This is a project that I have been working on (mentally) for quite some time now. It is great that I am actually making progress after several years of thought organizing. Basically I want to make a unmanned aerial vehicle (UAV) that will be attached to a weather balloon and at a certain altitude, cut itself free and navigate back to the launch site. I don’t think I can simplify that any more. The glider (which is what it really is since it has no way to power its flight) will use an on board computer to measure different instrument readings in order to find its location in real space. Using that data it will control its flight path to (hopefully) land where it was launched from, unless otherwise told so. My main goal for this project is to get high resolution (~8mp) pictures of the earth and horizon from ~ 100,000ft. On board Components (First Draft)
These are the systems that will control the glider as of now. The list shouldn’t change unless I come up with a better idea later on into the build. The current components that I already have are:
I am using Arduino micro controllers to basically run the glider. Since there are so many components and calculations that have to occur in real time, I plan on having two for redundancy and processing power. To prototype I am using the Arduino Diecimila because of its easy usb interface. For actual glider use, two Arduino Nano’s will be used because of their tiny footprint and weight. The I/O pins are the same between the Diecimila and the Nano which will make the swap over very easy. Cost (Diecimila) $35.00
The camera was purchased off of Ebay with a broken LCD screen… I am stripping all of the cover and excess weight off of the components anyway so this makes no difference. The less the glider weighs the better. Cost $20.00 Not any ordinary GPS unit can be used for these high altitude trips. Laws are in place that restrict gps manufacturers from making units that will work over 60,000ft or 1,000mph. I was however able to find the Garmin 35 Series, that were made before the laws became active. It uses a serial connection to talk with the Arduino. Cost $25.00
I actually had a friend that I traded some component for two super sub-micro 6g servo motors. Cost $0.00
The temperature sensors I decided to use are the 1 wire type and are free samples from Maxim-ic. Cost $0.00
One of the things that took the longest for me to decide was what type/style glider do I want to make? Having no prior aeronautics or RC Glider/Plane experience, I looked at all possibilities. Looking through articles and talking to people on forums is where I gained most my knowledge. I had to figure out if I wanted a glider that A: Stays in the air the longest. or B: Glides the furthest. I decided to go with B since I may eventually have the glider land at a far away location. It would be very interesting to have it fly X number of miles and to take pictures of its journey. The biggest issues that I have to think about when deciding which glider type to use are weight and strength. I need to make this entire project as light as possible as to meet FAA regulations for amateur ballooning. Looking at the results from Mark I, I found that great speeds and G forces will occur during initial cut down and pullout. Here are the designs I have been considering and why I am considering them.
Glider Types Part II:
(These images were taken from Google and I am in no way affiliated with them or their takers)
I found this military glider and really liked the wing design. It looks very stable and likely to give a very good glide ratio. Although that was before I knew what happens when you take a glider over its maximum speed.
That video is even more appropriate for this design. I found this one before I saw that video also. The things I liked about this design is that it has room up front for the electronic components and could be controlled by 1 vertical servo and 2 horizontal aileron servos. Which is better for redundancy.
I found this RC Glider and at first didn’t like the V tail. However, by using the V tail I can eliminate an entire servo and only use 2 for controlling the glider, this is still redundant if one decides to fail. The streamlined fuselage would have less wind resistance and the large wingspan would offer a better glide slope. The only downside is that I don’t think there is enough room for some of my larger components.
I must have spent at least an hour looking this single photo over. It is a great collection of RC Gliders and looks tremendously fun. I never expected to find them as large as shown here. I would love to build one as large as the tallest ones you see but don’t know if I can keep the weight down and strength up.
Finally I found the one I am gong to use. It can be made strong enough, light enough, and still have room for all components. Unless some other better design pops out of the wood work, this is the one. I doubt that any glider kit out there is strong enough for my needs so it is inevitable that I will have to make it from scratch. Which suits me just fine!
So now onto the components to go into the glider. Here is a small list of the main ones.
One of the main things I want to get back from this project is high resolution pictures of the earth from extreme heights. I was planning on using an old 4mp digital camera to do this but an ebay search and 25 dollars later I landed myself a nice Cannon 8mp digital camera complete with broken lcd.
Most gps manufactures have restrictions that are in place to keep people from using the receivers for guided missiles and such. They fail to work above 60,000 ft or over 1,000 mph. Now some manufactures take this request as an OR statement. Either shutoff above 60,000ft OR over 1,000mph not both. I was able to find one that was made before these regulations took effect. It is widely used by amateur balloonist. The Garmin 35 series.
So after a quick ebay search I found one for 20 bucks.. sweet. After stripping off the weatherproof outer shell here is what you’ve got.
The gps ended up having rs232 outputs and the Arduino accepts TTL level serial.. So a max232 circuit later and we are all set. The camera is extremely light after removing the cover and batteries and there was also quite a bit of change on the gps once its protective shell was gone.
So now there is only one main component left to find. Communications
Now the glider link above used a packet modem and the Payload link above used HAM radio.. Since I don’t have any ham radio equipment let alone a license, I think I’ll stick with a packet modem.
Oh yeah. Almost forgot that I am broke and in college (typical). Off to find a cheap way out. voila, Meet the Motorola C168i. This 9 dollar phone with 25 dollar airtime will give me 3 months or ~150 text/SMS messages of use. Also it can be refilled whenever I decide to launch again. One downside, GSM cell coverage typically cuts out ~10,000 ft. but then again it is going to have to come down eventually.. It absolutely amazed me how easy it was to interface. The headphone jack is also a TTL serial port.
And next to the other components with a 3″ cardboard tube in the background (probably going to be the (mold for the) fuselage.
Ok so now the GPS is ready to integrate and so is the Cell phone. I put off working on the camera any more because I didn’t know what all I wanted it to do. I finally came up with it taking pictures every ~20 seconds and videos at various moments (launch, cutaway, landing). I used some bent wire to make sure I had the correct traces located and proceeded to remove the plastic cover and dome switches.
Next I securely attached and soldered the wires to the camera.
Here is a picture of the really bare bones board (RBBB) Arduino Kit from Modern Device. They are much smaller than the Diecimila.
I used female headers on top instead of male headers on bottom. this will allow me to directly connect them into a breakout board im making from scratch.
Three of these will be in the glider and here are their individual functions.
And here is the flight computer and other major components awaiting code testing. I just need to finish wiring it up and am still waiting for the third flight computer (freeduino). Once that has arrived I’ll start testing some code out to make sure that I can get every system working individually before I put it all together.