Unexplained crash and log help

Log Analysis -- deprecated Copter Logs -- deprecated
1

I have a hex and I’m using the turnigy 9x TX and RX

For some reason, fail safe does not work on my hex. When I turn off the receiver, the values stay the same. I dont know what is wrong. I am using a 2 position and 3 position switch to get 6 flight modes so I dont know if that has anything to do with it. I updated firmware too so that shouldnt be it.

Since fail safe does not work, today I tried using a nito truck fail safe. turnigy mini failsafe from hobbyking.

When it senses a signal loss, it outputs a predetermined value (I set it at 960)
I bench tested and it worked great.
Also, if it senses 4.2 V or less on receiver it will output 960. (there is no way to turn this off on the turnigy mini failsafe)

THEN!!! I did my first flight with the new failsafe. Everything was normal until I flew pretty fast and stopped quickly. my hex shot up then back down into the ground, then flipped. Nothing is physically broke but I havnt flown sense.

A few thoughts. I think the receiver voltage may have dropped bellow 4.2V when I stopped suddenly because I pulled a lot of current on the motors. Sense the mini failsafe sensed 4.2V or less it triggered RTL.

Question: when failsafe RTL is activated, how do I regain control? Do i have to switch to another mode then back to stabilize?
According to the logs, chanel 3 only outputted 960 for a second or two then it went back to 1040 or so but the hex remained in RTL and never went back into stabilize.

I thought once signal was regained I would be able to regain control??

I am not good at reading the logs and I am trying to figure out exactly what happened.

The crash is at the very end (obviously) but the beginning is me setting up and calibrating new failsafe. I should have disconnected and reconnected before my flight to avoid such a long log.

the sensor health and a few other things seem to happen at the same time as RTL but I am not sure what all the values mean.

2

I also use a Turnigy 9X TX, but I am using a FrySky RX. I was able to successfully execute a throttle failsafe, but my quad had a quick “shudder in the air” before coming under control. Also note that I configured the failsafe on my FrkSky to output 0 signal (as opposed to specific PWM values) when it looses signal. Here is what I believe was going on in my case.

I am using PPM-SUM to connect my FrySky receiver to my APM. I believe that two things are happening in this case, although I would have someone more in the know than I confirm this:

  1. FrySky is outputting last known PWMs for some amount of time when the signal is initially lost. This would explain the delay I have observed between when I turn the transmitter off, and the “shudder”

  2. The wiki documentation indicates that in PPM-SUM mode, there is a 250ms delay, followed by a switch to mode 4, along with a 900 PWM setting on Channel 3.

So here is the sequence of events I suspect is actually occurring:

  1. Transmitter turned off
  2. FrySky looses signal and continues to transmit last known good for some short amount of time (perhaps half a second)
  3. FrySky executes my failsafe settings, which is to output 0 PPM Sum
  4. APM PPM Sum sends last known good settings for 250ms, then failsafe interrupt occurs
  5. APM PPM Sum begins sending 1500 PWM, 900 PWM to channel 3, and sets Mode 4 (Stabilize of me)
  6. 900 PWM is insuficient throttle, and my quad shudders
  7. APM detects 900 PWM on channel 3, and intiative my throttle failsafe, which is RTL
  8. Quad stabilizes and begins to RTL as desired

Here is the documentation from the Wiki on APM PPM Sum behavior in failsafe

Operation of Failsafe using a PPM Sum Receiver
If the APM is connected to a PPM Sum receiver in pass-through mode (APM input pins 2&3 shorted) the APM’s PPM encoder operates transparently.
There is no independent channel analysis and there is no channel manipulation.
If the receiver is disconnected or the PPM signal is cut off, after a 250 ms timeout the PPM encoder watchdog interrupt will start an 8 channel PPM-PWM generator with the following fail-safe values :
CH1, CH2, CH4, CH6, CH7, CH8 = 1500
CH3 = 900
CH5 = 1555 (flight mode 4)
If an input level change is again detected from the PPM receiver the fail-safe output is terminated and normal PPM pass-through operation is restored.
Alternatively, there is a provision on the APM 2.5 board to fully bypass the PPM encoder when using a PPM-Sum receiver…
The PPM signal is connected to APM 2.5 board’s input CH5 (instead of CH1 for pass through).
Also, a pad needs to be modified on the APM 2.5 board and soldering is required.

Finally, one thing I intend to do in the future is configure my FrkSky to output known values (including a 900 PWM on channel 3) for its failsafe. I believe this will mean that the PPM-Sum failsafe will not get triggered, and my quad will go directly into RTL. The reason I didn’t do so initially is that I have to go through two steps of configuration on my Turnigy. Set the low PWM for channel 3 to 900, set the failsafe settings on the FrkSky, then reset the low PWM on my Turnigy back to its default setting.