Introduction: Sending Camera to Near Space by Helium Balloon, and Track Where It Lands.

About: Handaxe_Maker

Journey to space is a dream beyond reach for ordinary people, but we already seen many weird objects fly to “space”, like chair, LEGO toy, hamburger, Hello Kitty and so on, usually people sent camera or payload to near space by balloon. We can’t wait to release our own, in April 2014, thanks for varies of open source techniques, we send a helium balloon to near space and track where it lands.

Step 1: Prepare Balloon and Helium

Some people may think is easy, just release a helium balloon, take a camera to record video at high altitude. In fact it is not that simple. To get the flight video file, we have to retrieve HD video recorder, means we must track and recovery our payloads, the whole work is as complicated as launching a returnable satellite. Let’s check out what kinds of technique is needed for launching this balloon.

The balloon should be sounding balloon, manufactured from natural latex. Compared to ordinary plastic one, latex balloon is much bigger, and most important things is latex balloons are specially designed, be able to expand to a very large diameter. Take the balloon we used for example, it is about 1.5 meters in diameter at launch, can expand to a diameter of 9 m before burst. As the altitude rises, the pressure in the atmosphere decreases, and the balloon will expand. If balloon unable to expand to large diameter, the point where it bursts will be too low, cannot climbs to the edge of space.

The gas we filled in the balloon is industrial high purity helium, helium is an inert gas, won’t explode. Although it is not the lightest gas (density 0.1786 g/L at standard temperature and pressure), it’s much safer than hydrogen. Helium Specific Gravity (Air = 1) is 0.14, a cubic meter of helium provides a buoyancy force equal to the weight of 1.115 kilogram mass. One industrial size helium cylinder can provide about 6 cubic meter helium gas, balloon filling will use a little more than 4 cubic meter, rest gas can be used to play game like inhaling it and make your voice go high-pitched. The only disadvantage is its price is too high (over 300 dollars per cylinder).

Step 2: Prepare Camera

We use several camera in the payloads, main recorder is Gopro Hero 3+. Extreme sports fan all know its excellent capability. Austrian skydiver Felix Baumgartner wear a Gopro on his helmet when he jumped to Earth from a helium balloon in the stratosphere on 14 October 2012, skydiving an estimated 39 km. Gopro camera has wonderful waterproof housing, we also added Anti-Fog inserts to prevent lens fog while climb through troposphere.

Step 3: Trajectory Predicting, Thanks for Spirit of Sharing

We did plenty of works on predicting the landing location, read many papers, include Meteorology over China, atmospheric circulation, aerodynamics, sounding balloon thermodynamics analyses etc. Then we found several universities already have high quality latex sounding balloon flight path prediction model, say, Cambridge University Spaceflight Landing Predictor (

This predictor uses data from the NOAA GFS models to predict the flight patch, burst altitude and landing location. All we need to do is enter the payload mass, balloon mass, parachute descent rate and other data in the relevant box. Pressing enter will run the calculations and displaying the results. The most convenient part is the predictor can generate .kml file which can be opened in Google Earth like software, it is very helpful in finding the balloon. Thanks for spirit of sharing!

Cambridge University Spaceflight Landing Predictor, a tool to predict the flight path and landing location of latex sounding balloons.

Step 4: Devices to Track the Balloon's Position

Devices to track balloon’s position is the key point of payloads recovering. We tried many plans to track the sounding balloon. First plan is radio location, we choose one method called Automatic Packet Reporting System (APRS), APRS is an Automatic Packet Reporting System, retrieve current location via GPS or other methods, and encode information into audio signal and broadcast it through 144.64MHZ amateur radio frequency.

Receiving equipment then decode audio signal to restore the coordinate information, get payload’s location and trajectory. In early experiments, we also use OSD(On Screen Display), which are often used by model aircraft players. OSD devices will add many gesture and location info in video then sent it back to ground operator, so we can know the position via watching live video.

Also we learn from other balloon project’s experience, choose “GPS tracker”, a reliable merchandised product. Most GPS trackers rely on a cellular signal, send GPS coordination back to specified cellphone via text message, so we have to pray mobile operators network service is robust enough in that area.

Step 5: Prepare Parachute

Calculate diameter and area of parachute

D = sqrt( (8 m g) / (p r Cd v2) )

D is the chute diameter in meters

m is the mass in kilograms

g is the acceleration of gravity

p is π

r is the density of air = 1.22 kg/m3

Cd is the drag coefficient of the chute

v is the speed we want at impact

Use UAV to test the speed of landing

Whole device is very simple (simple means more reliable) Balloon at the top, a line attach to the balloon, with parachute suspended at the other end. Our payload attach to chute cord. We must calculate the size of chute, if the wished land speed is 4m/s, area will be 0.75 square meters, and diameter less than 1 m. We open a hole on top of the chute, so the land velocity will be much higher. Some data suggests paratrooper’s max land velocity is 7m/s, UAV chute landing is at the same rate. So all velocity less than 7m/s will be safe. Fast descent speed means less distances to run!

Step 6: Must-do Tests

Use helium balloon launch a camera to nearspace and recovery it, is almost like launch a returnable satellite, need test of many systems, including balloon, camera system, payload, tracking system and parachute.

In fact we failed in first trial, although we do lots of subsystem test, our team member scattered far and wide over Beijing, and seldom have opportunity to test the whole system together. In first trial many tracking device happened to defunct at same time, and our SUV can’t catch balloon’s speed in desert, the signal was lost after 13km tracking. So in second trial we simplified the track device, made double backup for more reliable devices like GPS tracker, also we optimized launch process and tracking routine. After balloon launch we immediately sent a SUV drive to predict landing location ASAP.

Before the launch we tested the device under low temperature, we put all devices in payload (usually a foam box, with an hole to let camera take video from inside), seal inner side by aluminum foil tape, then put into refrigerator’s freezer (minus 30 Celsius) to test device’s performance in low temperature. Top layer of troposphere can reach -60℃, but temperature will rise in stratosphere.

Since theory result may be differ from actual speed, we should also test parachute descent rate. Descent rate is a crucial parameter for predicting trajectory. We use a UAV to bring the chute and payload to high altitude, then release and test the descent speed.

Step 7: Epilogue

We send payload to near space and recover it on Apr 26th 2014, and luck is a factor in this success. Our APRS didn’t work after launching and GPS tracker can’t send signal at high altitude. Cellphone signal in grassland and desert is bad, our GPS tracker get through -40℃ and return 4 location data before landing. Since there are no altitude data, we were not sure if it is landed. We reached the final coordinate location after 130 km driving and found nothing but vast grassland and desert, had almost lost hope. At this time GPS tracker just sent another 2 signal and the data suggested it was only 400 meter away from us. We climbed a dune and saw one herder wave hand to us, the payload land at a sand dune and there were no signal in this area, but after herder pick up the payload GPS tracker just got signal again! We are so lucky!

Step 8: Future Plan

Use more reliable tracking devices,multi-system backup. Change payload cabin’s shape to glider shape, so we can add flight control system and autopilot to let it land on specified location.

Step 9: Safety Related Issue, Avert Airline Route

If we want to fly a stratosphere balloon and let it land safe on a location easy to find, it is important to choose a safe launch site. We can use Cambridge University Spaceflight Landing Predictor, but in China is difficult to find a good site. China locate at mid - latitude westerly winds belt, there are many populated cities in east coastal area, balloon landing at those area will be dangerous even with parachute, not to mention some people may take the payload away before we arrive. Best site should be least populated places, like desert or grassland in northwest China. Also we must avert airport and airline route.

At last, we should remind that the activity of flying unmanned free balloons and captive balloons inside the territory of the People's Republic of China shall be subject to the relevant provisions. We must file an application to the flight control department in accordance with the relevant provisions of the state, and shall not launch balloon before approval.