**** 2005-01-11:

For some time now I've been dreaming about putting up my own "UAV balloon", flying over Sweden and maybe even one or two other nordic countries, while sending back data about it's position.

Not because it's useful for anything, but because it's my idea of fun.

What I'd need is:
1) one or more balloons filled with helium.
2) an "on-board" computer, with custom software (preferably Linux-based with USB, to make enabling/disabling of GPS and cellphone easy. Plus maybe a serial port for other things such as driving a servo?).
3) a GPS (which will also work as an altimeter).
4) a camera to send back digital postcards from it's voyage.
5) a cellphone.
6) a battery

and maybe also

7) suncells for re-charging the battery/batteries.
8) a servo and weights, to be able to keep it in flight even as the amount of helium in the balloon(s) decrease.
9) a parachute, to save the electronics (and people?) when the balloon finally fails.
10) a radar reflector, so airplanes and traffic controllers will see it. This shouldn't be needed if I manage to find an aluminium balloon though.

Of course all of these things would need to function at an altitude of 6.000-8.000 meters, where the pressure is ~0,5 atmospheres and the temperature is ~-40 degrees C (-40 degrees F) or even colder.

For a reference on radars and aluminium, see: http://en.wikipedia.org/wiki/Chaff_(radar_countermeasure)

Oh, and just in case someone outside little Sweden wants to read about this, I'll try to keep it all in english. =)

**** 2006-01-17:

My current thoughts on data transfer via GSM is that the "standard" 9600 baud will probably be best for sending home data; not GPRS, 3G or similar. This is because the faster the data goes, the closer you need to be to a base station. I have historically had alot of problems accessing GPRS while normal GSM calls work fine.

Did some research on use of cellphones at high altitudes. Found a couple of discouraging links.


According to this, you can't expect any calls to go thru above ~8.000 feet (~2.500 meters).

I found some information about the possibility of analogue (NMT?) systems being more reliable at altitude though. Hmm.

**** 2006-01-18:

I looked around for an onboard computer which can work at ~-40 C.
As it needs to be very small, lightweight and energy-efficient, finding one is actually harder than I thought.

This far, I have only found one supplier that looks promising; Arcom. - http://www.arcom.com/

As far as batteries go, I have done some reading about temperatures, function and self-discharge.
Se for instance: http://www.batteryuniversity.com/partone-15.htm
If I don't go with a rechargable battery (recharge via suncells), I can buy a Lithium battery which can function as far down as -60 C. Those are available easily & cheaply from the swedish electronics supplier ELFA.

I also looked at GPS:es. Looks like that's not going to be a problem as long as my "onboard computer" has USB host functionality. I found several USB-GPS'es which functions at -40 C at swedish supplier Dustin, as cheaply as 545 SEK even (about ~56 Euro).

For an onboard camera, I'm leaning towards a "laptop webcam". This is because it gets it's power from USB, it's lightweight, and you don't need an external way of turning the camera on, getting the pictures to the computer, etc. Plus there are no mechanically operated parts which might fail. I doubt webcams are intended to be operated at -40 C though...

**** 2006-01-24:

I found a few other embedded Linux boards with USB host. Some even seem to be priced at less than $180 USD for a version capable of working at -40 C.
And it has a built-in watchdog timer. Might be a good idea, since a manual reset will hardly be possible once it's airborne... "Options" include an industrial grade Flash memory for $51, too. Might be a good choise?

But I might need a "developer's board" aswell to get it running? That costs about $180 USD too, though. Ugh.

From what I can read out of resources on the net, the embedded board should have a RedBoot loader in flash when you get it. It even has an emergency "boot from COM port" jumper for emergency restores.
Should be pretty plug & play, then?
See for instance: http://www.comp.leeds.ac.uk/jj/linux/arm-sbc.html

**** 2006-01-25:

I might aswell write down some of the thoughts I previously had on being able to "stear" the balloon and control it's altitude.

Thoughts on stearing:
1) Electric motors - too energy consuming for an online battery and even for solar cells.
2) Stearing by shaping the payload as a "glider airplane", which can use it's airspeed for rudder control - will hardly work as the balloon will typically have the same speed as the air surrounding it.
3) Stearing by changing altitude, since the atmosphere typically moves in "layers", with different speed and direction.

Thoughts on controlling altitude (and altitude-dependent stearing):
1) Electric motors - not a good choice, see above.
2) Expanding or deflating the balloon by a helium pump - a pump solution might be complex and likely leak helium, plus a needed "online helium container" would probably weight too much.
3) Expanding or deflating the balloon by heat control inside the balloon - too complex, to energy consuming.
4) Controlling height by having a "water collector" which extracts the water from the air surrounding the balloon with a little help of sunlight, and then dumps the water when it's time to gain altitude - might be hard, as it is typically -40 C at 8.000 meters...
5) Simple control using only some weights which can be released by a servo - not renewable, can not be used to descend. But relatively simple.

Currently I am planning to use altitude method "5" (and having something which will not harm or pollute anything on the ground on impact), and programming the onboard computer to release a payload if altitude gets below ~1.500 meters, or do an emergency drop of several payloads if altitude goes below ~1.000 meters.

**** 2006-01-27:

I did some thinking about possible water-related problems at high altitude... First, if some electronical parts within the payload manages to push the temperature above 0 C (somewhat depending on pressure), water might condensate onto the stuff in the payload. Therefore, I should try to keep the payload box as air-tight as possible. Second, as the payload will need a "window" for the camera, that could become covered with ice crystals effectively blocking sight for the camera. As far as I know, these crystals don't form on surfaces colder than ~0 C. So I should make sure the payload is already colder than 0 C when it is released, and I should place the window so that the surface is not warmed by sunlight, to avoid this.

Hmm I as thinking it's time to put this log online now. Now if I only had a "working name" for the project... I did some googling on different combinations and ended up with "Cloud Searcher"... That doesn't seem to be "taken" by a company/product... plus it has a nice dreamy sound to it, kinda? =)

**** 2006-02-13:

Last night I contacted Luftfartsverket (The swedish airspace authority) by email, with an inquiry about which rules apply to my construction, and what permissions I need to apply for.
I figured it's a good idea to do this before I start buying hardware, as I might otherwise get something I'm not allowed to use.

**** 2006-03-10:

This is weird. I've been waiting almost a month to hear back from Luftfartsverket, but I haven't heard _anything_ from them.
I read up on Swedish airspace authorities today, and apparently Luftfartsverket was just recently split up into Luftfartsverket (airspace authority) and Luftfartsstyrelsen (Airspace board). The first now handling only issues concerning airport airspace, and the second handling civil aviation and airspace in general.
So I sent another email today, this time with Luftfartsstyrelsen, with Luftfartsverket as BCC.

**** 2006-03-20:

I have done some more thinking about how I will get data and images back from the balloon.
In my previously discussions with 'jcoxon' (the guy who builds the "Pegasus" series of balloons), he suggested that I use APRS for position reporting.
However, although the APRS coverage seems to be quite good, it seems very limited in sending additional data (ie. images) back, as relatively small amounts of data can clog up the network.
And Iridium and similar satellite phone/data systems cost from above €600 (that's something well over $600USD) for just basic hardware, normally with a very high data transfer cost.

So I'm leaning towards using a simple tri-band GSM phone instead. That would however require the balloon to stay at about 2.000 meters (~6.000 feet), but it might still be a good idea as it will make the temperature and pressure requirements on the hardware a lot easier. The major drawbacks would be
1. The fact that the balloon might float into a mountain somewhere, and
2. The fact that the pictures won't be as pretty. =)

I am however pretty excited about the possibility of having a "live webcam" in the balloon, which transfers back images to a website in realtime.
This might be possible as the "normal" data rate of GSM (9600 baud) is enough to transfer small images every few seconds. And the cost of calls isn't all that high within Sweden.
Also, Vodafone had an offering a while back of cellphone calls abroad with the same "minute price" as calls within Sweden. Although the opening cost for each of those calls is about €2.

The webpage for jcoxon's balloon projects is located at: http://www.srcf.ucam.org/~jac208/pegasus/
For additional information about APRS, look at http://www.aprs.net/
For additional information about different satellite phone/data networks, see

**** 2006-03-25:

Still no replies from either airspace authority. I double-checked the email addresses, and they appear to be correct and official.

So now I'm sending yet another email...
This time, if they still haven't replied in a week or two, I'm going to start harassing them by telephone aswell...

**** 2006-04-03:

It doesn't come as much of a surprise that neither airspace authority has replied. After almost 2 months!
And email is in fact an official means of contacting these authorities about airspace and aircraft issues, which is exactly what I'm trying to do!

Anyway, I called Luftfartsstyrelsen today and ended up talking to a very helpful lady who immediately started investigating what had been happening with my emails.
Later a man from luftfartsstyrelsen finally emailed me back with exactly the information I needed concerning laws, permissions, etc.
The applying rules may be found (in swedish!) in "BCL-T, section D" at http://www.lfs.luftfartsstyrelsen.se/
Click on "Gällande författningar" => "BCL" => "BCL-T" => "Bilaga D - Obemannade friballonger".

**** 2006-04-04:

This morning the helpful lady called me back again with additional information about my inquiry!
Apparently I may be able to get help with technical expertice aswell!

**** 2006-04-09:

I've been trying to figure out what onboard computer I should get, etc. If I'm going to stay at 1,000 (2,000?) meters altitude, a gumstix should be enough as the temperature conditions aren't that hard.
But still... I need it to be able to run a few servos, and there are alot of different solutions out there...

I'm becoming somewhat uncomfortable with the thought of attempting such a low altitude though, as air pressure naturally varies somewhat, as there are both upwards and downwards winds, and then there's turbulence.
The difference between 0 MSL (sea level) and 1,000 meters MSL is only about 10 %... Not a lot of margin for errors.

Another thing I've been spending some time thinking about is using pressure differences to generate speed, to stear a zeppeline-like balloon into the direction you want.
That is, when the aircraft is heavier than the air, it will sail forward like a sailplane while moving downwards. And when it's lighter than the air, it will sail forward while moving upwards!
The change in lift could be generated by moving some of the helium between the balloon and a compressed container.
The biggest problem though, seems to be that all lighter compressor only work at ~10-15 Bars. And if you want a lift difference of 0.5 kilos between minimum and maximum lift, you'd still need a ~50 liter container for the compressed air.
Kind of hard to do if you want to keep the complete payload/systems capsule under 1 Kilo...
Also, I've had another re-think of the possibility of having electric engines change the altitude. But as you'll always be trying to move a very large balloon, the friction against the air would mean large amounts of energy would be needed to make even small changes in altitude.
But I might be wrong?

Anyways, hopefully I'll be able to decide on an onboard computer within the next few days!