Pi in the sky

18th of July 2012 by liz

I think more of you have emailed me about this than about anything else that anybody has ever done with a Raspberry Pi.

What you’ve just looked at are, we think, the highest ever photographs transmitted live from an amateur device in the UK world. Dave Akerman hooked a Raspberry Pi with a webcam and GPS up to a hydrogen balloon, which got nearly 40km up (39,994m, to be precise) before bursting. This means that Dave’s is the first Raspberry Pi to visit near space (it returned unharmed, and Dave was able to recover it), and also means that Eben does not have to eat that hat he mentioned.

"And if you introduce a lighted spill to the neck of the test tube..."

Payload ready to go

If you want to learn more, visit Dave’s blog, where he has documented the flight minutely, to find out how he did it. He’s also put up a couple of Flickr sets: one full of images sent from the Raspberry Pi on its journey, and one of photos from the ground. Here at the Foundation, we’ve been reading his page on the project over and over, jaws on the floor. Well done Dave. We’re sending you a big virtual round of applause.

News Archives

- May 2013
- April 2013
- March 2013
- February 2013
- January 2013
- December 2012
- November 2012
- 2013
- 2012
- 2011

Supported by

In the forums

agreene, 1 minute ago: Fedora • Re: Pidora 18 (Raspberry Pi Fedora Remix) Release
tedsluis wrote:agreene wrote:In response to headless-mode not working I just tested and confirmed that it does work although admittedly it might be a little bit confusing because once headless-mode is enable, the run level remains at 5. So it might…
box, 4 minutes ago: Русский • Re: резервная копия
april22 wrote:в пути назначения понятно . а вот источник - как указать что весь диск (sd карта ) .Возможно уже не надо, но все же...Насколько я понял проблема только в том чтобы выяснить источник (source).Выполни :----------------------Code: # fdisk -lDisk /dev/sda:…
gordon@drogon.net, 6 minutes ago: Automation, sensing and robotics • Re: Programming for PiFace to support relay…
stlouistechy wrote:Thanks Gordon, not sure why it would not execute the .c program without sudo, but a fresh make after some additional edits to the file seemed to have cleared it up and I can now launch the program without…
forumisto, 8 minutes ago: Español • Re: Ventilación para uso 24x7
controla la temperatura de los chipslas cajas transparentes suelen tener tener unos agujeros, pero aún así se nota un aumento considerable en la temperatura.yo huiría de esas cajas y buscaría alguna otra algo más grande y pondría algún ventilador de…
Sandgrounder, 8 minutes ago: Troubleshooting • Re: Headless Model A
FTrevorGowen wrote:Interesting ... see my experiences in this thread viewtopic.php?t=35959.What USB LAN dongle are you using?The USB LAN dongle is an SMC EZconnect. It works perfectly when the HDMI monitor is connected.Your experiences in the other thread look well worth…

Follow @Raspberry_Pi

On Twitter

unclewilco, 8 hours ago
@Raspberry_Pi met a bloke last night in the pub who works making the Pi in Sony Bridgend - he was so passionate about it #credit a Shed, South Wales, UK
ModMyPi, 9 hours ago
We've got a load of those elusive @Raspberry_Pi camera modules in stock! All ready for immediate Worldwide Dispatch! https://t.co/poKsTQUaAF Exeter
Raspberry_Pi, 12 hours ago
Only 2 days left to enter Andrew Edney's competition to win a copy of the Raspberry Pi for Dummies books: http://t.co/seP1EckWuM Good luck! Cambridge, UK
FOSSwiki, 20 hours ago
announcement by @Fedora of Pidora 18 for @Raspberry_Pi computers http://t.co/pfMQv69P1d #linux #opensource #maker http://t.co/XeNGDPrQG5 San Francisco

Friends of Pi

48 comments

Rick on
July 18, 2012 at 3:34 pm said:

TheRegister will be all over this! Looks like a much more interesting (and cheaper) method than their ‘LOHAN’ and ‘PARIS’ launches – though PARIS did produce some fantastic photos!

Comment ↓
- liz on
  July 18, 2012 at 3:36 pm said:
  
  I think they actually got to it yesterday, before we did! http://www.theregister.co.uk/2012/07/17/pi_ascent/ (We had a bit of a hold-up here because it takes such a long time to push new images out to the mirrors and we really wanted to post about Raspbian yesterday.)
  
  Comment ↓
Guilherme de Sousa on
July 18, 2012 at 3:38 pm said:

Finally a post about this great project!

I think this video is missing here, can you add it Liz?
http://www.youtube.com/watch?v=KGLB9-LdpYM&feature=youtu.be

best regards !

Comment ↓
- liz on
  July 18, 2012 at 3:57 pm said:
  
  That video, sadly, wasn’t taken with the Raspberry Pi, but with a GoPro Hero 2 camera (attached to the same balloon), which is why I haven’t included it in the blog post.
  
  Comment ↓
  - Dave Akerman on
    July 18, 2012 at 5:26 pm said:
    
    I’m thinking about things to add for the next flight, and having the Pi take video footage (possibly using a second webcam, or that new camera) is on the list.
    
    Comment ↓
    - liz on
      July 18, 2012 at 5:33 pm said:
      
      The camera board would be a good option; it’s very light. I remain pathetically excited about the whole thing; please mail me next time you try something!
      
      Comment ↓
    - Gert van Loo on
      July 19, 2012 at 10:08 am said:
      
      Dave, just contact us before you send up your next balloon. Even if there is not yet an official camera, we might be able to have enough trial SW so we can lend you our 14MP prototype. (And yes, at the moment there is only ONE prototype). Also I am looking at reducing power consumption by (hopefully) 40% so you can use smaller batteries.
      
      Comment ↓
  - Guilherme de Sousa on
    July 18, 2012 at 7:16 pm said:
    
    That’s true, but they only got the camera back because the Pi was transmiting the GPS location:P
    
    Comment ↓
    - JBeale on
      July 18, 2012 at 10:24 pm said:
      
      great project, very inspirational! But my reading of the writeup is the Pi’s GPS got stuck at the launch location, and they got the GPS location readings from one of the other two packages on the string.
      
      Comment ↓
      - Dave Akerman on
        July 19, 2012 at 9:46 am said:
        
        Correct. My fault – I knew from testing I had to keep the GPS receiver away from everything else. After assembling the payload it got a lock very quickly, but I should have checked afterwards that it hadn’t lost it.
        
        A friend (Anthony, who launched with me) is designing a PCB now to hold the GPS and radio, with options for SPI and one-wire sensors. That will connect to the Pi via a ribbon cable, and it will have a much better GPS receiver, so this issue will disappear.
      - JBeale on
        July 19, 2012 at 3:31 pm said:
        
        Given the comment about heat build up inside the R-Pi package, also thinking about a bimetallic strip (simple thermostat) to open a window for cooling as needed.
Jörg on
July 18, 2012 at 4:03 pm said:

Oh, really? Vertical video?
http://www.youtube.com/watch?v=Bt9zSfinwFA

Comment ↓
- Dio on
  July 18, 2012 at 4:21 pm said:
  
  Very good – off topic slightly, but you need to laugh now and again – Bookmarked!
  
  Comment ↓
- Dave Akerman on
  July 18, 2012 at 5:29 pm said:
  
  In fairness to my better half Julie, who took the video, it did help get that long flight train in the video.
  
  LOL at the YT video BTW.
  
  Comment ↓
Dio on
July 18, 2012 at 4:26 pm said:

I watched this from a link in the earlier thread and was impressed, though one thought crossed my mind – do you need to get air clearance to launch one of these balloons?

Then whilst watching the embedded video an aircraft can quite clearly be heard flying over – I would hate to think what damage a balloon sucked into a jet engine could do!

Comment ↓
- liz on
  July 18, 2012 at 4:29 pm said:
  
  You do, and Dave discusses it in the very first paragraph of his blog post about the project. http://www.daveakerman.com/?p=592
  
  Comment ↓
  - Dio on
    July 18, 2012 at 4:39 pm said:
    
    Sorry, I skim subjects and immediately question things without looking – must do better in future
    
    Comment ↓
- AndrewS on
  July 18, 2012 at 6:03 pm said:
  
  “I would hate to think what damage a balloon sucked into a jet engine could do!”
  Probably not that much? They have to be able to withstand http://en.wikipedia.org/wiki/Bird_strike
  
  Additionally, latex is quite soft, and a quick bit of google-searching says that latex melts at 250-350F but jet fuel burns at 500-600F
  
  Comment ↓
  - Biggles on
    July 19, 2012 at 2:04 pm said:
    
    Yes, but the ballon is also filled with hydrogen…
    
    Comment ↓
    - Norman Dunbar on
      July 19, 2012 at 2:30 pm said:
      
      Hydrogen? Helium surely.
      
      I know it says hydrogen in the into from Liz, but I watched the video yesterday and I’m sure it was mentioned that it was helium.
      
      Hydrogen is too “Hindenberg”!
      
      Cheers,
      Norm.
      
      Comment ↓
      - Dave Akerman on
        July 19, 2012 at 3:54 pm said:
        
        It was hydrogen, and the Hindenburg disaster was not a fault of the hydrogen.
      - liz on
        July 19, 2012 at 4:18 pm said:
        
        Wasn’t it to do with the lacquer they’d used on the shell? Fascinating stuff (and ruined hydrogen, which got the blame, for everyone else).
  - Kev on
    July 20, 2012 at 8:26 am said:
    
    Yes, I did the 3D animation for Rolls Royce when they launched the Trent engine (many moons ago…), and they even go to the lengths of shooting them the right way round so that the breast bone goes in first as it’s the hardest part.
    
    Comment ↓
Ronald on
July 18, 2012 at 4:34 pm said:

Rapsberry heaven
http://www.youtube.com/watch?v=I37i8wvFpXw

Comment ↓
Davespice on
July 18, 2012 at 4:49 pm said:

OH EM GEE!! That’s bloody brilliant!! =)

Comment ↓
Will on
July 18, 2012 at 6:12 pm said:

Google did this with Android a while ago (http://android.hibal.org/) but I have been wanting to try it myself as well. Just got my board in the mail yesterday and it’s very cool! Can’t wait to see what kind of fun things I can do with it. Also looking forward to seeing a production version of that onboard/ribbon camera

Comment ↓
- Dave Akerman on
  July 18, 2012 at 7:01 pm said:
  
  Yes, the Android thing was an interesting project, though phones didn’t really do much themselves – the tracking and photography were done by other devices.
  
  The video where an Android toy robot gets sent by a flailing nylon cord to its doom is a classic.
  
  Comment ↓
MuddyDogs on
July 18, 2012 at 7:41 pm said:

Brilliant!!
To boldly go where no Raspberry Pi has gone before
(sorry, couldn’t resist)

Comment ↓
Chewy on
July 18, 2012 at 7:59 pm said:

Well done Dave an excellent achievement.

Comment ↓
SN on
July 18, 2012 at 8:19 pm said:

One word. Awesome.

Comment ↓
Tom on
July 18, 2012 at 8:44 pm said:

I think the project Horus boys just edge out on the altitude record: http://projecthorus.org/?page_id=1858. It’s almost too close to call

Comment ↓
- liz on
  July 18, 2012 at 9:12 pm said:
  
  It’s altitude plus live photos that we believe Dave now has the record for. Not many people have done that yet, because until now it’s not been easy to do it cheaply.
  
  Comment ↓
  - Dave Akerman on
    July 18, 2012 at 9:33 pm said:
    
    Exactly so. Previously I’ve flown Arduinos but you can’t plug a USB webcam into one of those. You can get serial cameras but one of those is twice the price of a Pi, and then you have to find a processor with enough power and memory to do the conversion from jpeg to the download format.
    
    The highest transmitted images previously worldwide were at about 35km (a flight in Holland), so unless there’s a flight I’m not aware of then the Pi got the highest transmitted images by 5km.
    
    The absolute altitude record is at 43721m, held by Steve Randall (if you’ve seen any of these flights on TV, Steve was probably behind it). He and I flew altitude attempts on the same day, and I first broke his previous record by a 1km before he took the record back by 82m. I held that record for 5 minutes 40 seconds, but I’m not bitter
    
    Comment ↓
    - liz on
      July 18, 2012 at 10:00 pm said:
      
      Just out of interest, would a significant weight reduction in the payload have an impact on altitude? I’m thinking heatsink/battery weight – we’re wondering if there’s anything we can do to help you go higher.
      
      Comment ↓
      - Dave Akerman on
        July 18, 2012 at 10:17 pm said:
        
        My switched-mode power modules arrived today, and they will help on that side. When I get time I’ll do some tests to see what I can get the batteries down to.
        
        Every gram costs (depending on who you believe) 1 – 3 metres in altitude, so the odd few grams here or there don’t make an awful lot of difference (says the person who missed the record by 82m!).
        
        There’s no doubt that a Pi could be flown a lot higher than 40km – remember that that was achieved with a medium size balloon and 2 other payloads hanging below!
  - John - K7VE on
    July 18, 2012 at 10:02 pm said:
    
    Not quite the record, but a great achievement.
    http://www.arhab.org/
    
    Here is another interesting flight: http://www.arrl.org/news/amateur-radio-balloon-flight-crosses-atlantic-sets-records
    
    The Raspberry Pi is getting a following in Amateur Radio. I have built .deb files for D-STAR (a Digital Radio system in Amateur Radio) — http://nw7dr.dstar-relay.net/debian
    
    Comment ↓
Lluís on
July 18, 2012 at 9:26 pm said:

Brilliant! Next step, unmanned drones with our nice device

Comment ↓
- AndrewS on
  July 19, 2012 at 12:01 am said:
  
  Y’mean something like this?
  http://www.raspberrypi.org/phpBB3/viewtopic.php?f=63&t=8202
  
  Comment ↓
Kev on
July 19, 2012 at 9:53 am said:

Reading this with great interest (at work and radio starts playing Star Man by Bowie. Nice coincidence.
I still haven’t found time to get the 2 buttons and some LED’s system going – I have a long way to go I think.

Comment ↓
Ian Davidson on
July 19, 2012 at 3:19 pm said:

How were the images transmitted live? By what means?

Comment ↓
- Dave Akerman on
  July 19, 2012 at 3:51 pm said:
  
  Using 300 baud rtty over a 10mW 434.65MHz radio link to a network of enthusiasts each using an amateur radio receiver hooked up to a PC to decode the signals. The entire infrastructure was built by other enthusiasts in the UKHAS group, including students at the Cambridge University SpaceFlight and Philip Heron who wrote the code to encode and decode the image data, and I’m indebted to all of them. More details on the image encoding at http://ukhas.org.uk/guides:ssdv and in the tracking system in general at http://ukhas.org.uk/guides:tracking_guide.
  
  Comment ↓
  - Ian Davidson on
    July 19, 2012 at 9:37 pm said:
    
    Thank you
    
    Comment ↓
reiuyi on
July 20, 2012 at 3:29 pm said:

This is worth its weight in gold

Up until a couple of years ago, no amateur would even dream of a live motion-pictures feed from near space

Comment ↓
- Barroon on
  July 22, 2012 at 6:33 pm said:
  
  Just to be pedantic, radio hams in the US were sending back live TV from weather balloons in the late 80s. Definitely more technical foo required for that back then, granted
  
  Comment ↓
Davethewave on
July 22, 2012 at 8:05 pm said:

This is what they means when talking about “computing on the cloud”???
Great!

Comment ↓
Tony on
July 26, 2012 at 12:30 pm said:

An alarming lack of basic physics/maths knowledge displayed by these two. Always climb a gate at the hinge end. Principle of levers.

I’m guessing there was a good reason for not just opening it.

Comment ↓
- Dave Akerman on
  July 27, 2012 at 7:43 pm said:
  
  The gate lock is jammed and doesn’t open.
  
  Thanks for the input.
  
  Comment ↓
carlos on
August 4, 2012 at 6:08 pm said:

Lo único que puedo decir es fantástico, felicitaciones

Comment ↓

Buy a Pi!

News Archives

Supported by

In the forums

On Twitter

Friends of Pi