of giant, circular farms, and the capsule could be in any one of them. This was no "needle in a haystack". It was "small red parachute in thousands of acres of farmland", "needle in a haystack on steroids". We drove for what seemed like hours, really just one hour, endlessly looking out the window. "Was that it?", "No, just a plastic bag". Spirits plunged to an all-time low. We found ourselves driving West on an undeveloped gravel road. The road turned South, back towards the Columbia, towards home. One last stretch of road before giving the balloon up for lost. I began to try and come to terms with the fact that we would never see the capsule again. The car was silent as we drove South down Alderdale Road towards the Columbia River. I could not have expected what came next. 

At exactly 11:58 AM, the silence was harshly broken by a sound from one of the cell phones. It was a text message. I quickly reached for the phone and read the message. Sure enough, it was a message from the capsule, containing GPS coordinates. Another phone received a message from the capsule while I looked at the first. The car was bustling with excitement as message after message came in, responses to every individual "ping" we sent the capsule. We pulled off on a side road and loaded the GPS coordinates into our devices. When we saw the location of the capsule, we became quite confused. In straight lines, the capsule was 73 miles East of the launch location, and 45 miles East from the predicted landing location. By road, the capsule was 72.5 miles away from where we stopped the truck. In an instant, we went from a hopeless search to knowing the exact location of the capsule, accurate to 5 meters. We drove East to the coordinates given, and tried to make sense of what happened. The balloon's flight had lasted 4 hours. But if we had put more helium than we planned, the balloons flight should have been shorter, not longer. As it turns out, even though we put in more helium, we had not planned for enough positive lift to give it a fast enough ascension rate. The various resources we consulted had different recommendations for positive lift. One recommended 1.5 times the weight of the capsule for lift in helium. Another recommended 3 or 4 pounds of positive lift. We gave the balloon 600 grams, as recommended by a third resource. This was clearly not enough, giving us only half the ascension rate we desired. We drove on, crossing the Columbia at Umatilla, and followed the river East and then North, before turning inland around Burbank. The terrain in this area was not looking good, the capsule could have landed in a stockyard, a small pond, or the tightly packed, inaccessible corn fields. We drove up a small hill, following the GPS coordinates, and found two farmhouses. We knocked on both doors to ask permission to search the fields, but no one was home. We drove down a road bordering the huge circular fields, and there it was. Lying in the dirt and weeds between two circle fields, the capsule was clearly visible and only a hundred or so feet from the road. The recovery may not have been quick, but it was definitely easy. We quickly brought the capsule and parachute back and began the long drive home. In total, we had driven 648 miles.