Servers by jDrones

Maiden Flight Loiter Failure

Hello all! I have been following here for a while over my last few Arduplane/copter/quadplane builds and y’all have been very helpful so thank you! However, I am now stumped on my most recent build which has lead to the plane being completely destroyed. It was the maiden flight and the plane performed beautifully in manual and stabilized flight modes for about 3-4 minutes flight time. I then attempted loiter, which (assuming GPS had good lock which I did at 13/14 Sat) shouldn’t be too difficult. However when I switched to loiter the plane then took off and accelerated very fast, and by the time I realized it was going at ridiculous speeds it was too late and it ploughed into the ground at almost 35m/s :frowning: I have never had this happen before and the only clue I had was in the Log CTUN ThrDem skyrocketed for no reason up to 170%. The plane seem to have compensated the correct direction by the looks of the servo directions, however, at those speeds the controls would have flexed/locked up. I have never experienced this with something as simple as loiter before, any help would be appreciated! I have attached the link to the Bin Log file and parameter list to this post. Thanks all!

https://drive.google.com/file/d/1GnrzJJgnse9ioGV2G5WQi1UVSDe44c2P/view?usp=sharing
https://drive.google.com/file/d/1RxHUhT0deQgXG-lrHO0j9OHp8eNn3zp0/view?usp=sharing

1 Like

Something similar happened to me, but unfortunately I wasn’t recording the LOG! In my case, I had already flown this plane 4 times without any problems! In this specific flight, the only thing I changed in the parameters was to enable ARSPD_AUTOCAL (keeping ARSPD_USE at 0) … When the plane entered the RTL loitering step (I was going to leave it in this condition for 5 minutes so the airspeed sensor would be calibrated), the plane simply engaged a full throttle towards the ground!

DVR FOOTAGE: https://youtu.be/q2GZKhLeIEQ

That sucks.

Switching ARSPD_AUTOCAL shouldn’t have caused that. I’ve flown for ages with that left on (ARSPD_USE was still 0).

What kind of FC are you using?

OH NO! I’m glad I am not the only one. I am very confused as to what has happened :frowning:

Hi Allister!
I am using a Matek F765-Wing. I agree I made sure that arspd_use was 0 prior to flying since it was not calibrated. It was a relatively windless day so GPS speed should have been suffice. Moreover that in the log GPS and arspd was both pretty close in blues by the time it was put into loiter… Still stumped :frowning:

That’s what I thought, too, Allister! But it really was the only parameter changed on this last flight… On the other flights, I had already done the autotune and a flight of 30 minutes after the autotune… The calibration of the airspeed sensor would be the last adjustment that would be made so that I could make bigger flights, but unfortunately it happened… I was using a Matek F405-Wing, running Arduplane 4.0.5 (stable)!

That’s unfortunate that it was never solved. Wonder if it’s some rare bug that is in ardu somewhere

@Gary_Wang, I’m still learning to read and interpret the log files but here’s my take on what happened:

Your flaps deployed when you went into loiter mode because the FLAP_1_SPEED = 50, or 50m/s. Since you were below that speed the flaps, and flaperons both went to their 30% position. The flight controller continued to request an increasing roll rate, to the 45 degree max set, but only reached about 9 degrees before it began to roll in the opposite direction.

Looking back at the log the roll appears to follow along with the flaperons pretty well. However after the flaps deployed the roll seems to just go off on it’s own. I wonder if one of the flaperons or flaps got stuck or failed to deploy. This would have triggered the roll that the plane could not correct for.

As for the throttle, the FC is trying to use throttle to maintain altitude in the turn. You can see at one point the descent started to slow, so it backed off the throttle for a moment, but as soon as the descent increased again, so did the throttle. The initial nose down pitching is probably a result of the flaps and flaperons being deployed.

I’m an amateur at reading these logs so I hope somebody else with some more experience has a look to see if I’m even on the right track.

Servers by jDrones