Autotuning Crash / First Flight Crash

[pUiE (jakob)]

Hey…any suggestions for a Tutorial how to connect ESCs via arduino. Tried this one (Link), but i got a timeout error (stk500v2)

PS: I ordered hobbywing pentium ESCs…just in case

[fnoop]

@pUiE

okay. I will give my BLHeli ESCs another try. Would you recommend to not connect the 5V Cable from ESC to Controller, like denoted in some posts?

Why have you got the ESC BEC connected to the flight controller? I believe you should almost never do that.

@pUiE

Here is my log file with the hard crash -> Hard crash log

Here is my log file with extreme tilting, when autotuning -> Extreme tilting during autotune

The good news is there doesn’t seem to be any indication of ESC desync or propulsion failure from these logs. It does look like it all goes a bit squiffy immediately after autotune starts. Does the copter fly in stabilize OK, with sharp attitude changes? You need a reasonable manual tune before you attempt autotune. It looks like the autotune commands a sudden movement and the controller can’t cope with this, possibly due to poor initial tuning.

[pUiE (jakob)]

@fnoop

@pUiE

okay. I will give my BLHeli ESCs another try. Would you recommend to not connect the 5V Cable from ESC to Controller, like denoted in some posts?

Why have you got the ESC BEC connected to the flight controller? I believe you should almost never do that.

On the emlid documents for navio2 i’ve read about power redundancy (controller gets power, even if the primary power source stops working), this sounds resonable for me. Also the ESCs have 3 wires, so I’m carefully cutting a wire, when i don’T exatly know, what i’m doing.
To be honest, I don’t really get it, why I should NOT connect these extra cables :D…Maybe you can sort things out for me…Thank you.

@fnoop

@pUiE

Here is my log file with the hard crash -> Hard crash log

Here is my log file with extreme tilting, when autotuning -> Extreme tilting during autotune

The good news is there doesn’t seem to be any indication of ESC desync or propulsion failure from these logs. It does look like it all goes a bit squiffy immediately after autotune starts. Does the copter fly in stabilize OK, with sharp attitude changes? You need a reasonable manual tune before you attempt autotune. It looks like the autotune commands a sudden movement and the controller can’t cope with this, possibly due to poor initial tuning.

Okay, thank you for taking a look at my logs. Maybe you can say, where to look for sync issues in the logs.

At my very first try to lift of, my copter was totally not flyable. It wobbled extremly and it was not possible to climb a few centimeters. So I changed the PIDs and after that I was able to fly for several minutes. I even climed around 15 meters or something like that without any Problem.
But it was not that sable as I expected (Saw some cool videos). So I thought autotune would be the way to got.

Would it help to lower the aggressiveness of the autotune as it is described here?

Thanks for your help. I’m always willing to learn

[fnoop]

@pUiE

On the emlid documents for navio2 i’ve read about power redundancy (controller gets power, even if the primary power source stops working), this sounds resonable for me. Also the ESCs have 3 wires, so I’m carefully cutting a wire, when i don’T exatly know, what i’m doing.

To be honest, I don’t really get it, why I should NOT connect these extra cables :D…Maybe you can sort things out for me…Thank you.

Sorry I was wrong! I mis-remembered that you shouldn’t connect opto ESC power wires:
http://ardupilot.org/copter/docs/connect-escs-and-motors.html?highlight=esc
If you are connecting the 5v power from the ESCs then you should also add a zener diode:
http://ardupilot.org/copter/docs/common-powering-the-pixhawk.html?highlight=esc#powering-backup-off-the-servo-rail
Yes it’s a good thing to add an extra power source to your flight controller if it supports it.

@pUiE

Okay, thank you for taking a look at my logs. Maybe you can say, where to look for sync issues in the logs.

Have a look at this thread for a clear indication of a desync:

Pixhawk 2 RCOUT C3 FLAT-LINED CAUSED CRASH

Pixhawk 2 RCOUT C3 went to zero and flat-lined causing my quad to crash. Anybody else see anything like this? Newbie here. Log file too large to be uploaded, but can be downloaded from my Google drive here: https://drive.google.com/open?id=0B0OMLBwWNbKjal9NNjJPU1l5eW81 Trying to confirm this is what actually happened through testing - hardware? Software? Any inputs would be appreciated.

@pUiE

At my very first try to lift of, my copter was totally not flyable. It wobbled extremly and it was not possible to climb a few centimeters. So I changed the PIDs and after that I was able to fly for several minutes. I even climed around 15 meters or something like that without any Problem.

But it was not that sable as I expected (Saw some cool videos). So I thought autotune would be the way to got.

Would it help to lower the aggressiveness of the autotune as it is described here?

Sure you can try to reduce the aggressiveness to start with, but you’re much better off getting a half-decent manual tune first. Don’t try autotune before you can fly in stabilize with sharp movements, without losing control. Here’s a good way of doing it:
http://ardupilot.org/copter/docs/ac_rollpitchtuning.html

[OldGazer]

[pUiE (jakob)]

So.
I’ve did a Second attempt. New frame (s500) and new controllers (hobbywing x-rotor).
Everything fine Now. Copter flies pretty stable with Standard parameters. I think some fine Tuning is still necessary, but for now it looks great.
I don’t know the exact reasons for my crashs earlier. Maybe a mix of different Things:D