Weather Balloon Photography
Discussion
Early this year there was some talk here about taking photos/video from the stratosphere using cameras attached to a weather balloon. I remember at least one other member was going to do this. Did anyone? Will I be the first? ;-)
My own project is coming together nicely, and I'm hoping for a launch this weekend though the weather may have other ideas! Photos will be from a Canon compact using special firmware running an intervalometer script; video will from a Kodak camcorder. I'll be tracking it using 3 methods (to be sure to be sure) - a GSM/GPS tracker gadget, and separate GPS receiver connected to a flight computer transmitting the position via a small UHF transmitter and also via an old mobile phone. It will also have pressure and temperature sensors and a 3-axis accelerometer, for no better reason than I had spare pins on the computer doing nothing :-).
Dave
My own project is coming together nicely, and I'm hoping for a launch this weekend though the weather may have other ideas! Photos will be from a Canon compact using special firmware running an intervalometer script; video will from a Kodak camcorder. I'll be tracking it using 3 methods (to be sure to be sure) - a GSM/GPS tracker gadget, and separate GPS receiver connected to a flight computer transmitting the position via a small UHF transmitter and also via an old mobile phone. It will also have pressure and temperature sensors and a 3-axis accelerometer, for no better reason than I had spare pins on the computer doing nothing :-).
Dave
You need to get your application in to the CAA *at least* 4 weeks before launch. The guy there is friendly and helpful but I get the impression he's quite busy too! You can apply for several dates, in case the weather isn't good for your first date. Mine is for "every weekend in July, launch time between 10am and 1pm". He then issues a NOTAM (NOtice To AirMen) so anyone flying through the area has warning. You may well then get a call if you happen to be in someone's flightpath, if they're wondering whether to divert or not!
I ordered the Helium online, for collection from a local depot. There's a spreadsheet online so you can calculate how much you'll need. In my case with a 1kg payload and 1kg balloon I need about 3 cubic metres, depending on how quickly I want it to rise and what target height to get to. Many of the online sites describe their cylinders in terms of how many party balloons they will fill, and it's not always easy to get the volume in cu m. The site I chose did have all the information, but even so when I called the conversation started thus:
"How many balloons do you need to fill?"
"Just the one."
LOL. My cylinder has just over 5 cubic metres so I won't need to worry if I lose some when filling. Cost was a little under £100. Fortunately one of my customers wants me to do a sponsored launch for them, and they're paying for the helium and balloon for this one too :-).
Dave
I ordered the Helium online, for collection from a local depot. There's a spreadsheet online so you can calculate how much you'll need. In my case with a 1kg payload and 1kg balloon I need about 3 cubic metres, depending on how quickly I want it to rise and what target height to get to. Many of the online sites describe their cylinders in terms of how many party balloons they will fill, and it's not always easy to get the volume in cu m. The site I chose did have all the information, but even so when I called the conversation started thus:
"How many balloons do you need to fill?"
"Just the one."
LOL. My cylinder has just over 5 cubic metres so I won't need to worry if I lose some when filling. Cost was a little under £100. Fortunately one of my customers wants me to do a sponsored launch for them, and they're paying for the helium and balloon for this one too :-).
Dave
There's an online path predictor where you enter the launch site and date/time, plus the expected rise/fall rates and maximum altitude, then it looks at predicted wind speeds and direction and tells you roughly where to expect the balloon to land. IIRC that's what was shown in the video, but I could be wrong. It's certainly possible to log all the GPS locations (I will be) then plot those on a map or lpay back into Google Earth.
When I checked yesterday, the predicted path of my balloon for the weekend took it from the launch site (between Newbury and Wantage) out to Milton Keynes. Comparing that with the prediction I ran the day before, it's edging closer to landing at Silverstone. During qualifying. Which at least could get me on TV LOL! Seriously though, if the predictor has it landing anywhere unsafe then I'll postpone. It's also possible to adjust the amount of helium - more means it rises faster and pops at a lower altitude, thus it won't travel as far.
The prediction software doesn't start getting accurate till a couple of days before launch, so I'll decide on Thursday or Friday if it's OK to launch this weekend. I'll run it again on the morning of the launch, and then it should be accurate to within a few miles.
Dave
When I checked yesterday, the predicted path of my balloon for the weekend took it from the launch site (between Newbury and Wantage) out to Milton Keynes. Comparing that with the prediction I ran the day before, it's edging closer to landing at Silverstone. During qualifying. Which at least could get me on TV LOL! Seriously though, if the predictor has it landing anywhere unsafe then I'll postpone. It's also possible to adjust the amount of helium - more means it rises faster and pops at a lower altitude, thus it won't travel as far.
The prediction software doesn't start getting accurate till a couple of days before launch, so I'll decide on Thursday or Friday if it's OK to launch this weekend. I'll run it again on the morning of the launch, and then it should be accurate to within a few miles.
Dave
Sounds good, interested to see the results.
What happens when it lands on a roundbout in Milton Keynes!
Do you track the GPS info live and make an effort to chase it? I appreciate it will move faster than you can, but could you have a chase vehicle somewhere close to the predicted landing site?
What happens when it lands on a roundbout in Milton Keynes!
Do you track the GPS info live and make an effort to chase it? I appreciate it will move faster than you can, but could you have a chase vehicle somewhere close to the predicted landing site?
gary71 said:
What happens when it lands on a roundbout in Milton Keynes!
Gets eaten by one of the concrete cows? ;-)I ran the predictor for Sunday, and that has it landing only 13 miles from home - much more like it! I could then track the entire flight from somewhere high, like the Ridgeway :-)
gary71 said:
Do you track the GPS info live and make an effort to chase it? I appreciate it will move faster than you can, but could you have a chase vehicle somewhere close to the predicted landing site?
Yes, I'm hoping to track it live, and if I have a chance I'll try and be in the area to watch it land! Total flight time is around 2.5 hours so if it lands 100 miles away that's not going to happen, but 13 miles away should be do-able :-). We may well have a couple of vehicles in which case one can head for the expected landing site whilst the other perches on a hill with UHF receiver + antenna + laptop (the laptop decodes the signal and shows the balloon's current position, and heading).A really neat thing about the UHF scheme is that if you announce the launch on the UKHAS site, others in the area will tune in and decode the transmissions using a program that uploads the data to a web site. So I should know where the balloon is so long as someone listening has a good signal.
Dave
I can now report that .... It flew, reaching 98,810 feet, landed (hard) and recorded video throughout 
The stills camera didn't work - I think I stopped the script by touching a button when I was closing off the box. However the video worked right up to impact.
And I mean "impact". The balloon is supposed to shred into lots of pieces, but instead most of it stayed attached in one big lump, tying itself around the parachute. Fortunately although a 3' parachute would have been OK for the payload weight, I went for a 4' one in case this happened. It took just 27 minutes to descend (should have been 40), at an average speed of 42mph and an estimated final speed of 26mph. The fall did get broken a bit by a tree, but the impact was pretty hard regardless.
Tracking was great aside from a bug in my code where the longitude went wrong when it was east of the -1 line. So it jumped on the map from Berkshire to somewhere off the coast of Essex! People from around the country picked up the tracking signal (a piddly little 10mW transmitter), with one guy in Yorkshire getting a signal for most of the flight and someone in Northern Ireland joining in too!
The signal was lost when the balloon came down below people's horizons. We had a receiver in one of the chase cars, and went for the estimated landing area. Unfortunately the estimate was out by 8 miles (usually it's much better than that, I'm told) and we lost the signal. Eventually we found somewhere with 3G coverage and I went online to grab the latest position anyone had, which was at a height of 4000 metres. So we aimed for there, all the time listening to static on the receiver (connected to magmount aerial on the roof). At one spot the static stopped for a second and I heard the familiar warble of the balloon's data stream. So I shouted for the car to stop, then got out the Yagi aerial which is much more sensitive and very directional. That picked up the signal and gave us a direction. The signal sounded weak but I connected to the laptop decoder and got a clear GPS location of where the balloon thought it was. I could see that the balloon's GPS had failed at some point and the flight computer was retransmitting the same time and position again and again. The computer had rebooted so I guessed from that the landing was a hard one.
So we got back in the car, connected up the magmount aerial, and headed for the last known GPS position. As we got closer the signal got stronger and stronger. We parked up nearby and walked down the hill, with me waving the Yagi around to get a direction fix. It didn't take long to locate the rough position, and it seemed to be in someone's back garden. Fortunately the owner was there and after I explained why 7 geeks had descended on him waving an aerial and carrying a warbling radio device, he was happy to help.
Everywhere around there were trees, Very tall trees. It seemed hardly possible that the payload could get past those to the ground. Anyway, we walked around the (very large, think "small forest") garden trying to get a location, when I heard a separate whistle. I switched the receiver off and could still hear it - the piezo bleeper I put on the payload box! Then we just wandered around trying to get an audible fix on where it was coming from. As it got louder we spotted the box in the garden next door. RESULT !!!
So, here's the view from 98,810 feet:
The Highest Point by daveake, on Flickr
and more in my flickr page:
Flickr
The stills camera didn't work - I think I stopped the script by touching a button when I was closing off the box. However the video worked right up to impact.
And I mean "impact". The balloon is supposed to shred into lots of pieces, but instead most of it stayed attached in one big lump, tying itself around the parachute. Fortunately although a 3' parachute would have been OK for the payload weight, I went for a 4' one in case this happened. It took just 27 minutes to descend (should have been 40), at an average speed of 42mph and an estimated final speed of 26mph. The fall did get broken a bit by a tree, but the impact was pretty hard regardless.
Tracking was great aside from a bug in my code where the longitude went wrong when it was east of the -1 line. So it jumped on the map from Berkshire to somewhere off the coast of Essex! People from around the country picked up the tracking signal (a piddly little 10mW transmitter), with one guy in Yorkshire getting a signal for most of the flight and someone in Northern Ireland joining in too!
The signal was lost when the balloon came down below people's horizons. We had a receiver in one of the chase cars, and went for the estimated landing area. Unfortunately the estimate was out by 8 miles (usually it's much better than that, I'm told) and we lost the signal. Eventually we found somewhere with 3G coverage and I went online to grab the latest position anyone had, which was at a height of 4000 metres. So we aimed for there, all the time listening to static on the receiver (connected to magmount aerial on the roof). At one spot the static stopped for a second and I heard the familiar warble of the balloon's data stream. So I shouted for the car to stop, then got out the Yagi aerial which is much more sensitive and very directional. That picked up the signal and gave us a direction. The signal sounded weak but I connected to the laptop decoder and got a clear GPS location of where the balloon thought it was. I could see that the balloon's GPS had failed at some point and the flight computer was retransmitting the same time and position again and again. The computer had rebooted so I guessed from that the landing was a hard one.
So we got back in the car, connected up the magmount aerial, and headed for the last known GPS position. As we got closer the signal got stronger and stronger. We parked up nearby and walked down the hill, with me waving the Yagi around to get a direction fix. It didn't take long to locate the rough position, and it seemed to be in someone's back garden. Fortunately the owner was there and after I explained why 7 geeks had descended on him waving an aerial and carrying a warbling radio device, he was happy to help.
Everywhere around there were trees, Very tall trees. It seemed hardly possible that the payload could get past those to the ground. Anyway, we walked around the (very large, think "small forest") garden trying to get a location, when I heard a separate whistle. I switched the receiver off and could still hear it - the piezo bleeper I put on the payload box! Then we just wandered around trying to get an audible fix on where it was coming from. As it got louder we spotted the box in the garden next door. RESULT !!!
So, here's the view from 98,810 feet:
The Highest Point by daveake, on Flickr
and more in my flickr page:
Flickr
Edited by daveake on Monday 11th July 18:10
Gassing Station | Photography & Video | Top of Page | What's New | My Stuff