Sharing Knowledge

Space: the final frontier, we came close to it with a balloon and have pictures to prove it. The idea is to launch a weather balloon with an attached camera to try and get pictures from near space. I have Reddit to thank for this as that is where I read the original article that started off this adventure. As a tribute, the balloon was named ReHAB, this was suggested by Reddit user Maxd and stands for Reddit High Altitude Balloon.

Sure, this has been done before but in this case I wanted to make it as simple as possible and to track the balloon in real time without needing a special amateur radio license. Cost was an issue but not the primary one; I did not want to make this as cheap as possible, rather my concern was to reduce the complexity and make this easy for others to do. With that in mind I’m going to explain all the steps right down to the basic details in my next post in the hopes it helps more people launch their own near space adventure, this post will just cover the story of the launch.

“Hey, the weather looks good for Sunday” was the message that greeted me when I logged on to IRC. This was Gussy, part of the balloon team and a fullsize hotair balloon pilot so he’s good at judging the winds. I open the balloon trajectory website, put in the coordinates for Melbourne, Australia along with our balloon data and sure enough it’s perfect. As the wind is a southerly it means our balloon can be launched near the city without it going out to sea, which is great as we don’t have a boat.

Launching near the city is exactly what we wanted, firstly it means we can get interesting pictures early on in the flight and secondly as its an early morning launch, it means I can stay in bed longer. The trajectory calculator estimates a landing in the area of Shepparton, which is about 100km north of Melbourne. Its now late Thursday afternoon, this gives me Friday and Saturday to get some Helium.

Getting Helium was something that Kev, another member of the team was going to look in to. This man has contacts and can get just about anything, I call him but it goes to voice mail. Not good, it is now Friday afternoon and I’m not sure if the Helium shops are open on Saturday. I look in the yellow pages and call a few of the gas supply places, no luck they all want to lease a cylinder for a year. I call Gussy to discuss and he suggests a party supply shop close by. I call them and they have a full cylinder, exactly the right size for $98. I call by and pick it up. We are all set for Sunday morning.

On Saturday I doubled checked everything in the capsule was working, put new batteries in the camera, charge the batteries for the tracking setup  and prepared a check list for Sunday morning. I also put a small message on the outside of the capsule explaining it was a harmless science experiment and with mine and Gussy’s’ phone number on it in case someone found it before we did.

Launch Day

Everyone turns up at my place and we do a final check of the trajectory tracker, the winds have changed slightly overnight and it indicates we should land north of Yea and very close to Euroa, I start google earth and make sure the area around the landing spot and the predicted flight path is cached at multiple levels of zoom. So we load the balloon and Christine my girlfriend (the adventure’s photographer) in to Gussy car; he has a 4 wheel drive and there is every chance that we will need it. We also taped the tracking modem for the laptop to the roof of his car and the USB cable is run via the sunroof.

The launch spot is just 5 minutes away from my place, we unload all the bits, I consult the check list and Christine starts taking photos. Right, turn on the trackers and test them. There are two trackers in the capsule, the first is a GSM / GPS tracker, you simply call it and it returns its location via SMS. I power it on, wait a couple minutes for it to get a GPS fix and call it, I put the coordinates it returns in to Google Earth and they are exactly where we are, excellent. Next is the live tracking, I plug in the Xbee modem on the roof of the car into the laptop via USB and start Earth Bridge, select the right com port and hit connect. Next the xbee and GPS unit in the capsule is powered on and we start getting NMEA data displayed in Earth Bridge, 30 seconds later we have a fix and google earth is showing the correct location.

Flight string awaits the balloon

Next it is time to fill the balloon, this is the part I am most concerned about. The lady at the Party shop said we didn’t need a regulator as that would take us forever to fill the balloon, so we put on our latex gloves (except for Kev as they were too small for his shovel hand), opened the balloon packet and just stretched the balloon nozzle over the neck of the helium cylinder. I opened the valve slowly.

With some odd noises, the helium rushes in and our balloon starts to inflate! The biggest question is where do we stop? The 350g balloon we are using is meant to be filled to 1.03 cu meters of volume and our cylinder contains 1.2 cu meters. However we are slightly heavy and if we under fill the balloon it won’t go high enough to burst and could stay in the Stratosphere for days. Conversely, if we over fill it, it won’t reach its maximum height and land well short of the predicted landing zone, possibly in a forest. The manufacturers state a diameter of 125cms at release, it looks about that size so we get out the tape measure and are trying to measure it when the gas starts to run out, good enough we think.

One full balloon.

After a bit of a struggle we get the balloon off the cylinder and Kev ties the balloon off and attaches the flight string. It wasn’t easy to tie but Kev has a lot of experience with balloon knots so he got the job done. The camera is started and the capsule lid is placed on and is sealed with tape, slowly we let the balloon take the strain on the line and I’m left holding the capsule with balloon bobbing above it in the air.

Flight string getting attached to the balloon

Is this it? Finally it’s going to get launched? A lot of preparation and testing with only the final step to go: release the balloon.

Bear in mind that once I let go of this thing I have no idea what is going to happen. I hope all the calculations were right and it will rise swiftly in to the air but I am now questioning the idea of launching it next to the Freeway. A couple of thoughts are running through my mind: What if it hops the fence and the only picture we get is the face of a terrified motorist as the balloon bounces of their car windscreen? And, why the hell didn’t I put just Gussy’s phone number on it? Oh well no turning back now I thought as I let go.

The balloon rose quickly in to the air, much faster than I thought it would. We watch it for a few minutes until it becomes just a dot in the sky and then we head off to the chase vehicle to check the laptop and see if we are getting good tracking information. Tracking is working fine and the balloon is heading away on the predicted track at 18MPH, oh, it appears I forgot to switch Earth Bridge to metric, never mind as the hunt is now on and we head off in pursuit!

Up and Away

For the first hour tracking the balloon was fairly leisurely, we were close to the city and had a large choice of roads we could use to stay under it, the balloon wasn’t going that fast either around 25 MPH, we easily stayed ahead of it and stopped a couple of times to let it catch up. The main debate was how long we were going to get tracking data for, the 900Mhz xbee XSC modems are rated to 20 km but this is generally viewed as an over estimation and we didn’t expect to get this, even though we were using the higher gain omni antennas and mounted them in a horizontal polarisation to use them more effectively.

On the way

Then at around 30,000 ft the balloon really started to move, 40mph, 50mph and finally settled to around 65 mph (105kmh). We were still getting a perfect track from the on board GPS via the Xbees but the balloon was pulling away quickly as we were on country roads and ones that took us away from the actual flight path of the balloon. We decided to just take the fastest path towards the balloons predicted landing point as at this point we were convinced we’d lose the xbee signal. Just as we approached Yea and still with a perfect track from the xbees, the balloon started to slow down, it was now at 68,000ft and moving at only 10 mph but still climbing.

Leaving the city behind

We caught up and stopped the car directly where the GPS was showing us the balloon’s position was, although we couldn’t see it, we knew that somewhere high above us was a balloon and beer esky, it had already traversed the Troposhere, and just crossed the Tropopause where the outside temperature was around -55 degrees Celsius and the balloon had increased it’s size to almost 4 meters in diameter due to the pressure differential.

Over Melbourne's Suburbs

Kaymont state that this particular balloon has a bursting altitude of around 80,000ft, if we over filled it too much, this would be lower and we kept an eye on the GPS for any decrease in altitude. Woo hoo, 70,000ft, then 75,000ft came up as we watched the laptop like hawks, 78,000ft, 80,000ft. Then the GPS stopped giving us readings except for time updates, turns out this is a bug in the firmware. A couple of minutes later, we get a reading: 78,500ft, we are on the way down! What seemed like a few seconds later, we get another reading, 72,000ft, it’s coming down FAST. 65,000ft, 52,000ft and we are trying to get to where we could watch the landing. Now we have two problems, firstly has the parachute failed to deploy and secondly there are no roads to take us to where we estimate it will land. We have to go 5km at right angles to the balloon before we can turn towards the predicted landing spot. 24,000ft, 21,500ft….

Victoria from 50,000ft

Finally at 19,000ft the altitude readings slow down considerably although we are certainly going to be too far away to see the landing and there is a large hill between us and the balloon, hence, we are certain we will lose have line of sight and will have the live tracking no longer . Our last reading is 11,000ft but the rate of decent has slowed dramatically and so the balloon appears to be under canopy! After a few minutes Gussy starts calling the GSM tracker in the balloon, tensions are high as we really need this to work to save searching for the balloon. 5 minutes pass, the GSM tracker number just goes to the default message that states the cell phone is either out of range or switched off. Not good and the mood starts to get gloomy, Oh well, Gussy states, I’ll keep trying. We start to look closely at the laptop to guess where the balloon might have come down. ITS RINGING!!! yells Gussy. Sure enough, 10 seconds later his phone announces a new SMS.

Highest point in the flight, over 80,000ft

The next call and SMS reported a position 100 meters further than the first, then the next few reported the same position so we were sure the balloon had landed and we had its location. 10 minutes of driving around country lanes and we arrive at the farm house of the farm the balloon has landed on, even though they think we are a bit mad they agree to escort us up to its expected location. Five minutes of following a bright yellow carbon monoxide emitting motorcycle engined “jeep”  up a dirt track, we stop at the top of hill and start looking for the esky and the parachute. Finally Kev spots the yellow of the parachute on the side of the hill a few hundred meters away!  It has landed safely on grass with no damage to the foam esky at all! No trees, rocks, cow pats or dams involved.  Pretty lucky really as the landing area is filled with rocks and it is a farm so there’s plenty of cow pats as well. We were 90% confident we would be able to find the balloon when it landed but it was always a possibility it would land in trees in the predicted landing area, this was a risk and we were thankful we got lucky.

Under canopy

More carbon monoxide poisoning as when followed the “jeep” back down the hill and surprisingly for a farm vehicle that operated in hilly countryside, it didn’t appear to have working brakes. Its primary means of stopping appeared to be selecting a low gear and just crashing in to things.

After showing the people at the farm some of the photos from the flight we jump back in the car and start heading back to Melbourne. There’s about 3000 photos to sort through. All in all lots of fun and a perfect day*.

Just before touch down

The raw image files of the highlights of the trip can be found here.

*The day was indeed perfect, however, the fish and chips we got on the way back were not and I had the wild shites for the next two days.