A Mishap at Cowra
I attended the Cowra round of the Millennium Cup on the weekend, and didn’t do too badly in the end. The conditions were interesting - weak lift, lots of broad sink including various spots around the field with guaranteed sink, and a fairly defined shear layer with winds aloft. The first day of the event saw me crack 2000 points from 8 flights with one drop. The leader of the day was around 2400 points. The second day was much more successful, and I finished in second place after some good flights. Overall, I came about 8th for the weekend.
However, the highlight for me was maidening my new Pulsar 3200 electric glider. I had my reservations about it whilst building it, but flying it was great! It was a bit squirrely going vertical under power, but going up at 45 degrees or so, it tracked straight, and was easy to fly. Power off, and it’d float around the sky, thermal quite nicely (though it needs to fly at speed to avoid tip-stalling), and rolled quite axially.
Unfortunately, I ignored a warning sign, and watched as it shredded itself in a tree.
I built the Pulsar 3200 to use a five cell flight pack, and a 2000mAh LiPO pack. The flight pack is connected to a toggle switch mounted on the side of the fuselage below the wing seating, which is then wired into a Schulze Alpha 8.35 receiver. The 2000mAh LiPO pack connects to the 110amp ESC. Because I’m not using the BEC feature of the ESC, there’s no way to power the receiver from the LiPO pack, only the flight pack.
I did this deliberately, as I’d kitted the glider with digital servos, which use more power than the typical ESC could provide.
Before I left home for an informal flying day at the Cowra field on Friday, I’d charged all the battery packs. The Friday’s flying was non-eventful, with the maiden flights used for trimming, and fine tuning of the various settings. We measured the current draw of the system at wide open throttle, at around 65 amps, so the glider could go up a prop size or two. I was averaging around 10 second motor runs for good height, and so with 6 motor runs, that’s about 1 minute of motor run, and about 1000mAh of battery top-up. I never recharged the flight pack. Another flight, 1:30 of motor time equated to 1500mAh of recharge.
The next morning (the morning of the competition) I intended to charge the LiPO battery, set up the glider, and fly around for a while, getting used to the conditions, before the competition proper. I pulled the fuse out of the car, pulled off the nose cone, attached the LiPO to the charger, and waited about an hour for it to charge. Once it was charged, I went out to the keyboard, put my key in, walked back to the plane, switched on the transmitter, and heard the control surfaces twitch and move as the receiver locked onto the transmitter signal.
“That should not have happened,” I mused to myself. The plane should have been switched off!
At this point, I should have switched off the plane, and checked the flight battery, as I didn’t know how long the plane had been switched on. I suspect that the toggle switch had been toggled as I’d pulled the plane out of the car, and thus it had sat on the ground switched on for an hour…
I was so itching to fly the glider, I ignored the warning signs, and thus knowingly took the risk of flying the plane with a flight pack that was almost out of charge.
I launched the plane, flew around for a bit, got down low, flicked the motor switch, gained some height, thermalled off way downwind, and was having a great time with it. As I brought it back upwind, and off to the right of the homesteads next to the field, I hit some turbulence. I thought it was a thermal with rather defined edges to it (there were thermals like this during the day), so I flew away from the area for a few minutes, flew back towards it, hit the same “turbulence”, and very quickly lost control of the plane.
It was flying around quite lazily, and quite out of control. I thought I had some control over it initially, flaps down, and using the rudder to slow the wild gyrations in roll. It was at about twice tree height (and I think over the field), when I thought to myself to start the motor. It started, pointed the nose at the sky, I quickly turned the motor off again, and watched as it started loop after loop after loop, with each one getting bigger and faster…
It eventually went vertical at speed into a tree, shredding the wing, stress cracking the fuselage below the wing seat, but left the tail surfaces intact and untouched. Interestingly, the elevator servo had parted itself from the epoxy seat it was in.
As a result, the wing is a writeoff - ribs are missing or shattered, the carbon/kevlar dbox has crushed sections, the right tip panel has been severed in half (right through the dbox and spar), the right half of the flap is destroyed, and the left tip has some deformities in it.
So, it looks like a new Pulsar 3200 is in order, because of the difficulty in getting a new wing. I could repair the fuselage, but acquiring a new wing is gonna be difficult.
Lessons Learnt and Mitigation Strategy
Because of a past life working in the IT Security industry, I tend to weigh up risks and benefits when confronted with problems. What could I do to mitigate this rather distressing event? I’ve thought about this problem for a few days now, and decided that I needed a visual indication of the flight battery strength, and whether the flight battery was “armed” or not. If I’d've known earlier that the plane was “on” whilst waiting for the LiPO battery to charge, I would have switched it off, and charged the flight battery - the glider gave me no indication at all that it was on until I turned the transmitter on.
Paul Daniels from nqrc.com has created a bunch of electronic gizmos for radio control aircraft, two of which would be quite useful to mitigate my problem:
- a 5-cell NiMH/NiCAD low voltage battery alarm; and
- an electronic on/off switch, using a jumper “tag” to disarm the glider.
I plan on installing the battery alarm in the wing, so I can see the LED, and place the jumper tag in a position where it cannot possibly be removed whilst the plane is transported, and provides a visual indicator that the plane is disarmed - remove the jumper, and the plane is armed and ready to fly.
Well, it’s time to save up for another Pulsar 3200.