I am thinking sending APM power glider to near space by balloon and return home. Anyone already made it?
The video below link below is interesting because the RC Explorer guys talk about how hard this is and that it took 4 tries to get satisfactory results. Well, it is hard.
Here's a link to the post flight video from North Texas Near Space 3 and the post flight report. The goals of the report were to report general information on results + root cause and corrective actions for any anomalies. Hardware for the next mission is on the bench.
These guys from BC, Canada did it, long before the RCexplorer guy. They may have been the first, even. But here is their website, they seem to have documented their adventures quite well.
We are doing the same project here at our university, and are hoping to research some novel designs in this scheme. Looking forward to working with all of you!
I had lost that link, thanks for posting. It was interesting to review. This was posted elsewhere, but thought you might like to see APM at altitude. Minor SW issues have been corrected for a re-try fairly soon.
Thanks for sharing that Larry, and congratulations, I'm going to be frank and say that you will be heavily cited in our research. But I was hoping you could help me out here:
I'm interested in your release mechanism, I've seen it work, but don't quite understand what happened there. We used nichrome wire that was heated by means of nickel cadmium batteries that would activate at a particular altitude detected by an aruduino. The nichrome would melt the nylon rope that was holding up our plane. Unfortunately due to a standoff with our airspace regulators, we couldn't launch our craft from near space. We demonstrated our release mechanism on the ground, and it worked in -40*C ground temperatures.
On that note, at the point of release, how did the APM kick in? You mentioned that you had some sample code that would detect freefall. We toyed around with the idea of using a timer based on calculations, but if we were off and the APM kicked in before burst, that would be a problem!
I'm reviewing your post-hoc reports. I don't quite understand a lot of the technical stuff, but I am willing to learn. It seems to me that the forces and speeds your stinger had to deal with are quite extreme, so structural integrity must be a key issue. I'm starting to doubt that our EPO foam plane could handle it (fingwing penguin airframe). I was toying around the idea of using a bio-organic composite material for an airframe for research purposes, but would need to anticipate the active forces on it so see if it is feasible for high altitude flight.
My release was servo activated and the servo was inside the insulated airframe (as was my elevator servo). The servo was controlled with APM based on altitude, longitude, freefall and time constraints. Some of that is probably not necessary. Altitude and free fall are necessary. I have used two different release mechanisms. On NTNS-2 I suspended the plane by the tail and used a sailplane tow release. This is readily available. NTNS-3/NTNS-4/NTNS-5 suspended the plane wings level from the CG. To make this fit I made a custom release from light ply, and two carbon fiber tubes with a cable running between the tubes. The cable is pulled by a servo to release the plane. Works fine. I set a flag when the vehicle drops and use the flag in a bunch of places.
I run the autopilot all the time, but disable servo outputs and I don't accumulate integration states in the Chalenger controllers. I'm controlling airspeed only. So, there's really no issue with delaying the autopilot since it wants to pitch down to gain airspeed. If you watch the video you'll see that a controllable airspeed happens pretty early and the plane does a dandy job of passively recovering to nose down. There's no reason the plane couldn't have recovered much earlier if I had not botched autopilot code intended to handle the Euler angle singularity at theta=-90. i.e. the stinger's brush with 500 mph greatness wasn't necessary. The max load factor was a little over 6 G's as the plane was in a barrel roll. Hopefully I've fixed this. It is interesting that the Stinger held together. The wing was stock (structurally) but the horizontals were reinforced with carbon spars as was the vertical.