I was flying around and experienced a radio failsafe. RTL was engaged for one second and seemed like all motors were cut off while in RTL. Failsafe went away & stabilize kicked in very quickly, but my full throttle was not enough to recover - thankfully slowed it down enough to have no damage.
I have tested RTL and it performed (previously) without issue. I always manually RTL once before taking longer flights. I checked the throttle position for my radio failsafe - it is at zero. But this should not matter since in RTL throttle is not used (correct?).
I've been going through the logs but it does not make sense to me why would RTL set throttle to 0. Under CTUN, when graphing BarAlt + ThrIn + Thr out, at line 3104 ThrIn/out goes all the way down to 0 (when RTL kicked?). Shortly after on line 3126 you see full throttle, but BarALT continues to dip. Here is a zoom of the area:
Video of so you can hear motors cut out (I was trying new settings on my gimbal which is causing it to twitch in this video):
I checked if maybe my RTL ALT was set to zero, but it is not:
Any help would be greatly appreciated.. I'm afraid to do any real flying until I can figure out what happened. Log attached.
here you go:
I unplug the ppm cable and do not re-plug. Now that I see the video closely, it seems to go to RTL for a split second before disarming.
thank you Joakim! I will try your suggestion.
In the off chance I am not able to get it going the way Joakim suggested- Randy, do you feel it is unsafe to set my FS to toggle RTL and leave 50% throttle so the motors do not cut off? or could this still have unforeseen consequences?
thanks again for all the help, really appreciate you guys taking the time.
Raph, I believe the problem is because of the way that you are using the Rx failsafe. Instead of programming it for the official <950uS failsafe, you have instead programmed it to switch to RTL Mode, with a zero throttle. As you point out, this should be OK, as you can in fact change to RTL, and then drop the throttle to zero and it will not fall.
However, I think your problem has to do with the way the frame packets come in. This is a bit of a guess as I don't actually have enough data to prove this, but I believe that your Rx is changing the throttle to zero *before* it changes Ch5 to RTL mode. You can sort of see this in the logs actually. You'll notice the throttle drops to zero, and the motors shut off on line 11589. However, the mode change to RTL doesn't come through until 11592.
So your testing is not quite the same. If you start in stabilize mode, and then drop the throttle to zero, then change to RTL, I believe you'll see if shuts off the motors exactly the same way.
The solution would be to utilize the official failsafe feature the way it was designed.
1. set radio failsafe to RTL with throttle in mid position. simulated failsafe by turning off my DX8 radio. RTL goes as planned.
2. set radio failsafe to RTL with throttle at zero position. simulated failsafe by turning off my DX8 radio. Copter falls out of sky.
Is this by design? I thought throttle was ignored in RTL until the land sequence, and definitely should not turn off motors completely. Can a developer please chime in here? If this is by design for whatever reason, I really feel there should be a clear mention!
I too had a cut off while descending in RTL:
It was very windy the day it happened to me.
Here is what my data looks like. My RTL was lower than the start altitude, but I was on final- too final!