[Canberrauav] flying this weekend
christopher.d.gough at gmail.com
Sun Nov 4 16:23:46 EST 2012
About removing the broken spar, I've been thinking. Manually twisting hard, I could feel/hear the glue creaking. I think some percussive encouragement on the end (hammer) while twisting would persuade the glue joint to shear. Might want to plug/tape the impact end to prevent damage, in case it doesn't work.
On 04/11/2012, at 3:26 PM, Andrew Tridgell <tridge at samba.org> wrote:
> Hi All,
> We flew the X8 again today at CMAC, launching first with Stans bungee,
> then with the one Chris brought.
> We first flew with one battery, same as at Lake George. The aim was to
> get an idea of endurance, given we ultimately want to fly for over an
> The bungee takeoff went well after we stretched it out enough, and Steve
> helped with the APM tuning from the ground station while we dialed in
> the navigation parameters. The X8 flew for 25 minutes doing a simple
> circuit auto mission before the battery reported 20% remaining and we
> brought it down for an easy landing.
> The next test flight was with 2 batteries, giving us 10Ah of 4S
> LiPo. That requiired a bit more bungee stretch (with the bungee from
> Chris this time), and the takeoff went fine.
> After a minute or so of flying OK it started to lose altitude quickly in
> AUTO mode. I put it into FBWA and found that it was still losing
> altitude! It kept going down and crashed quite hard a couple of hundred
> meters from the runway. Damage includes some dented leading edges, and a
> broken wing spar.
> The wing spar looked to be the main issue, and we had a long discussion
> on how to fix it. Once I got home I found that inserting 3 pieces of
> narrower carbon fiber tube hard up against each other inside the spare
> repaired it nicely, so the X8 is ready for another go.
> I'm still analysing the logs, but thus far it looks as though we had
> very marginal GPS lock (losing lock completely several times), and that
> had a huge impact on the GPS velocity measurements used to correct the
> accelerometers. The result was a huge pitch error, which caused the
> plane to dive hard. If I had switched to manual it would have been OK.
> I'm planning on adding some some code to the AHRS to use both estimator
> methods for acceleration correction in parallel, as a way of detecting
> poor GPS velocity numbers.
> Cheers, Tridge
More information about the Canberrauav