Hi, I'm working with my students on a gpio coding project and the idea we've decided on was to use the Raspberry Pi to control, monitor, and record a voyage into the upper atmosphere. We've done quite a lot of research and seen other similar projects using digital cameras but we'd like to take it a step further.
We're looking for information from someone who is familiar with very high-altitude requirements: temperature expected during transit, effects of radiation (?), effects of near vacuum on electronic components, etc. It seems feasible as we've seen some projects in the UK where students have sent digital cameras to the upper atmosphere using weather balloons.
The idea is to use a weather balloon to get the craft to the maximum altitude when the ballon would normally not provide lift. Then using GPS interface to determine altitude, detach the tether to the ballon.
The raspberry pi would store and execute (using gpio) things like GPS input, collect temperature data, and take some photos of course (perhaps by controlling a digital camera).
As it's been done several times by hobbyists using weather balloons it seems like a feasible and interesting project for the budding computer scientists I'm working with.
We've received some clearance from the local aviation authorities to operate in a specific zone.
Recovery is not in the planning (so trying to keep costs low!).
-
mahjongg
- Forum Moderator
- Posts: 12407
- Joined: Sun Mar 11, 2012 12:19 am
- Location: South Holland, The Netherlands
Re: operation in extreme environments
CCD cameras will be susceptible to cosmic radiation, somewhat, but not too much to be unuseable.
Not too concerned about radiation problems with standard logic, but use a watchdog system to reset the system if it goes haywire.
Remember in a near vacuum there is no convection cooling, only way to loose heat is by radiating infrared energy, so you will probably need a heat-pipe to remove energy from the SoC, obviously you need no Ethernet so you should remove te LAN9512 chip, (for heat and power consumption reasons) the SoC would still have one USB port in case you need one..
Not too concerned about radiation problems with standard logic, but use a watchdog system to reset the system if it goes haywire.
Remember in a near vacuum there is no convection cooling, only way to loose heat is by radiating infrared energy, so you will probably need a heat-pipe to remove energy from the SoC, obviously you need no Ethernet so you should remove te LAN9512 chip, (for heat and power consumption reasons) the SoC would still have one USB port in case you need one..
Re: operation in extreme environments
I'm curious why you don't plan a recovery. Many people do high-altitude balloon experiments and routinely recover the payload, even in very low budget projects. If you've got a radio transmitter onboard, it's that much easier. I guess you must have a transmitter, otherwise there would be no point to taking photos if you're not going to recover the camera ?
Re: operation in extreme environments
great info, thanks and very helpful! recovery not in planning as not sure yet if they can get it done. It's a summer school project so am trying to hold them back a bit as these kids just want it to do everything. probably should break this up into phases to give them something that is doable before september. maybe just something simple like have it record some data, store it on the sim, test it before launch and give them the grade based on pre-launch test, then launch it... recovery seems very difficult to do before September rolls around.
so, far we've got a temperature sensor that seems to work. then trying to get it to read data from a gps...
at a high level how would you go about getting it to simply read gps data (altitude) and then use that in the logic. Perhaps simply a usb connection to the garmin device then write drivers on debian to read the data?
these kids are brilliant so I'm confident they can get this to work!
so, far we've got a temperature sensor that seems to work. then trying to get it to read data from a gps...
at a high level how would you go about getting it to simply read gps data (altitude) and then use that in the logic. Perhaps simply a usb connection to the garmin device then write drivers on debian to read the data?
these kids are brilliant so I'm confident they can get this to work!
Re: operation in extreme environments
we've identified a product that will shield the RPi from cosmic rays!
http://www.hybridplastics.com/docs/tds/neushield.pdf
provides both x-ray and gamma ray shielding up to 95% and sold in glue gun sticks (or components) or tape (to cover wiring).
the temperature problem can possible be controlled using dry-ice? Perhaps packing the RPi in styrofoam w/ layer of dry ice surrounding the unit?
http://www.hybridplastics.com/docs/tds/neushield.pdf
provides both x-ray and gamma ray shielding up to 95% and sold in glue gun sticks (or components) or tape (to cover wiring).
the temperature problem can possible be controlled using dry-ice? Perhaps packing the RPi in styrofoam w/ layer of dry ice surrounding the unit?
Re: operation in extreme environments
When I saw this thread it reminded me of a Youtube video I'd seen some time ago. I did a quick search and it seems that quite a few private individuals have launched weather balloons (and captured great footage!).evan wrote:we've identified a product that will shield the RPi from cosmic rays!
http://www.hybridplastics.com/docs/tds/neushield.pdf
provides both x-ray and gamma ray shielding up to 95% and sold in glue gun sticks (or components) or tape (to cover wiring).
the temperature problem can possible be controlled using dry-ice? Perhaps packing the RPi in styrofoam w/ layer of dry ice surrounding the unit?
http://www.youtube.com/watch?v=y6ZMscMp8UM
Rather than reinventing the wheel (and guessing stuff) I'd advise having a good search through the internet to see what problems (and solutions) others have found. Keeping a battery warm enough to function for the duration of the ride might be one of the most immediate.The guys in the video above used handwarmers. Ensuring chip cooling without convection? Might not be a problem, but who knows (unless you do datalogging!) For instance, a suitably shaped thin aluminium sheet heatsink could harvest heat from the chips and dump it to your battery. Cosmic rays? More guesswork, but do some research and if you decide they might be a problem then that heatsink (or a bit of carefully placed lead foil) etc. would keep it simple.
Anyway, let us know how it goes!
-
secretagent
- Posts: 36
- Joined: Wed Mar 07, 2012 10:09 am
Re: operation in extreme environments
Note that there are some ambiguous regulations meant to prevent commercial GPS units from being used for missile guidance and so many units will not work properly at high altitudes. You may want to take a look at http://showcase.netins.net/web/wallio/G ... s60kft.htm for receivers that have been verified to work.
Re: operation in extreme environments
I've flown several high altitude balloon payloads. Usually people choose simple processors such as Atmels and PIC (the various Arduinos are very popular). For data logging and telemetry, those are plenty powerful enough. They are also cheap and require little power.
The Raspberry Pi isn't ideal for the purpose, partly because of its relatively high power requirement (2W vs maybe 50mW), and because the main thing you want in a flight computer is for it to be reliable, and you gain that by simplicity. On an Arduino there's so little to go wrong - no SD card, no USB, etc.
All of that said, there's no reason for not using a Pi if you want to use one. You do get a far more powerful processor, with much more memory, and with easily available SD file storage.
I'm flying a Raspberry Pi payload quite soon. I made it for 2 reasons. One reason is that the USB provides access to cheap webcams, meaning I can send down live images with the usual telemetry. The other reason is .... "because"
. My pi was sitting there just begging to be used for something interesting!
Now onto your questions:
"temperature expected" - outside the pyload it will get down to -50 degrees or so. Inside however is another matter. With good insulation (EPS), the heat within from your batteries/regulators/Pi will keep it quite warm. I doubt it will go much below zero.
"effects of radiation" - never heard of any issues.
"effects of near vacuum on electronic components" - none (not for the time they're up there, anyway.
"Then using GPS interface to determine altitude, detach the tether to the ballon." - That's possible, but mostly people just rely on the (latex) balloon bursting, usually at an altitude of 30km or so. It is possible to detach from the balloon, typically using some nichrome wire to cut the cord. Remember this has to be *above* the parachute for obvious reasons!
"The raspberry pi would store and execute (using gpio) things like GPS input, collect temperature data, and take some photos of course (perhaps by controlling a digital camera)."
All OK. I have GPS and webcam on mine. I'm not bothering with other sensors this time. Most people use a Canon digital camera with the CHDK firmware which then runs an intervalometer script to automate the picture taking, in which case there's no need to interface a camera to the Pi. In my case I want live images this time, so I'm using a webcam. There's not enough bandwidth to download large images during the flight hence no need to use anything of decent quality.
The Raspberry Pi isn't ideal for the purpose, partly because of its relatively high power requirement (2W vs maybe 50mW), and because the main thing you want in a flight computer is for it to be reliable, and you gain that by simplicity. On an Arduino there's so little to go wrong - no SD card, no USB, etc.
All of that said, there's no reason for not using a Pi if you want to use one. You do get a far more powerful processor, with much more memory, and with easily available SD file storage.
I'm flying a Raspberry Pi payload quite soon. I made it for 2 reasons. One reason is that the USB provides access to cheap webcams, meaning I can send down live images with the usual telemetry. The other reason is .... "because"
. My pi was sitting there just begging to be used for something interesting!Now onto your questions:
"temperature expected" - outside the pyload it will get down to -50 degrees or so. Inside however is another matter. With good insulation (EPS), the heat within from your batteries/regulators/Pi will keep it quite warm. I doubt it will go much below zero.
"effects of radiation" - never heard of any issues.
"effects of near vacuum on electronic components" - none (not for the time they're up there, anyway.
"Then using GPS interface to determine altitude, detach the tether to the ballon." - That's possible, but mostly people just rely on the (latex) balloon bursting, usually at an altitude of 30km or so. It is possible to detach from the balloon, typically using some nichrome wire to cut the cord. Remember this has to be *above* the parachute for obvious reasons!
"The raspberry pi would store and execute (using gpio) things like GPS input, collect temperature data, and take some photos of course (perhaps by controlling a digital camera)."
All OK. I have GPS and webcam on mine. I'm not bothering with other sensors this time. Most people use a Canon digital camera with the CHDK firmware which then runs an intervalometer script to automate the picture taking, in which case there's no need to interface a camera to the Pi. In my case I want live images this time, so I'm using a webcam. There's not enough bandwidth to download large images during the flight hence no need to use anything of decent quality.
Re: operation in extreme environments
As planned I flew my Raspberry Pi on Saturday, with the balloon bursting at an altitude of 39,994 metres (nearly 25 miles). It sent down some amazing photos during the flight. I've written the project up at http://www.daveakerman.com/?p=592.
Dave
Dave
-
liz
- Raspberry Pi Foundation Employee & Forum Moderator
- Posts: 5202
- Joined: Thu Jul 28, 2011 7:22 pm
- Contact: Website
Re: operation in extreme environments
Dave, that's *crazy* and brilliant and we love it. Thanks so much for sharing! I'll be featuring it on the front page later this week. Just out of interest - does that mean that a Raspberry Pi now holds the UK amateur record for altitude?
Director of Communications, Raspberry Pi
Re: operation in extreme environments
No, it's currently at #12 - the full table is here - http://ukhas.org.uk/general:uk_records.
I happen to hold the #2 position, using a very light payload and a larger balloon. That's the combination needed for a record attempt. My RPi flight had 3 separate payloads so lots of weight, so it was never going to break the record!
I happen to hold the #2 position, using a very light payload and a larger balloon. That's the combination needed for a record attempt. My RPi flight had 3 separate payloads so lots of weight, so it was never going to break the record!
Re: operation in extreme environments
Great report and some brilliant pictures. There's lots of information which will doubtless be relevant to builders of all sorts of other project too. Bloody well done!
Re: operation in extreme environments
I should have added that the Raspberry Pi does hold the "highest live amateur images record", worldwide, having taken and transmitted images during the entire flight which peaked at 39,994 metres.liz wrote:Dave, that's *crazy* and brilliant and we love it. Thanks so much for sharing! I'll be featuring it on the front page later this week. Just out of interest - does that mean that a Raspberry Pi now holds the UK amateur record for altitude?
. Of the flights that went higher, only one carried a camera and that just stored on SD card and did not transmit the images.
Last edited by daveake on Mon Mar 11, 2013 11:05 am, edited 1 time in total.
Re: operation in extreme environments
Out of curiosity, how long does it take to beam each (low-res) image back to ground?
Re: operation in extreme environments
Between 3.5 and 5 minutes each one - a bit better than I was expecting. The images are all jpeg compressed and that makes short work of black skyAndrewS wrote:Out of curiosity, how long does it take to beam each (low-res) image back to ground?
.You can see the full set on the site that they got uploaded to - http://sanslogic.co.uk/ssdv/live
Re: operation in extreme environments
A work of genius! 