Attached is the log file and kmz. I did CompassMOT a few weeks ago and it was 4-7%. Vibration looks ok to me. It was a mystery and my first thought was ESC sync issue or motor failure but now I'm wondering if it was due to bad radio calibration causing PWM conflicts.

Before flight, after radio cal in MP; didn't notice lower numbers changed from original setup.

31j0KY0.png

oj6Imsa.png

Could the above have caused a flight mode error and/or GPS glitch or conflict in the Pixhawk?

After crash, recalibrated tx and in MP; numbers are what they were when copter was flying good, no problems, missions worked flawlessly. Over 100 flights. I must have messed with the Taranis and didn't recal it before doing the pre-flight checks and radio cal. I don't know.

qDRJzDS.png

Available flight modes in MP. Note FLTMODE2 value of 16. that was the error in the log.

phu2QIV.png

Here is the flight mode error and also GPS error when the copter was first armed. I thought that was but it locked back so didn't give a thought The copter acted as if it didn't know what mode it was in. It should be noted the flight modes were in super simple mode. Normally I leave stab mode normal.

GrHFz3b.png

I literally lost complete control; copter would not respond to stick moves, then it just took off and made a bee line to the ground. Fortunately so far two props broke and no other visible damage. At no time was there a warning from the Taranis of lost radio contact. I did not have telemetry working on this flight (for lack of a better word).

JIEk5Fm.png

Side note: as I understand it, per Randy COMPASS_ORIENT in parameter settings should be set to 8. It was 0. Correct if that is mistaken. Compass calibration was done with Pixhawk checked off.

http://diydrones.com/xn/detail/705844:Comment:1758426

If anyone can confirm my radio calibration was muffed and could have been the cause of the out of control condition and subsequent crash, or any other reason(s), please advise. I've had so much fun with this Y6 for a good month or two, to have this happen is discouraging.

Thanks.

143.log

143.kmz

You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –

Replies

  • Developer

    DG,

    I've looked at your logs and it looks like a GPS glitch (see last comment) but it's very slow to hit and I'm wondering which GPS you're using.  Other comments and finding below.

    • On the pixhawk, the COMPASS_ORIENT stays at zero even if it's external.  It's only on the APM2 where it's 8.  Your vehicle's compass seems fine.  We can see this by checking the NTUN message's desired velocity vs actual velocity.

    3701814605?profile=original

    • Flight mode 16 is PosHold which is a new mode for AC3.2.  It's not available in AC3.1.5 but due to a limitation in the way we make our parameter descriptions available it appears even for AC3.1.5.  This is on the list to fix.  When it fails to switch into a mode it remains in the previous mode, which is Loiter in this case so it doesn't explain the crash.
    • The GPS error at the beginning is not a problem because I think you were flying initially in stabilize mode.
    • The RC calibration seems ok.  At least the RCXX_MIN and RCXX_MAX values for each channel are not crazy like the ones in the picture you've kindly attached above.  For example RC3_MIN is 982 and RC3_MAX is 2006.  The throttle failsafe is 960 which is lower than the RC3_MIN so a throttle/radio failsafe won't trigger unexpectedly... and you've left the ARMING_CHECK to "1" so it wouldn't let you arm if there was an issue there anyway.
    • The altitude loss is because the throttle was below mid and the vehicle was in loiter mode.  So it was accepting and acting on the throttle input.

    3701814530?profile=original

    I think the speed off in loiter is caused by a GPS position glitch.  The glitch was quite slow in coming on as opposed to a sudden drastic change in position which is why it wasn't caught by the GPS Glitch protection.  Can I ask, what GPS are you using?

    3701814377?profile=originalBy the way, it looks like you switch to stabilize mode just after it hit the ground which is fine.  I guess it happened quickly so you weren't able to get to it as it was speeding off.

    Hope this helps.

    • Thanks so much Randy.

      It is a 3DR Y6B kit I bought in late May. Last week it was acting very strange and I had no control over it, so I contacted 3DR support and they thought it could be a bad GPS/MAG and sent me a new one......just received and installed Saturday, then recalibrated everything.I assume it is the Ublox

      Again, Super Simple mode was used as was the case for dozens of flights because obviously it makes it easier, but also my eyesight isn't what it used to be so going out too far it is difficult to orient the copter. I should have one stabilize mode without SS mode and another with.

      The flight you see here was done with 2 x 6000 mah batteries because I forgot to bring the "dummy" battery (wood blocks) to take up the slack of the long velcro straps supplied in the kit. That's why the throttle hover position was so high. Normally it takes 1/2 throttle or one line below to hover. In this case the extra weight required more power.

      The problem is not simply losing altitude, it was total loss of control and FAST moving to the ground; no response from stick movements.

      Tonight it was even worse. See post at Arducopter:

      http://ardupilot.com/forum/viewtopic.php?f=80&t=9437&p=2232...

      This was with one 6000 mah battery as I normally use and took off normal. Would changing the COMPASS_ORIENT to 0 have wreaked the havoc when I switched to RTL?  According to Otherhand's reply it could. Whatever is wrong, lifting off was straight up, all looked fine until I tried to fly it around. It started to act like I was losing control (thought it was in PosHold......was actually still Stab) so flipped to RTL. Once I did that, it took off like a rocket at 20-30 deg angle toward the ground at 45A. If changing the COMPASS_ORIENT to 0 caused that, shame on me. 

      The flight mode error is answered. All I can say is......doh!

      I think resetting the Pixhawk firmware and starting from scratch may be the way to go at this point. IIRC the easiest way to do that is loading Arduplane then reloading Arducopter. Yes?

      143.log

This reply was deleted.

Activity

Neville Rodrigues liked Neville Rodrigues's profile
Jun 30
Santiago Perez liked Santiago Perez's profile
Jun 21
More…