Alta Crash

Discussion in 'ALTA' started by Dylan Glockler, Jun 14, 2016.

  1. Dylan Glockler

    Joined:
    Jun 14, 2013
    Messages:
    112
    Likes Received:
    23
    I was going to wait until FreeFly had an opportunity to review this incident but I submitted the incident and requested an RMA a week ago and was told there because of the Alta 8 grounding they could give me no estimate on when I will receive an RMA to get the unit in for repair.

    The Incident
    We were flying the Alta with the MOVI m5 carrying a Sony a7s ii with a Canon 24-105 L lens and an Amimon Connex video transmitter. Batteries were a pair from FreeFly - 10,000 mah. A pretty light load using our normal, teal dampeners. This is the same configuration we've used countless times.

    We started the day with a compass calibration and normal checklists and systems checks with no issues. We flew the first set of batteries, no problem - had a nice flight for just over 8 minutes with little to no wind. We swapped the batteries, completed our check-lists and start up procedures, again with all systems go as normal.

    The pilot launched the Alta in manual, as normal, hovered for a few, then switched into GPS mode. The aircraft banked instantly, hard at about 20 degrees from the nose in an easterly direction and towards the ground (incidentally NOT holding the altitude as expected). The pilot tried to correct but it was only seconds before it hit the grass. This was an area we had flown numerous times previously with no issues and nothing we expected to cause any interference.

    One boom has significant damage to the buckle with some props damaged. A couple pieces of carbon fiber on the m5 are damaged and will need to be replaced and one landing gear leg snapped. Camera was not damaged, so all-in-all we were pretty lucky. I suspect total damage is probably to about 5-8 basic parts.

    I looked through the data and tried to decipher but nothing jumped out at me (no LOS or loss in GPS, etc). I'll attach the data if any good samaritans are willing to have a look.

    I certainly understand FreeFly is a smaller company and quite busy with this grounding issue they're having but I sure hope some repair resources free up for us old customers somewhat soon.
     

    Attached Files:

  2. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,380
    Likes Received:
    1,164
    Dylan: At first sight I thought this might be caused by a solar flare disrupting the GPS, but I've checked the Kp Index values:
    June 7: https://www.spaceweatherlive.com/en/archive/2016/06/07/kp

    The last significant event was the day before your flight: https://www.spaceweatherlive.com/en/archive/2016/06/07/kp

    The log file appears to be incomplete -- normally there should be a "Flight Summary" at the end. This incompleteness is something that often happens if power is cut to the controller in a crash -- there is not time to flush the last few data rows of the log file out to the micro-SD card.

    I'll study the file some more when I have a moment -- I just got back from a two-week trip so I'm still digging down to find my desk.
    Andy.
     
    Humberto Barchini likes this.
  3. Chris Jackson

    Chris Jackson New Member

    Joined:
    Dec 5, 2015
    Messages:
    17
    Likes Received:
    3
    What do the props look like? Looking at the log vibrations are massively increasing right before impact, and accelerate when the pilot inputs full throttle. Was there an orientation issue as well?, looks like its uncommanded pitch for the most part, and pilot never inputs anything to correct it, but it does roll and yaw to command input right before she hits.
     
  4. Cody Lewis

    Cody Lewis Member

    Joined:
    May 1, 2013
    Messages:
    56
    Likes Received:
    4
    Anyone think it is weird in the alta data viewer that the aircraft is rolling in the opposite direction that it is moving? Is that a bug in the data viewer or something up with the altas IMU? (or I suppose more likely the GPS?)
     
  5. Dylan Glockler

    Joined:
    Jun 14, 2013
    Messages:
    112
    Likes Received:
    23
    Cody, Chris, Andy - thanks for looking it over - working on-site for two weeks, so a little slow in responding.

    We checked all props (as we do each time) before flight and there was no play or discernible issue. A number of them are a broken and shattered after the crash. I believe the pilot first attempted a fairly aggressive throttle up - again, it was flying away from the launch point and did not appear to be holding attitude and also altitude (banking away and down).

    Cody, your note about the roll direction is disturbing....? Sending the thread to the pilot in case he has anything to add.

    Dylan
     
  6. Cody Lewis

    Cody Lewis Member

    Joined:
    May 1, 2013
    Messages:
    56
    Likes Received:
    4
    I think I know why your alta crashed. We did some testing on this and have replicated an issue that sounds exactly like yours. If the alta is spinning at all on boot up (as it would on top of a movi) when switching into GPS mode, it will bank VERY hard in one direction, without sufficient altitude and quickness on the switches I could see how this could lead to a crash. The Alta behaving this way is repeatable each time it is spun on boot up.
     
  7. Dylan Glockler

    Joined:
    Jun 14, 2013
    Messages:
    112
    Likes Received:
    23
    It does sound like the most likely culprit and we were aware of service bulletin SB12 but i'm quite certain that the Alta was not spinning during boot up and we absolutely did check that there were no faults, though as indicated in the SB this can be the case and still cause an issue. Perhaps the data could tell for certain whether there was any movement on initialization.

    I'm not sure why they call SB12 a service bulletin when it's really an advisory - one would think that a service bulletin would issue a recommendation for service to correct the issue (there is a distinct and important difference between an SB and an advisory in aviation).

    I would advise anyone flying to secure the alta from movement using a foam prop wedge as suggested in SB12! I'll be adding it to my checklist.

     
    Rolands Zelders likes this.
  8. Cody Lewis

    Cody Lewis Member

    Joined:
    May 1, 2013
    Messages:
    56
    Likes Received:
    4
    And they only mention "degraded" performance. Which is a total understatement. In every test I've done I've been able to get the Alta to boot while moving without faults. In gps hold its not just poor performance, it banks very hard and flies off in a direction to the point I'm not comfortable leaving it on for more than a second. At low altitude it would definitely cause a crash. Not only that, but with faster yawning on boot up, the Alta's yaw will be unstable even in manual mode. Ideally they would have a better way of detecting these faults on boot, and hopefully that is something that will come in a firmware update.
     
  9. Larkin Carey

    Larkin Carey Member

    Joined:
    Jun 12, 2014
    Messages:
    30
    Likes Received:
    1
  10. Dylan Glockler

    Joined:
    Jun 14, 2013
    Messages:
    112
    Likes Received:
    23
    The update was released a couple weeks after our incident.

    Seems to be the culprit for our crash, though there was no discernible movement or rotation during boot-up. I really wish the note in the service bulletin, about FreeFly test pilots wedging a piece of foam between the Alta and MOVI during startup, had been issued as an Air Worthiness notice rather than an afterthought.


     
  11. Dylan Glockler

    Joined:
    Jun 14, 2013
    Messages:
    112
    Likes Received:
    23
    Following up here - FreeFly checked in our Alta for repair and said that the issue addressed in SB12 caused the crash. I'm not certain if they determined that there was some movement with the aircraft on start-up, we didn't notice any and there were no faults indicated by the software. From what I can tell, it appears there was no more than a couple degrees movement from initialization of the log to the first throttle input.

    We really wish we had employed the tip at the end of SB12: "Often the pilots at Freefly will use one of the foam prop protectors to wedge between the ALTA and the Gimbal to ensure the aircraft does not spin / move on boot up."

    In any case, FreeFly has been very professional and responsive in dealing with the issue. We did suggest that they consider delineating their Service Bulletins into informational Service Bulletins and Airworthiness Directives, as we do in traditional aviation. I would recommend anyone reading future SBs to incorporate any tips like the above as an AD!
     
  12. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,380
    Likes Received:
    1,164
    Good suggestions, Dylan. An AD certainly gets everyone's attention in the manned aviation world.

    Andy

    Forensic Software & sUAV / Drone Analyst : Photographer : Videographer : Pilot (Portland, Oregon, USA): Trees=2, Ground=1, Props=11. :(
    The Ground Is The Limitâ„¢
    ---------- Forensic Drone Analyst : Forensic sUAV Analyst : Forensic Unmanned Aircraft Analyst : Forensic Drone Expert
     
  13. Stefan Jannides

    Stefan Jannides New Member

    Joined:
    Aug 16, 2014
    Messages:
    10
    Likes Received:
    1
    This might seem like a dumb question, but how do you then remove the wedge after start-up without slicing your kneecaps open?
     
  14. Dylan Glockler

    Joined:
    Jun 14, 2013
    Messages:
    112
    Likes Received:
    23
    I haven't been using the foam to hold the aircraft in place as this issue was resolved in recent firmware updates. Instead, the pilot maintains a readiness to switch back to manual mode in the case of erratic behavior.
     
  15. Stefan Jannides

    Stefan Jannides New Member

    Joined:
    Aug 16, 2014
    Messages:
    10
    Likes Received:
    1
    Thanks Dylan, that's good to hear.
     
  16. Dylan Glockler

    Joined:
    Jun 14, 2013
    Messages:
    112
    Likes Received:
    23
    You bet - to be absolutely clear, this was the response I received from FF when I asked about that issue (relating to SB12):

    "From what I've learned the latest firmware is very good at preventing corrupted booting from aircraft movement. That being said, it is always best practice to make sure flight controllers are kept still during boot up. In the rare case where switching to Position or Height mode do produce undesired response, an immediate switch back to manual to correct the problem is recommended."
     
    Stefan Jannides likes this.

Share This Page