Success!

After an early start, and about an hour or more at glider assembly and checklist stage due to missing a step in the checklist, we had a good launch.

The target altitude was decided to be 10,000 to 15,000 feet, depending on the wind as it went up, and our ability to see the glider in the video camera on high zoom.  A primary goal this time was to learn in detail what happened when it released from the balloon, and video seemed a good way to ensure that.

The winds had came up to be 10 knots or so as we assembled the glider, which didn't help with nerves.  But the launch went great despite some minor delays and wind, however, and the flight was just fantastic, as explained below.

(Photos and video by Grant, Tanya, and Darya)

A pensive look from Simon.  Getting the nose cone on with the O-ring seated, is a pain in the ***.  (No jokes about O-rings, please)

The balloon handlers (Owen, Aaron, and Steph) always look like they're having way more fun than those of us at the glider assembly table.

One of the touchiest moments is letting the line out gradually, in wind, before the release.

The balloon handlers also walk quickly with the wind on release, to try to reduce swinging.  It globs around like a giant soap-bubble in any sort of wind.

Up, Up and Away! 

Video Clip of the Launch (1.9mb, 21s)

The winds were fairly strong after launch, but they changed direction 180 degrees with altitude, so the balloon went up in a spiral that strayed only a few kilometres from the launch site.

We used a video camera with a (claimed) 76x zoom, on a tripod, to try and get the cut-away from the balloon on video.  We knew the camera used "digital zoom" for part of its zoom capability, which means Big Pixels (i.e., "false advertising zoom"), but it was thought the glider would still show large enough in the video to avoid that.  Sadly, no.  

The original plan was to go to 15,000 feet, but at 11,000 feet it was getting pretty small and blurry, so we told the glider to release right there.  The video shows a tiny cross-shaped thing drop and instantly arc down and away from a round blurry thing - makes for riveting viewing.  Nonetheless, we now have a release on video. 

Video Clip of the Balloon Release  (1.4mb, 13s)

The autopilot recovered immediately into a nice glide at just about exactly the expected airspeed, trimmed itself out straight in seconds, and turned back to the launch site.  All flight instruments and systems seemed to work perfectly.  Even the onboard wind estimation algorithm worked - which is stunning, as it requires a low average error margin on almost every instrument at once.

Soon after the release, when it was obviously flying itself properly, the autopilot was commanded to fly a "panorama turn" - a 360 turn with photos taken every 45 degrees, which worked great.  The "Expert" high-level autopilot then guided the glider back down to the landing site without help from the ground station, although the autopilot turn-gains were experimented with a bit (to no noticeable effect). 

The approach pattern was not bad, but could stand improvement - at times it was a bit more of a figure-8 than the ordinary race-track shape, it seemed due to the strong headwinds.  Overall the landing behavior worked as well as could possibly be expected though, with a good approach, and a touchdown less than 65m away from the launch point.  This is well within the expected landing error margin, given GPS error, wind changes from altitude to ground level, and altitude estimate error.  In fact, it's better than expected.

The aerial photos are also just fantastic:

35mm Aerial Photos

Over-flying the launch site on one of the last patterns, at maybe 2000 feet AGL.  The descent became a game of spot-the-glider, as it planned and flew its own approach path, leaving us little to do but watch.  The winds up there were fairly strong, 25 knots plus, so it would approach the launch site slowly, and then appear to really boot it on the downwind turn.

 

 
It flew right overhead, went into the wind the right distance, and hit the chute at about 550 feet (170m) above ground level. 

 

Note the point that  the orange shroud lines spread from - they knotted themselves a bit in the chute box.  This had never happened before, but something will need to be done to make sure it doesn't happen for next time. 

Otherwise, a beautiful staged chute opening, and smooth descent.

It landed gently, only 65m from the launch point.  Not bad at all.

Two years of work to get to this point!