iris+ suddenly started to fall from 80m high and crashed

Hi 

Yesterday I was flying normally and switching between ALT HOLD and LOITER mode back and forth during the mission. 

At the last moment I was at 80m high and was trying to come back, so I switched to ALT HOLD and cut the throttle all the way down. 

It started to descend normally but some seconds later suddenly it just started to fall, at that moment I looked at the quad, it looked like one of the motors was not working as one the sides was not level as the others. I went all the way down without spinning whats so ever.

I thought maybe one of the propellers went off, but the four of them were still there when I get to the crashing place. (see picture)

I'm attaching the logs, could anyone help me solve this mystery?

Actually the same thing happen to me 6 months earlier, by the time I thought it might have been a propeller, but now I know it must be something else.

Views: 487

Attachments:

Replies to This Discussion

You have to remove the sticker! It's bad luck...

It looks like the throttle in dropped to zero and the copter started to drop, and at 70m detected a crash. But the vibrations seem to go very high just after the decent started. Something came loose?

Thank you Bill for your help. 

So I've read about vibrations in http://copter.ardupilot.com/wiki/common-diagnosing-problems-using-l...

Is there anything else I can read to understand more?

Can we assume that the 70m crash heigh is because the baro was already giving bad input due to vibrations?

I really can't tell if anything came loose, the four propellers were in it's place when I found the copter, as you can see in the picture.

To plot the graphics, are you using APM Planner on Mac o Mission Planner on windows?

I'm using APM Planner 2.0 you can download the latest version for OSX here (it's stable) from here (it has faster log loading than previous versions)

apm_planner_2.0.19-rc4-59-gbbaa3f7_osx.dmg

The altitude reading is correct, the vibrations effect the INAV and crash detector. It looks like something came loose and the vibrations went high. It's hard to know for sure.

You can see in the alt vs GPS alt they correlate

Ron check the motors. I've had times where the bolts that hold the motors to the arms start to come loose and the motor can actually wiggle around inside of that cup that hold them.  On mine, the motor was loose enough that the outer bell of the motor was rubbing against the wall of the arm and was causing the motor output signal to saturate on that one motor (which is what made me look at it in the first place after seeing that in the logs).  It's good practice to check things like that before every flight (lesson learned).  Open the shell up as well to make sure that nothing came loose inside and that no foreign material found it's way into the shell.  I had a buddy whose Iris was flying erratically.  I looked at the logs and there was terrible vibration.  I opened it up and a small rock had gotten into the shell and lodged itself under the Pixhawk.  And one last tip, always have Stabilize mode setup as one of your flight modes.  When things go wrong, sometimes Stabilize mode is your best chance at getting control back.  When in an altitude controlled flight mode (AltHold, Loiter, PosHold), bad vibrations in the z axis can cause it to either climb at full throttle or drop at min throttle (depending on whether it samples the Accels at the peak or valley of the vibrations).  When that happens, stabilize mode can give you control back.  Not saying that's what happened here, but just a tip for future reference.

Bill, Dax

thank you so much for taking the time to answer! 

will take note of all the advices. 

I did found a screw falling of from the inside, but can't tell if it was loose before or after crash.

Do you think the lack of some of the vibration rubber on the gimbal could contribute to vibration all on the quad? (I only had two of the four installed) you can see that in the picture.

RSS

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service