Welcome to InspirePilots.com

Join the leading DJI Inspire community for free!

Finally Happened On 27th Flight

Discussion in 'Inspire 1 Discussion' started by Lesmess, May 23, 2015.

  1. Lesmess

    Joined:
    Mar 18, 2015
    Messages:
    99
    Likes Received:
    29
    Yesterday took my Inspire to film a local monument flight distance to the monument was just over a KM, windy conditions took off flew her 750m before losing the video link. Flew her back immediately,video link restored after about 5 seconds, landed switched everything off changed the battery. On the second attempt made it to the monument then flew her back, no issues, really pleased how she responded to the conditions.

    Today went to my local park at 7 am for a short fly about to discharge my TB47 battery that was fully charged but not used for several days. K Index was in the green zone, preflight checks gave a compass MOD reading of 1435 (normal) and sensor values were spot on. I had full GPS showing 17 satellites, took off manually and hovered for about 30 seconds everything stable. Did several loops around the park, 6 minutes into the flight took her up to 28 meters, she started to turn with out control inputs. I thought it was the higher winds at that altitude and dropped her down to 10 meters at this point I did not think there was anything wrong. I noted she was moving from side to side in the hover and brought her in close to me then all of a sudden 7min 30 sec into the flight she started toilet bowling. I have heard of this but did not know exactly what it meant, my inspire without control input started to go round in circles with the nose pointing in the same direction in a clockwise direction. I still had throttle control and tried to get her close to the ground to initiate gear down, this was impossible so quickly manually dropped the gear. At the slowest point in the loop she was doing I dropped her onto the grass and killed the throttle due to the sideways momentum she tilted up as the two left feet caught the ground thankfully she did not tip over and the props did not hit the ground.

    I did not put her into ATTI mode as she was so close to me and no more than three meter high when she finally lost it. I did not calibrate the compass before the flight (about seventh flight without compass calibration) the compass was last calibrated at this flying spot. I used to calibrate the compass each flight however I have taken the advice of several experienced folks that this is not required. Preflight checks confirmed IMU and compass values were in the green values. My I-pad mini 2 was in flight mode, everything has been updated to the latest firmware, everything fully charged. Vast majority of my flights including the first has been at this park, I did nothing different.
    This has shattered my confidence, if this had happened the day before on my Ythsie prop flight I might have lost her. Post flight actions, fully checked out my Inspire, inspected the propellers, QR locks & motors all were in good condition. Did a full IMU calibration on a dead level surface. Will do a compass calibration before my next flight. Link to yesterdays flight, PS I'm still learning so technically not so hot, this is purely a hobby for me.



    Les
     
    Canuck32 likes this.
  2. Frank508

    Joined:
    May 9, 2015
    Messages:
    171
    Likes Received:
    92
    Location:
    SF Bay Area
    I end each flight with at little bit of Atti mode flying practice, both visually, and FPV. I usually run the battery down to return with about 20-25% battery and then run them down to 10% right in front of me. I figure a cycle is a cycle, so why not use it up. Anyway, Atti mode in this quad isn't a big deal and is a part of a back-up matrix to create a safe flight.

    If the Quad has impaired navigation, we have Atti mode, if we loose radio link we have GPS, we would need to loose both to loose the aircraft.

    Anyway, glad it turned out ok for you.
     
    Lesmess likes this.
  3. Lesmess

    Joined:
    Mar 18, 2015
    Messages:
    99
    Likes Received:
    29
    Thanks good advice, problem I live in the North of Scotland very windy here most days just now. I'm ok with Atti close in with low winds, further out I have difficult with the orientation of the bird without looking at the screen. The disturbing thing for me, the inspire has behaved flawlessly for 26 flights. I have read all the problems some folks are having and have put most of it down to poor preflight checks. I am methodical with my checks and 100% sure everything was ok before my flight and even then she almost gave me heart failure. Back to calibrating the compass before flying in a new location.
     
  4. Frank508

    Joined:
    May 9, 2015
    Messages:
    171
    Likes Received:
    92
    Location:
    SF Bay Area
    Scotland is a beautiful country, everything is brown here in California this time of year.

    If Atti mode isn't a option, it might be a good idea to reconsider flying that day. The automated navigation is not that strong a pillar to hang the fate of your quad, yourself and all those around you on.

    Anyway, if you accumulate 20 or 30 minutes of Atti practice you'll feel fine. Remember you can just keep it pointed away from you and fly home backwards so you don have control reversal.
     
    Lesmess likes this.
  5. Lesmess

    Joined:
    Mar 18, 2015
    Messages:
    99
    Likes Received:
    29
    Question - Why did the vision positioning system not help to stop the toilet bowling when she was less than two meters off the ground in gps mode with no stick inputs? If it happens again I switch to Atti mode and land, does anyone know if pushing the return to home button has any effect?
     
  6. Eric

    Joined:
    Mar 8, 2015
    Messages:
    243
    Likes Received:
    91
    Location:
    San Jose
    I've had two instances, when on the current FW .3 my I1 decided to fly her own missions (both ended without damage) the first well above VPS, but the second was at 3.3ft when she started to pitch hard without commands and start to "toilet". Not as mild as .17 toilet bowling was at all. Set her down rechecked Mods and the only difference was a drop in compass to low 1400's. My point is for this thread, the second unstable flight should have had good VPS as I was on a not so great poured asphalt with many defects to lock onto. She didn't seem to care about her VPS logic and or most stick commands at this point. The commands she was responding to was enough for me to land at a otherwise uneventful skidding stop :-( . I recalibrated the compass (back to low 1500's) and was back in the air in less than 3 minutes and have had stable flights since. Not confident at all that this issue won't return under current FW :-(. My first flight was in a sustained 10+Mph winds with hills to hit in every direction and the second instance only lasted 22 seconds, both I remained in GPS with no drop in green "safe to fly".
     
  7. Lesmess

    Joined:
    Mar 18, 2015
    Messages:
    99
    Likes Received:
    29
    Just back from the park where I had my issues yesterday, she flew rock solid today. Did a IMU calibration at home and then the compass dance before flying. The sensor values were almost identical to yesterday. The bird just had a mind if it's own yesterday. Most folks will say must have been a compass error.
    I had no issues doing the compass dance every flight. Then after the latest firmware update I calibrated the compass and left it alone for the next 7 flights or so as recommended by the more experienced folks on this forum, most flights were in different locations within 20 miles of one another. With the compass readings, is it only the MOD value I should be concerned with or should I take note of the other values as well? My Mod value is always around the mid 1400's.
     
  8. Eric

    Joined:
    Mar 8, 2015
    Messages:
    243
    Likes Received:
    91
    Location:
    San Jose
    If my MOD values look normal I fly. The glich might be from over Dancing them.??? I believe the compass is sending false data to the flight control computer and then it tries to "fix" the problem off that incorrect data. Unfortunately, I don't believe anybody quite knows for sure :-(.